From f0fa992e18f6e06364d1a71d5784f2e800e10dcd Mon Sep 17 00:00:00 2001 From: shayaharon Date: Sun, 12 Nov 2023 16:50:21 +0200 Subject: [PATCH 1/5] updated all notebooks --- Makefile | 3 + README.md | 7 +- notebooks/quickstart_segmentation.ipynb | 1590 ++++++++++++++++ .../segmentation_connect_custom_dataset.ipynb | 980 ++++++++++ ...nsfer_learning_semantic_segmentation.ipynb | 1680 +++++++++++++++++ 5 files changed, 4257 insertions(+), 3 deletions(-) create mode 100644 notebooks/quickstart_segmentation.ipynb create mode 100644 notebooks/segmentation_connect_custom_dataset.ipynb create mode 100644 notebooks/transfer_learning_semantic_segmentation.ipynb diff --git a/Makefile b/Makefile index 53a669d05e..0ac1f3384c 100644 --- a/Makefile +++ b/Makefile @@ -37,6 +37,9 @@ NOTEBOOKS_TO_RUN += notebooks/what_are_recipes_and_how_to_use.ipynb NOTEBOOKS_TO_RUN += notebooks/transfer_learning_classification.ipynb NOTEBOOKS_TO_RUN += notebooks/how_to_use_knowledge_distillation_for_classification.ipynb NOTEBOOKS_TO_RUN += notebooks/PTQ_and_QAT_for_classification.ipynb +NOTEBOOKS_TO_RUN += notebooks/quickstart_segmentation.ipynb +NOTEBOOKS_TO_RUN += notebooks/segmentation_connect_custom_dataset.ipynb +NOTEBOOKS_TO_RUN += notebooks/transfer_learning_semantic_segmentation.ipynb # If there are additional notebooks that must not be executed, but still should be checked for version match, add them here NOTEBOOKS_TO_CHECK := $(NOTEBOOKS_TO_RUN) diff --git a/README.md b/README.md index 252ee4dc98..14a77653b8 100644 --- a/README.md +++ b/README.md @@ -211,9 +211,10 @@ model = models.get("model-name", pretrained_weights="pretrained-model-name") ### Semantic Segmentation -* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://bit.ly/3qKx9m8) [Segmentation Quick Start](https://bit.ly/3qKx9m8) -* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://bit.ly/3qKwMbe) [Segmentation Transfer Learning](https://bit.ly/3qKwMbe) -* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://bit.ly/3QQBVJp) [How to Connect Custom Dataset](https://bit.ly/3QQBVJp) +* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/quickstart_segmentation.ipynb) [Segmentation Quick Start](https://bit.ly/3qKx9m8) +* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/transfer_learning_semantic_segmentation.ipynb) [Segmentation Transfer Learning](https://bit.ly/3qKx9m8) +* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/segmentation_connect_custom_dataset.ipynb) [How to Connect Custom Dataset](https://bit.ly/3qKx9m8) + ### Pose Estimation diff --git a/notebooks/quickstart_segmentation.ipynb b/notebooks/quickstart_segmentation.ipynb new file mode 100644 index 0000000000..63150b51c5 --- /dev/null +++ b/notebooks/quickstart_segmentation.ipynb @@ -0,0 +1,1590 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "HY_HuQbxn7X0" + }, + "source": [ + "![SG - Horizontal.png](data:image/png;base64,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)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "oA_p5zIsoAJQ" + }, + "source": [ + "# SuperGradients quick start Semantic Segmentation\n", + "\n", + "In this tutorial we will train PPLiteSeg model on Supervisely semantic segmentation dataset\n", + "\n", + "The notebook is divided into 7 sections:\n", + "1. Experiment setup\n", + "2. Dataset definition\n", + "3. Architecture definition\n", + "4. Training setup\n", + "5. Training and Evaluation\n", + "6. Predict\n", + "7. Convert to ONNX\\TensorRT" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "GqH4VGMroWec" + }, + "source": [ + "#Install SG" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Q8uA6AWEhHN6" + }, + "source": [ + "The cell below will install **super_gradients** which will automatically get all its dependencies." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "-mm-E4xRoNEm", + "outputId": "ce0b8873-49f3-44a4-f8f1-e53087c4f96b" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + " Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", + " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m135.8/135.8 kB\u001B[0m \u001B[31m4.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m684.5/684.5 kB\u001B[0m \u001B[31m14.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.9/2.9 MB\u001B[0m \u001B[31m26.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m42.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m408.6/408.6 kB\u001B[0m \u001B[31m36.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m154.5/154.5 kB\u001B[0m \u001B[31m24.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.5/79.5 kB\u001B[0m \u001B[31m13.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m4.5/4.5 MB\u001B[0m \u001B[31m62.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m13.5/13.5 MB\u001B[0m \u001B[31m62.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m68.0/68.0 kB\u001B[0m \u001B[31m9.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", + " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m17.0/17.0 MB\u001B[0m \u001B[31m16.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m3.3/3.3 MB\u001B[0m \u001B[31m89.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.2/2.2 MB\u001B[0m \u001B[31m84.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m455.0/455.0 kB\u001B[0m \u001B[31m44.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.0/46.0 kB\u001B[0m \u001B[31m7.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m11.3/11.3 MB\u001B[0m \u001B[31m115.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.8/79.8 kB\u001B[0m \u001B[31m12.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m108.3/108.3 kB\u001B[0m \u001B[31m17.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m176.0/176.0 kB\u001B[0m \u001B[31m26.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m117.0/117.0 kB\u001B[0m \u001B[31m19.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m575.5/575.5 kB\u001B[0m \u001B[31m57.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m121.1/121.1 kB\u001B[0m \u001B[31m19.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m86.8/86.8 kB\u001B[0m \u001B[31m14.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m18.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m17.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.6/120.6 kB\u001B[0m \u001B[31m20.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m14.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m14.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m84.7/84.7 kB\u001B[0m \u001B[31m15.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m16.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m12.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.8/99.8 kB\u001B[0m \u001B[31m15.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m14.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m14.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m90.6/90.6 kB\u001B[0m \u001B[31m14.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m15.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m14.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m16.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m94.0/94.0 kB\u001B[0m \u001B[31m16.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m407.7/407.7 kB\u001B[0m \u001B[31m54.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m107.7/107.7 kB\u001B[0m \u001B[31m16.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m277.4/277.4 kB\u001B[0m \u001B[31m40.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m113.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m913.9/913.9 kB\u001B[0m \u001B[31m80.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m105.0/105.0 kB\u001B[0m \u001B[31m16.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.2/46.2 kB\u001B[0m \u001B[31m7.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m106.8/106.8 kB\u001B[0m \u001B[31m17.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m194.6/194.6 kB\u001B[0m \u001B[31m28.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m58.1/58.1 kB\u001B[0m \u001B[31m8.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Building wheel for super-gradients (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for pycocotools (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for termcolor (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for treelib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for coverage (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for antlr4-python3-runtime (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for stringcase (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for xhtml2pdf (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for svglib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", + "lida 0.0.10 requires fastapi, which is not installed.\n", + "lida 0.0.10 requires kaleido, which is not installed.\n", + "lida 0.0.10 requires python-multipart, which is not installed.\n", + "lida 0.0.10 requires uvicorn, which is not installed.\n", + "tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.23.0 which is incompatible.\u001B[0m\u001B[31m\n", + "\u001B[0m" + ] + } + ], + "source": [ + "! pip install -qq super-gradients==3.4.1\n", + "\n", + "! pip install -qq prettyformatter\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "892xArqDsGsQ" + }, + "source": [ + "# 1. Experiment setup\n", + "We will initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", + "\n", + "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", + "\n", + "```\n", + "ckpt_root_dir\n", + "|─── experiment_name_1\n", + "│ ckpt_best.pth # Model checkpoint on best epoch\n", + "│ ckpt_latest.pth # Model checkpoint on last epoch\n", + "│ average_model.pth # Model checkpoint averaged over epochs\n", + "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", + "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", + "└─── experiment_name_2\n", + " ...\n", + "```\n", + "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", + " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "pl0WPz1HisFz" + }, + "source": [ + "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "HAff--HysJmP", + "outputId": "aa92f470-b3ce-448d-c1d9-39aa8b9cca72" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-08 10:54:04] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is logged into /root/sg_logs/console.log\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-08 10:54:04] WARNING - __init__.py - Failed to import pytorch_quantization\n", + "[2023-11-08 10:54:04] INFO - utils.py - NumExpr defaulting to 2 threads.\n", + "[2023-11-08 10:54:17] WARNING - calibrator.py - Failed to import pytorch_quantization\n", + "[2023-11-08 10:54:17] WARNING - export.py - Failed to import pytorch_quantization\n", + "[2023-11-08 10:54:17] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n", + "[2023-11-08 10:54:17] INFO - env_sanity_check.py - Library check is not supported when super_gradients installed through \"git+https://github.com/...\" command\n" + ] + } + ], + "source": [ + "from super_gradients import Trainer\n", + "\n", + "CHECKPOINT_DIR = '/home/notebook_ckpts/'\n", + "trainer = Trainer(experiment_name=\"segmentation_quick_start\", ckpt_root_dir=CHECKPOINT_DIR)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "dwVMY4gMjQSL" + }, + "source": [ + "# 2. Dataset definition\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fpIWhnR9j2rm" + }, + "source": [ + "\n", + "For the sake of this presentation, we'll use **Supervisely** semantic segmentation dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ZACgRb-qjzDJ" + }, + "source": [ + "SG trainer is fully compatible with PyTorch data loaders, so you can definitely use your own data for the experiment below if you prefer." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6ulV6Hpao3IN" + }, + "source": [ + "## 2.A. Download data\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mVwslNv-j-2C" + }, + "source": [ + "Feel free to change the download path by editing SUPERVISELY_DATASET_DOWNLOAD_PATH" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "id": "dfR18Rmbo00y", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "8ed988c8-190a-4637-c0a5-ebcb173e7329" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Downloading and extracting supervisely dataset to: /home/data\n", + "/home/data\n", + "--2023-11-08 10:54:17-- https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", + "Resolving deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)... 52.216.49.177, 52.217.138.241, 52.217.171.137, ...\n", + "Connecting to deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)|52.216.49.177|:443... connected.\n", + "HTTP request sent, awaiting response... 200 OK\n", + "Length: 3564001012 (3.3G) [application/zip]\n", + "Saving to: ‘supervisely-persons.zip’\n", + "\n", + "supervisely-persons 100%[===================>] 3.32G 41.2MB/s in 72s \n", + "\n", + "2023-11-08 10:55:30 (47.0 MB/s) - ‘supervisely-persons.zip’ saved [3564001012/3564001012]\n", + "\n" + ] + } + ], + "source": [ + "import os\n", + "\n", + "SUPERVISELY_DATASET_DOWNLOAD_PATH=\"/home/data\"\n", + "\n", + "supervisely_dataset_dir_path = SUPERVISELY_DATASET_DOWNLOAD_PATH + os.path.sep + 'supervisely-persons'\n", + "\n", + "if os.path.isdir(supervisely_dataset_dir_path):\n", + " print('supervisely dataset already downloaded...')\n", + "else:\n", + " print('Downloading and extracting supervisely dataset to: ' + SUPERVISELY_DATASET_DOWNLOAD_PATH)\n", + " ! mkdir $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " %cd $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " ! wget https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", + " ! unzip --qq supervisely-persons.zip" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": false, + "id": "V9ZcklupX8Qx" + }, + "source": [ + "## 2.B. Create data loaders\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3Mk_YixjlEhj" + }, + "source": [ + "The dataloaders are initiated with the default parameters defined in the [yaml](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/recipes/dataset_params/supervisely_persons_dataset_params.yaml)\n", + "file. Parameters as batch_size, transforms, root_dir and others can be overridden by passing as `dataset_params` and\n", + "`dataloader_params`, as implemented bellow." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "S3BzMRhSX8Qx" + }, + "outputs": [], + "source": [ + "from super_gradients.training import dataloaders\n", + "root_dir = supervisely_dataset_dir_path\n", + "batch_size = 8\n", + "\n", + "train_loader = dataloaders.supervisely_persons_train(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})\n", + "valid_loader = dataloaders.supervisely_persons_val(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6dHIwvs46-dk" + }, + "source": [ + "As you can see, we didn't have to pass many parameters into the dataloaders construction. That's because defaults are pre-defined for your convenience, and you might be curious to know what they are. Let's print them and see which resolution and transformations are defined." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "id": "76tzhKxi6aS-" + }, + "outputs": [], + "source": [ + "from prettyformatter import pprint\n", + "\n", + "print('Dataloader parameters:')\n", + "pprint(train_loader.dataloader_params)\n", + "print('Dataset parameters')\n", + "pprint(train_loader.dataset.dataset_params)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "I4QEOkKyy93R" + }, + "source": [ + "We can take a look at some images from the dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "l5GcDAg_pUGJ" + }, + "source": [ + "# 3. Architecture definition\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "id": "xXPMJQCJzmb4" + }, + "outputs": [], + "source": [ + "from super_gradients.training import models\n", + "from super_gradients.common.object_names import Models\n", + "\n", + "model = models.get(model_name=Models.PP_LITE_T_SEG,\n", + " arch_params={\"use_aux_heads\": False},\n", + " num_classes=1)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fU8orO7wlwIK" + }, + "source": [ + "SG includes implementations of many different architectures for semantic segmentation tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-oGSU3V8lqcm" + }, + "source": [ + "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", + "and extra Auxiliary heads aren't used for training.\n", + "\n", + "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "X-_dBewgr1dG" + }, + "source": [ + "# 4. Training setup\n", + "\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "H1Rll8Orl-Dy" + }, + "source": [ + "\n", + "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n", + "\n", + "We will be using an average of BCE and Dice loss for segmentation, with different learning rates for the replaced segmentation head layer, and the rest of the network- this is controlled by the `multiply_head_lr` parameter which is the multiplication factor of the learning rate for the newly replaced layer.\n", + "\n", + "As our `metric_to_watch`, we will be monitoring the `target_IOU` which is one of the components of `BinaryIOU` torchmetrics object (the other components are `mean_IOU` which is the mean of the background and target IOUs, and `background_IOU`)." + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "id": "NShu3zLgr5qD" + }, + "outputs": [], + "source": [ + "from super_gradients.training.metrics.segmentation_metrics import BinaryIOU\n", + "from super_gradients.training.utils.callbacks import BinarySegmentationVisualizationCallback, Phase\n", + "\n", + "train_params = {\"max_epochs\": 15,\n", + " \"lr_mode\": \"cosine\",\n", + " \"initial_lr\": 0.01,\n", + " \"lr_warmup_epochs\": 5,\n", + " \"multiply_head_lr\": 10,\n", + " \"optimizer\": \"SGD\",\n", + " \"loss\": \"bce_dice_loss\",\n", + " \"ema\": True,\n", + " \"zero_weight_decay_on_bias_and_bn\": True,\n", + " \"average_best_models\": True,\n", + " \"metric_to_watch\": \"target_IOU\",\n", + " \"greater_metric_to_watch_is_better\": True,\n", + " \"train_metrics_list\": [BinaryIOU()],\n", + " \"valid_metrics_list\": [BinaryIOU()],\n", + " \"loss_logging_items_names\": [\"loss\"]\n", + " }" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qTECVyhcs506" + }, + "source": [ + "# 5. Training and evaluation\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "S1K5MU2kmmDb" + }, + "source": [ + "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", + "\n", + "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "id": "u6roEj9ktFTi", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "9b530548-7596-4db2-f1cf-2c461e9cb5bd" + }, + "outputs": [ + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "Dataloader parameters:\n", + "{\"batch_size\": 8, \"shuffle\": True, \"drop_last\": True}\n", + "Dataset parameters\n", + "{'root_dir': '/home/data/supervisely-persons', 'list_file': 'train.csv', 'cache_labels': False, 'cache_images': False, 'transforms': [{'SegRandomRescale': {'scales': [0.25, 1.0]}}, {'SegColorJitter': {'brightness': 0.5, 'contrast': 0.5, 'saturation': 0.5}}, {'SegRandomFlip': {'prob': 0.5}}, {'SegPadShortToCropSize': {'crop_size': [320, 480], 'fill_mask': 0}}, {'SegCropImageAndMask': {'crop_size': [320, 480], 'mode': 'random'}}]}\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "[2023-11-08 10:56:13] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231108_105613_531244`\n", + "[2023-11-08 10:56:13] INFO - sg_trainer.py - Checkpoints directory: /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244\n", + "/usr/local/lib/python3.10/dist-packages/super_gradients/common/registry/registry.py:72: DeprecationWarning: Object name `bce_dice_loss` is now deprecated. Please replace it with `BCEDiceLoss`.\n", + " warnings.warn(f\"Object name `{name}` is now deprecated. Please replace it with `{deprecated_names[name]}`.\", DeprecationWarning)\n", + "[2023-11-08 10:56:13] INFO - sg_trainer.py - Using EMA with params {}\n", + "[2023-11-08 10:56:13] WARNING - ema.py - Parameter `decay` is not specified for EMA params. Please specify `decay` parameter explicitly in your config:\n", + "ema: True\n", + "ema_params: \n", + " decay: 0.9999\n", + " decay_type: exp\n", + " beta: 15\n", + "Will default to decay: 0.9999\n", + "In the next major release of SG this warning will become an error.\n", + "[2023-11-08 10:56:13] WARNING - ema.py - Parameter decay_type is not specified for EMA model. Please specify decay_type parameter explicitly in your config:\n", + "ema: True\n", + "ema_params: \n", + " decay: 0.9999\n", + " decay_type: constant|exp|threshold\n", + "Will default to `exp` decay with beta = 15\n", + "In the next major release of SG this warning will become an error.\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "The console stream is now moved to /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/console_Nov08_10_56_13.txt\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "[2023-11-08 10:56:14] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", + " - Mode: Single GPU\n", + " - Number of GPUs: 1 (1 available on the machine)\n", + " - Full dataset size: 2477 (len(train_set))\n", + " - Batch size per GPU: 8 (batch_size)\n", + " - Batch Accumulate: 1 (batch_accumulate)\n", + " - Total batch size: 8 (num_gpus * batch_size)\n", + " - Effective Batch size: 8 (num_gpus * batch_size * batch_accumulate)\n", + " - Iterations per epoch: 309 (len(train_loader))\n", + " - Gradient updates per epoch: 309 (len(train_loader) / batch_accumulate)\n", + "\n", + "[2023-11-08 10:56:14] INFO - sg_trainer.py - Started training for 15 epochs (0/14)\n", + "\n", + "Train epoch 0: 100%|██████████| 309/309 [02:08<00:00, 2.41it/s, BCEDiceLoss=0.405, background_IOU=0.539, gpu_mem=1.14, mean_IOU=0.604, target_IOU=0.669]\n", + "Validating: 100%|██████████| 65/65 [00:16<00:00, 3.86it/s]\n", + "[2023-11-08 10:58:39] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 10:58:39] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.6919947266578674\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 0\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.4047\n", + "│ ├── Target_iou = 0.6685\n", + "│ ├── Background_iou = 0.5393\n", + "│ └── Mean_iou = 0.6039\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3857\n", + " ├── Target_iou = 0.692\n", + " ├── Background_iou = 0.4666\n", + " └── Mean_iou = 0.5793\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 1: 100%|██████████| 309/309 [01:57<00:00, 2.63it/s, BCEDiceLoss=0.338, background_IOU=0.606, gpu_mem=1.14, mean_IOU=0.663, target_IOU=0.719]\n", + "Validating epoch 1: 100%|██████████| 65/65 [00:16<00:00, 4.02it/s]\n", + "[2023-11-08 11:00:56] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:00:56] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7162820100784302\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 1\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.3375\n", + "│ │ ├── Epoch N-1 = 0.4047 (\u001B[32m↘ -0.0672\u001B[0m)\n", + "│ │ └── Best until now = 0.4047 (\u001B[32m↘ -0.0672\u001B[0m)\n", + "│ ├── Target_iou = 0.7191\n", + "│ │ ├── Epoch N-1 = 0.6685 (\u001B[32m↗ 0.0506\u001B[0m)\n", + "│ │ └── Best until now = 0.6685 (\u001B[32m↗ 0.0506\u001B[0m)\n", + "│ ├── Background_iou = 0.6064\n", + "│ │ ├── Epoch N-1 = 0.5393 (\u001B[32m↗ 0.0671\u001B[0m)\n", + "│ │ └── Best until now = 0.5393 (\u001B[32m↗ 0.0671\u001B[0m)\n", + "│ └── Mean_iou = 0.6628\n", + "│ ├── Epoch N-1 = 0.6039 (\u001B[32m↗ 0.0588\u001B[0m)\n", + "│ └── Best until now = 0.6039 (\u001B[32m↗ 0.0588\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.388\n", + " │ ├── Epoch N-1 = 0.3857 (\u001B[31m↗ 0.0023\u001B[0m)\n", + " │ └── Best until now = 0.3857 (\u001B[31m↗ 0.0023\u001B[0m)\n", + " ├── Target_iou = 0.7163\n", + " │ ├── Epoch N-1 = 0.692 (\u001B[32m↗ 0.0243\u001B[0m)\n", + " │ └── Best until now = 0.692 (\u001B[32m↗ 0.0243\u001B[0m)\n", + " ├── Background_iou = 0.3871\n", + " │ ├── Epoch N-1 = 0.4666 (\u001B[31m↘ -0.0795\u001B[0m)\n", + " │ └── Best until now = 0.4666 (\u001B[31m↘ -0.0795\u001B[0m)\n", + " └── Mean_iou = 0.5517\n", + " ├── Epoch N-1 = 0.5793 (\u001B[31m↘ -0.0276\u001B[0m)\n", + " └── Best until now = 0.5793 (\u001B[31m↘ -0.0276\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 2: 100%|██████████| 309/309 [01:57<00:00, 2.64it/s, BCEDiceLoss=0.326, background_IOU=0.629, gpu_mem=1.14, mean_IOU=0.679, target_IOU=0.729]\n", + "Validating epoch 2: 100%|██████████| 65/65 [00:16<00:00, 3.94it/s]\n", + "[2023-11-08 11:03:11] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:03:11] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7318407297134399\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 2\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.3256\n", + "│ │ ├── Epoch N-1 = 0.3375 (\u001B[32m↘ -0.0119\u001B[0m)\n", + "│ │ └── Best until now = 0.3375 (\u001B[32m↘ -0.0119\u001B[0m)\n", + "│ ├── Target_iou = 0.7294\n", + "│ │ ├── Epoch N-1 = 0.7191 (\u001B[32m↗ 0.0102\u001B[0m)\n", + "│ │ └── Best until now = 0.7191 (\u001B[32m↗ 0.0102\u001B[0m)\n", + "│ ├── Background_iou = 0.6291\n", + "│ │ ├── Epoch N-1 = 0.6064 (\u001B[32m↗ 0.0227\u001B[0m)\n", + "│ │ └── Best until now = 0.6064 (\u001B[32m↗ 0.0227\u001B[0m)\n", + "│ └── Mean_iou = 0.6792\n", + "│ ├── Epoch N-1 = 0.6628 (\u001B[32m↗ 0.0164\u001B[0m)\n", + "│ └── Best until now = 0.6628 (\u001B[32m↗ 0.0164\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3604\n", + " │ ├── Epoch N-1 = 0.388 (\u001B[32m↘ -0.0275\u001B[0m)\n", + " │ └── Best until now = 0.3857 (\u001B[32m↘ -0.0252\u001B[0m)\n", + " ├── Target_iou = 0.7318\n", + " │ ├── Epoch N-1 = 0.7163 (\u001B[32m↗ 0.0156\u001B[0m)\n", + " │ └── Best until now = 0.7163 (\u001B[32m↗ 0.0156\u001B[0m)\n", + " ├── Background_iou = 0.4626\n", + " │ ├── Epoch N-1 = 0.3871 (\u001B[32m↗ 0.0755\u001B[0m)\n", + " │ └── Best until now = 0.4666 (\u001B[31m↘ -0.004\u001B[0m)\n", + " └── Mean_iou = 0.5972\n", + " ├── Epoch N-1 = 0.5517 (\u001B[32m↗ 0.0455\u001B[0m)\n", + " └── Best until now = 0.5793 (\u001B[32m↗ 0.0179\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 3: 100%|██████████| 309/309 [01:57<00:00, 2.63it/s, BCEDiceLoss=0.306, background_IOU=0.642, gpu_mem=1.14, mean_IOU=0.695, target_IOU=0.748]\n", + "Validating epoch 3: 100%|██████████| 65/65 [00:17<00:00, 3.80it/s]\n", + "[2023-11-08 11:05:27] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:05:27] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7394765019416809\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 3\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.306\n", + "│ │ ├── Epoch N-1 = 0.3256 (\u001B[32m↘ -0.0196\u001B[0m)\n", + "│ │ └── Best until now = 0.3256 (\u001B[32m↘ -0.0196\u001B[0m)\n", + "│ ├── Target_iou = 0.7483\n", + "│ │ ├── Epoch N-1 = 0.7294 (\u001B[32m↗ 0.0189\u001B[0m)\n", + "│ │ └── Best until now = 0.7294 (\u001B[32m↗ 0.0189\u001B[0m)\n", + "│ ├── Background_iou = 0.6421\n", + "│ │ ├── Epoch N-1 = 0.6291 (\u001B[32m↗ 0.0131\u001B[0m)\n", + "│ │ └── Best until now = 0.6291 (\u001B[32m↗ 0.0131\u001B[0m)\n", + "│ └── Mean_iou = 0.6952\n", + "│ ├── Epoch N-1 = 0.6792 (\u001B[32m↗ 0.016\u001B[0m)\n", + "│ └── Best until now = 0.6792 (\u001B[32m↗ 0.016\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3468\n", + " │ ├── Epoch N-1 = 0.3604 (\u001B[32m↘ -0.0136\u001B[0m)\n", + " │ └── Best until now = 0.3604 (\u001B[32m↘ -0.0136\u001B[0m)\n", + " ├── Target_iou = 0.7395\n", + " │ ├── Epoch N-1 = 0.7318 (\u001B[32m↗ 0.0076\u001B[0m)\n", + " │ └── Best until now = 0.7318 (\u001B[32m↗ 0.0076\u001B[0m)\n", + " ├── Background_iou = 0.4784\n", + " │ ├── Epoch N-1 = 0.4626 (\u001B[32m↗ 0.0158\u001B[0m)\n", + " │ └── Best until now = 0.4666 (\u001B[32m↗ 0.0118\u001B[0m)\n", + " └── Mean_iou = 0.6089\n", + " ├── Epoch N-1 = 0.5972 (\u001B[32m↗ 0.0117\u001B[0m)\n", + " └── Best until now = 0.5972 (\u001B[32m↗ 0.0117\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 4: 100%|██████████| 309/309 [01:58<00:00, 2.61it/s, BCEDiceLoss=0.29, background_IOU=0.668, gpu_mem=1.14, mean_IOU=0.714, target_IOU=0.76]\n", + "Validating epoch 4: 100%|██████████| 65/65 [00:16<00:00, 3.86it/s]\n", + "[2023-11-08 11:07:45] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:07:45] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7402159571647644\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 4\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2896\n", + "│ │ ├── Epoch N-1 = 0.306 (\u001B[32m↘ -0.0164\u001B[0m)\n", + "│ │ └── Best until now = 0.306 (\u001B[32m↘ -0.0164\u001B[0m)\n", + "│ ├── Target_iou = 0.76\n", + "│ │ ├── Epoch N-1 = 0.7483 (\u001B[32m↗ 0.0117\u001B[0m)\n", + "│ │ └── Best until now = 0.7483 (\u001B[32m↗ 0.0117\u001B[0m)\n", + "│ ├── Background_iou = 0.668\n", + "│ │ ├── Epoch N-1 = 0.6421 (\u001B[32m↗ 0.0259\u001B[0m)\n", + "│ │ └── Best until now = 0.6421 (\u001B[32m↗ 0.0259\u001B[0m)\n", + "│ └── Mean_iou = 0.714\n", + "│ ├── Epoch N-1 = 0.6952 (\u001B[32m↗ 0.0188\u001B[0m)\n", + "│ └── Best until now = 0.6952 (\u001B[32m↗ 0.0188\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3405\n", + " │ ├── Epoch N-1 = 0.3468 (\u001B[32m↘ -0.0063\u001B[0m)\n", + " │ └── Best until now = 0.3468 (\u001B[32m↘ -0.0063\u001B[0m)\n", + " ├── Target_iou = 0.7402\n", + " │ ├── Epoch N-1 = 0.7395 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " │ └── Best until now = 0.7395 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " ├── Background_iou = 0.4861\n", + " │ ├── Epoch N-1 = 0.4784 (\u001B[32m↗ 0.0077\u001B[0m)\n", + " │ └── Best until now = 0.4784 (\u001B[32m↗ 0.0077\u001B[0m)\n", + " └── Mean_iou = 0.6131\n", + " ├── Epoch N-1 = 0.6089 (\u001B[32m↗ 0.0042\u001B[0m)\n", + " └── Best until now = 0.6089 (\u001B[32m↗ 0.0042\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 5: 100%|██████████| 309/309 [01:59<00:00, 2.58it/s, BCEDiceLoss=0.289, background_IOU=0.663, gpu_mem=1.14, mean_IOU=0.712, target_IOU=0.761]\n", + "Validating epoch 5: 100%|██████████| 65/65 [00:17<00:00, 3.71it/s]\n", + "[2023-11-08 11:10:03] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:10:03] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7486929297447205\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 5\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2888\n", + "│ │ ├── Epoch N-1 = 0.2896 (\u001B[32m↘ -0.0008\u001B[0m)\n", + "│ │ └── Best until now = 0.2896 (\u001B[32m↘ -0.0008\u001B[0m)\n", + "│ ├── Target_iou = 0.7608\n", + "│ │ ├── Epoch N-1 = 0.76 (\u001B[32m↗ 0.0008\u001B[0m)\n", + "│ │ └── Best until now = 0.76 (\u001B[32m↗ 0.0008\u001B[0m)\n", + "│ ├── Background_iou = 0.6632\n", + "│ │ ├── Epoch N-1 = 0.668 (\u001B[31m↘ -0.0048\u001B[0m)\n", + "│ │ └── Best until now = 0.668 (\u001B[31m↘ -0.0048\u001B[0m)\n", + "│ └── Mean_iou = 0.712\n", + "│ ├── Epoch N-1 = 0.714 (\u001B[31m↘ -0.002\u001B[0m)\n", + "│ └── Best until now = 0.714 (\u001B[31m↘ -0.002\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3203\n", + " │ ├── Epoch N-1 = 0.3405 (\u001B[32m↘ -0.0202\u001B[0m)\n", + " │ └── Best until now = 0.3405 (\u001B[32m↘ -0.0202\u001B[0m)\n", + " ├── Target_iou = 0.7487\n", + " │ ├── Epoch N-1 = 0.7402 (\u001B[32m↗ 0.0085\u001B[0m)\n", + " │ └── Best until now = 0.7402 (\u001B[32m↗ 0.0085\u001B[0m)\n", + " ├── Background_iou = 0.5274\n", + " │ ├── Epoch N-1 = 0.4861 (\u001B[32m↗ 0.0413\u001B[0m)\n", + " │ └── Best until now = 0.4861 (\u001B[32m↗ 0.0413\u001B[0m)\n", + " └── Mean_iou = 0.638\n", + " ├── Epoch N-1 = 0.6131 (\u001B[32m↗ 0.0249\u001B[0m)\n", + " └── Best until now = 0.6131 (\u001B[32m↗ 0.0249\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 6: 100%|██████████| 309/309 [01:58<00:00, 2.61it/s, BCEDiceLoss=0.273, background_IOU=0.686, gpu_mem=1.14, mean_IOU=0.728, target_IOU=0.77]\n", + "Validating epoch 6: 100%|██████████| 65/65 [00:16<00:00, 3.93it/s]\n", + "[2023-11-08 11:12:20] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:12:20] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.751246988773346\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 6\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2731\n", + "│ │ ├── Epoch N-1 = 0.2888 (\u001B[32m↘ -0.0158\u001B[0m)\n", + "│ │ └── Best until now = 0.2888 (\u001B[32m↘ -0.0158\u001B[0m)\n", + "│ ├── Target_iou = 0.7704\n", + "│ │ ├── Epoch N-1 = 0.7608 (\u001B[32m↗ 0.0096\u001B[0m)\n", + "│ │ └── Best until now = 0.7608 (\u001B[32m↗ 0.0096\u001B[0m)\n", + "│ ├── Background_iou = 0.686\n", + "│ │ ├── Epoch N-1 = 0.6632 (\u001B[32m↗ 0.0228\u001B[0m)\n", + "│ │ └── Best until now = 0.668 (\u001B[32m↗ 0.018\u001B[0m)\n", + "│ └── Mean_iou = 0.7282\n", + "│ ├── Epoch N-1 = 0.712 (\u001B[32m↗ 0.0162\u001B[0m)\n", + "│ └── Best until now = 0.714 (\u001B[32m↗ 0.0142\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3158\n", + " │ ├── Epoch N-1 = 0.3203 (\u001B[32m↘ -0.0045\u001B[0m)\n", + " │ └── Best until now = 0.3203 (\u001B[32m↘ -0.0045\u001B[0m)\n", + " ├── Target_iou = 0.7512\n", + " │ ├── Epoch N-1 = 0.7487 (\u001B[32m↗ 0.0026\u001B[0m)\n", + " │ └── Best until now = 0.7487 (\u001B[32m↗ 0.0026\u001B[0m)\n", + " ├── Background_iou = 0.5394\n", + " │ ├── Epoch N-1 = 0.5274 (\u001B[32m↗ 0.012\u001B[0m)\n", + " │ └── Best until now = 0.5274 (\u001B[32m↗ 0.012\u001B[0m)\n", + " └── Mean_iou = 0.6453\n", + " ├── Epoch N-1 = 0.638 (\u001B[32m↗ 0.0073\u001B[0m)\n", + " └── Best until now = 0.638 (\u001B[32m↗ 0.0073\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 7: 100%|██████████| 309/309 [01:58<00:00, 2.62it/s, BCEDiceLoss=0.26, background_IOU=0.699, gpu_mem=1.14, mean_IOU=0.739, target_IOU=0.78]\n", + "Validating epoch 7: 100%|██████████| 65/65 [00:17<00:00, 3.79it/s]\n", + "[2023-11-08 11:14:37] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:14:37] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7533503174781799\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 7\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.26\n", + "│ │ ├── Epoch N-1 = 0.2731 (\u001B[32m↘ -0.013\u001B[0m)\n", + "│ │ └── Best until now = 0.2731 (\u001B[32m↘ -0.013\u001B[0m)\n", + "│ ├── Target_iou = 0.7799\n", + "│ │ ├── Epoch N-1 = 0.7704 (\u001B[32m↗ 0.0095\u001B[0m)\n", + "│ │ └── Best until now = 0.7704 (\u001B[32m↗ 0.0095\u001B[0m)\n", + "│ ├── Background_iou = 0.6987\n", + "│ │ ├── Epoch N-1 = 0.686 (\u001B[32m↗ 0.0127\u001B[0m)\n", + "│ │ └── Best until now = 0.686 (\u001B[32m↗ 0.0127\u001B[0m)\n", + "│ └── Mean_iou = 0.7393\n", + "│ ├── Epoch N-1 = 0.7282 (\u001B[32m↗ 0.0111\u001B[0m)\n", + "│ └── Best until now = 0.7282 (\u001B[32m↗ 0.0111\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3143\n", + " │ ├── Epoch N-1 = 0.3158 (\u001B[32m↘ -0.0015\u001B[0m)\n", + " │ └── Best until now = 0.3158 (\u001B[32m↘ -0.0015\u001B[0m)\n", + " ├── Target_iou = 0.7534\n", + " │ ├── Epoch N-1 = 0.7512 (\u001B[32m↗ 0.0021\u001B[0m)\n", + " │ └── Best until now = 0.7512 (\u001B[32m↗ 0.0021\u001B[0m)\n", + " ├── Background_iou = 0.5443\n", + " │ ├── Epoch N-1 = 0.5394 (\u001B[32m↗ 0.0049\u001B[0m)\n", + " │ └── Best until now = 0.5394 (\u001B[32m↗ 0.0049\u001B[0m)\n", + " └── Mean_iou = 0.6488\n", + " ├── Epoch N-1 = 0.6453 (\u001B[32m↗ 0.0035\u001B[0m)\n", + " └── Best until now = 0.6453 (\u001B[32m↗ 0.0035\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 8: 100%|██████████| 309/309 [01:58<00:00, 2.60it/s, BCEDiceLoss=0.252, background_IOU=0.711, gpu_mem=1.14, mean_IOU=0.749, target_IOU=0.786]\n", + "Validating epoch 8: 100%|██████████| 65/65 [00:16<00:00, 3.92it/s]\n", + "[2023-11-08 11:16:56] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:16:56] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7535856366157532\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 8\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2517\n", + "│ │ ├── Epoch N-1 = 0.26 (\u001B[32m↘ -0.0084\u001B[0m)\n", + "│ │ └── Best until now = 0.26 (\u001B[32m↘ -0.0084\u001B[0m)\n", + "│ ├── Target_iou = 0.7861\n", + "│ │ ├── Epoch N-1 = 0.7799 (\u001B[32m↗ 0.0062\u001B[0m)\n", + "│ │ └── Best until now = 0.7799 (\u001B[32m↗ 0.0062\u001B[0m)\n", + "│ ├── Background_iou = 0.7112\n", + "│ │ ├── Epoch N-1 = 0.6987 (\u001B[32m↗ 0.0125\u001B[0m)\n", + "│ │ └── Best until now = 0.6987 (\u001B[32m↗ 0.0125\u001B[0m)\n", + "│ └── Mean_iou = 0.7487\n", + "│ ├── Epoch N-1 = 0.7393 (\u001B[32m↗ 0.0093\u001B[0m)\n", + "│ └── Best until now = 0.7393 (\u001B[32m↗ 0.0093\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.314\n", + " │ ├── Epoch N-1 = 0.3143 (\u001B[32m↘ -0.0003\u001B[0m)\n", + " │ └── Best until now = 0.3143 (\u001B[32m↘ -0.0003\u001B[0m)\n", + " ├── Target_iou = 0.7536\n", + " │ ├── Epoch N-1 = 0.7534 (\u001B[32m↗ 0.0002\u001B[0m)\n", + " │ └── Best until now = 0.7534 (\u001B[32m↗ 0.0002\u001B[0m)\n", + " ├── Background_iou = 0.5448\n", + " │ ├── Epoch N-1 = 0.5443 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " │ └── Best until now = 0.5443 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " └── Mean_iou = 0.6492\n", + " ├── Epoch N-1 = 0.6488 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " └── Best until now = 0.6488 (\u001B[32m↗ 0.0004\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 9: 100%|██████████| 309/309 [01:58<00:00, 2.61it/s, BCEDiceLoss=0.248, background_IOU=0.711, gpu_mem=1.14, mean_IOU=0.75, target_IOU=0.789]\n", + "Validating epoch 9: 100%|██████████| 65/65 [00:17<00:00, 3.76it/s]\n", + "[2023-11-08 11:19:14] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:19:14] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7540615200996399\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 9\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2479\n", + "│ │ ├── Epoch N-1 = 0.2517 (\u001B[32m↘ -0.0037\u001B[0m)\n", + "│ │ └── Best until now = 0.2517 (\u001B[32m↘ -0.0037\u001B[0m)\n", + "│ ├── Target_iou = 0.7895\n", + "│ │ ├── Epoch N-1 = 0.7861 (\u001B[32m↗ 0.0034\u001B[0m)\n", + "│ │ └── Best until now = 0.7861 (\u001B[32m↗ 0.0034\u001B[0m)\n", + "│ ├── Background_iou = 0.7109\n", + "│ │ ├── Epoch N-1 = 0.7112 (\u001B[31m↘ -0.0003\u001B[0m)\n", + "│ │ └── Best until now = 0.7112 (\u001B[31m↘ -0.0003\u001B[0m)\n", + "│ └── Mean_iou = 0.7502\n", + "│ ├── Epoch N-1 = 0.7487 (\u001B[32m↗ 0.0015\u001B[0m)\n", + "│ └── Best until now = 0.7487 (\u001B[32m↗ 0.0015\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3133\n", + " │ ├── Epoch N-1 = 0.314 (\u001B[32m↘ -0.0007\u001B[0m)\n", + " │ └── Best until now = 0.314 (\u001B[32m↘ -0.0007\u001B[0m)\n", + " ├── Target_iou = 0.7541\n", + " │ ├── Epoch N-1 = 0.7536 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " │ └── Best until now = 0.7536 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " ├── Background_iou = 0.5458\n", + " │ ├── Epoch N-1 = 0.5448 (\u001B[32m↗ 0.001\u001B[0m)\n", + " │ └── Best until now = 0.5448 (\u001B[32m↗ 0.001\u001B[0m)\n", + " └── Mean_iou = 0.6499\n", + " ├── Epoch N-1 = 0.6492 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " └── Best until now = 0.6492 (\u001B[32m↗ 0.0007\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 10: 100%|██████████| 309/309 [01:58<00:00, 2.61it/s, BCEDiceLoss=0.241, background_IOU=0.724, gpu_mem=1.14, mean_IOU=0.76, target_IOU=0.796]\n", + "Validating epoch 10: 100%|██████████| 65/65 [00:16<00:00, 3.98it/s]\n", + "[2023-11-08 11:21:31] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:21:31] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7544161081314087\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 10\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2409\n", + "│ │ ├── Epoch N-1 = 0.2479 (\u001B[32m↘ -0.007\u001B[0m)\n", + "│ │ └── Best until now = 0.2479 (\u001B[32m↘ -0.007\u001B[0m)\n", + "│ ├── Target_iou = 0.7962\n", + "│ │ ├── Epoch N-1 = 0.7895 (\u001B[32m↗ 0.0068\u001B[0m)\n", + "│ │ └── Best until now = 0.7895 (\u001B[32m↗ 0.0068\u001B[0m)\n", + "│ ├── Background_iou = 0.7243\n", + "│ │ ├── Epoch N-1 = 0.7109 (\u001B[32m↗ 0.0134\u001B[0m)\n", + "│ │ └── Best until now = 0.7112 (\u001B[32m↗ 0.0132\u001B[0m)\n", + "│ └── Mean_iou = 0.7603\n", + "│ ├── Epoch N-1 = 0.7502 (\u001B[32m↗ 0.0101\u001B[0m)\n", + "│ └── Best until now = 0.7502 (\u001B[32m↗ 0.0101\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3126\n", + " │ ├── Epoch N-1 = 0.3133 (\u001B[32m↘ -0.0007\u001B[0m)\n", + " │ └── Best until now = 0.3133 (\u001B[32m↘ -0.0007\u001B[0m)\n", + " ├── Target_iou = 0.7544\n", + " │ ├── Epoch N-1 = 0.7541 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " │ └── Best until now = 0.7541 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " ├── Background_iou = 0.5464\n", + " │ ├── Epoch N-1 = 0.5458 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.5458 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " └── Mean_iou = 0.6504\n", + " ├── Epoch N-1 = 0.6499 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " └── Best until now = 0.6499 (\u001B[32m↗ 0.0005\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 11: 100%|██████████| 309/309 [01:56<00:00, 2.64it/s, BCEDiceLoss=0.235, background_IOU=0.73, gpu_mem=1.14, mean_IOU=0.764, target_IOU=0.799]\n", + "Validating epoch 11: 100%|██████████| 65/65 [00:16<00:00, 4.00it/s]\n", + "[2023-11-08 11:23:47] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:23:47] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7546034455299377\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 11\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2349\n", + "│ │ ├── Epoch N-1 = 0.2409 (\u001B[32m↘ -0.006\u001B[0m)\n", + "│ │ └── Best until now = 0.2409 (\u001B[32m↘ -0.006\u001B[0m)\n", + "│ ├── Target_iou = 0.7988\n", + "│ │ ├── Epoch N-1 = 0.7962 (\u001B[32m↗ 0.0025\u001B[0m)\n", + "│ │ └── Best until now = 0.7962 (\u001B[32m↗ 0.0025\u001B[0m)\n", + "│ ├── Background_iou = 0.7297\n", + "│ │ ├── Epoch N-1 = 0.7243 (\u001B[32m↗ 0.0053\u001B[0m)\n", + "│ │ └── Best until now = 0.7243 (\u001B[32m↗ 0.0053\u001B[0m)\n", + "│ └── Mean_iou = 0.7642\n", + "│ ├── Epoch N-1 = 0.7603 (\u001B[32m↗ 0.0039\u001B[0m)\n", + "│ └── Best until now = 0.7603 (\u001B[32m↗ 0.0039\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3118\n", + " │ ├── Epoch N-1 = 0.3126 (\u001B[32m↘ -0.0008\u001B[0m)\n", + " │ └── Best until now = 0.3126 (\u001B[32m↘ -0.0008\u001B[0m)\n", + " ├── Target_iou = 0.7546\n", + " │ ├── Epoch N-1 = 0.7544 (\u001B[32m↗ 0.0002\u001B[0m)\n", + " │ └── Best until now = 0.7544 (\u001B[32m↗ 0.0002\u001B[0m)\n", + " ├── Background_iou = 0.5468\n", + " │ ├── Epoch N-1 = 0.5464 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " │ └── Best until now = 0.5464 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " └── Mean_iou = 0.6507\n", + " ├── Epoch N-1 = 0.6504 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " └── Best until now = 0.6504 (\u001B[32m↗ 0.0003\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 12: 100%|██████████| 309/309 [01:56<00:00, 2.65it/s, BCEDiceLoss=0.227, background_IOU=0.733, gpu_mem=1.14, mean_IOU=0.769, target_IOU=0.805]\n", + "Validating epoch 12: 100%|██████████| 65/65 [00:16<00:00, 3.95it/s]\n", + "[2023-11-08 11:26:03] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:26:03] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7549036145210266\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 12\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2272\n", + "│ │ ├── Epoch N-1 = 0.2349 (\u001B[32m↘ -0.0077\u001B[0m)\n", + "│ │ └── Best until now = 0.2349 (\u001B[32m↘ -0.0077\u001B[0m)\n", + "│ ├── Target_iou = 0.8051\n", + "│ │ ├── Epoch N-1 = 0.7988 (\u001B[32m↗ 0.0064\u001B[0m)\n", + "│ │ └── Best until now = 0.7988 (\u001B[32m↗ 0.0064\u001B[0m)\n", + "│ ├── Background_iou = 0.7333\n", + "│ │ ├── Epoch N-1 = 0.7297 (\u001B[32m↗ 0.0036\u001B[0m)\n", + "│ │ └── Best until now = 0.7297 (\u001B[32m↗ 0.0036\u001B[0m)\n", + "│ └── Mean_iou = 0.7692\n", + "│ ├── Epoch N-1 = 0.7642 (\u001B[32m↗ 0.005\u001B[0m)\n", + "│ └── Best until now = 0.7642 (\u001B[32m↗ 0.005\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.311\n", + " │ ├── Epoch N-1 = 0.3118 (\u001B[32m↘ -0.0008\u001B[0m)\n", + " │ └── Best until now = 0.3118 (\u001B[32m↘ -0.0008\u001B[0m)\n", + " ├── Target_iou = 0.7549\n", + " │ ├── Epoch N-1 = 0.7546 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " │ └── Best until now = 0.7546 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " ├── Background_iou = 0.5474\n", + " │ ├── Epoch N-1 = 0.5468 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.5468 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " └── Mean_iou = 0.6512\n", + " ├── Epoch N-1 = 0.6507 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " └── Best until now = 0.6507 (\u001B[32m↗ 0.0005\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 13: 100%|██████████| 309/309 [01:56<00:00, 2.65it/s, BCEDiceLoss=0.223, background_IOU=0.74, gpu_mem=1.14, mean_IOU=0.774, target_IOU=0.807]\n", + "Validating epoch 13: 100%|██████████| 65/65 [00:16<00:00, 3.99it/s]\n", + "[2023-11-08 11:28:18] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:28:18] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7553263306617737\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 13\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2229\n", + "│ │ ├── Epoch N-1 = 0.2272 (\u001B[32m↘ -0.0043\u001B[0m)\n", + "│ │ └── Best until now = 0.2272 (\u001B[32m↘ -0.0043\u001B[0m)\n", + "│ ├── Target_iou = 0.8075\n", + "│ │ ├── Epoch N-1 = 0.8051 (\u001B[32m↗ 0.0023\u001B[0m)\n", + "│ │ └── Best until now = 0.8051 (\u001B[32m↗ 0.0023\u001B[0m)\n", + "│ ├── Background_iou = 0.7402\n", + "│ │ ├── Epoch N-1 = 0.7333 (\u001B[32m↗ 0.0068\u001B[0m)\n", + "│ │ └── Best until now = 0.7333 (\u001B[32m↗ 0.0068\u001B[0m)\n", + "│ └── Mean_iou = 0.7738\n", + "│ ├── Epoch N-1 = 0.7692 (\u001B[32m↗ 0.0046\u001B[0m)\n", + "│ └── Best until now = 0.7692 (\u001B[32m↗ 0.0046\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3103\n", + " │ ├── Epoch N-1 = 0.311 (\u001B[32m↘ -0.0007\u001B[0m)\n", + " │ └── Best until now = 0.311 (\u001B[32m↘ -0.0007\u001B[0m)\n", + " ├── Target_iou = 0.7553\n", + " │ ├── Epoch N-1 = 0.7549 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " │ └── Best until now = 0.7549 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " ├── Background_iou = 0.548\n", + " │ ├── Epoch N-1 = 0.5474 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.5474 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " └── Mean_iou = 0.6517\n", + " ├── Epoch N-1 = 0.6512 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " └── Best until now = 0.6512 (\u001B[32m↗ 0.0005\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 14: 100%|██████████| 309/309 [01:56<00:00, 2.65it/s, BCEDiceLoss=0.219, background_IOU=0.744, gpu_mem=1.14, mean_IOU=0.777, target_IOU=0.811]\n", + "Validating epoch 14: 100%|██████████| 65/65 [00:16<00:00, 3.97it/s]\n", + "[2023-11-08 11:30:33] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", + "[2023-11-08 11:30:33] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7558110952377319\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 14\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2193\n", + "│ │ ├── Epoch N-1 = 0.2229 (\u001B[32m↘ -0.0036\u001B[0m)\n", + "│ │ └── Best until now = 0.2229 (\u001B[32m↘ -0.0036\u001B[0m)\n", + "│ ├── Target_iou = 0.8107\n", + "│ │ ├── Epoch N-1 = 0.8075 (\u001B[32m↗ 0.0033\u001B[0m)\n", + "│ │ └── Best until now = 0.8075 (\u001B[32m↗ 0.0033\u001B[0m)\n", + "│ ├── Background_iou = 0.7443\n", + "│ │ ├── Epoch N-1 = 0.7402 (\u001B[32m↗ 0.0041\u001B[0m)\n", + "│ │ └── Best until now = 0.7402 (\u001B[32m↗ 0.0041\u001B[0m)\n", + "│ └── Mean_iou = 0.7775\n", + "│ ├── Epoch N-1 = 0.7738 (\u001B[32m↗ 0.0037\u001B[0m)\n", + "│ └── Best until now = 0.7738 (\u001B[32m↗ 0.0037\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3095\n", + " │ ├── Epoch N-1 = 0.3103 (\u001B[32m↘ -0.0008\u001B[0m)\n", + " │ └── Best until now = 0.3103 (\u001B[32m↘ -0.0008\u001B[0m)\n", + " ├── Target_iou = 0.7558\n", + " │ ├── Epoch N-1 = 0.7553 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " │ └── Best until now = 0.7553 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " ├── Background_iou = 0.5487\n", + " │ ├── Epoch N-1 = 0.548 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " │ └── Best until now = 0.548 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " └── Mean_iou = 0.6523\n", + " ├── Epoch N-1 = 0.6517 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " └── Best until now = 0.6517 (\u001B[32m↗ 0.0006\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-08 11:30:39] INFO - sg_trainer.py - RUNNING ADDITIONAL TEST ON THE AVERAGED MODEL...\n", + "Validating epoch 15: 98%|█████████▊| 64/65 [00:16<00:00, 4.31it/s]" + ] + } + ], + "source": [ + "trainer.train(model=model, training_params=train_params, train_loader=train_loader, valid_loader=valid_loader)" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": { + "id": "X8BJq1crcbjl" + }, + "outputs": [], + "source": [ + "print(\"Best Checkpoint mIoU is: \"+ str(trainer.best_metric))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3Nybj15cchxd" + }, + "source": [ + "Now you can download your trained weights from this directory" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "id": "_iHsFgPSciQh" + }, + "outputs": [], + "source": [ + "print(trainer.checkpoints_dir_path)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "yuhYeXLA18q5" + }, + "source": [ + "# 6. Predict\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "VjRA1tu1mvXQ" + }, + "source": [ + "When the training is complete you can use the trained model to get predictions on the validation set, your data or some other image. Let's load some image and\n", + "run a model inference to create a binary segmentation mask." + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "id": "Ads7RyGN2JwQ", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "outputId": "a0da9ef8-2743-46a2-c03a-95875ab80dc8" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "\rValidating epoch 15: 100%|██████████| 65/65 [00:16<00:00, 4.37it/s]\rValidating epoch 15: 100%|██████████| 65/65 [00:16<00:00, 3.91it/s]\n", + "[2023-11-08 11:30:56] INFO - base_sg_logger.py - [CLEANUP] - Successfully stopped system monitoring process\n", + "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py:283: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n", + " and should_run_async(code)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Best Checkpoint mIoU is: 0.7558110952377319\n", + "/home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244\n" + ] + } + ], + "source": [ + "from torchvision.transforms import Compose, ToTensor, Resize, Normalize, ToPILImage\n", + "from PIL import Image\n", + "import torch\n", + "\n", + "pre_proccess = Compose([\n", + " ToTensor(),\n", + " Normalize([.485, .456, .406], [.229, .224, .225])\n", + "])\n", + "\n", + "demo_img_path = \"/home/data/supervisely-persons/images/ache-adult-depression-expression-41253.png\"\n", + "img = Image.open(demo_img_path)\n", + "# Resize the image and display\n", + "img = Resize(size=(480, 320))(img)\n", + "display(img)\n", + "\n", + "# Run pre-proccess - transforms to tensor and apply normalizations.\n", + "img = pre_proccess(img).unsqueeze(0).cuda()\n", + "\n", + "# Run inference\n", + "model = trainer.net\n", + "model = model.eval()\n", + "mask = model(img)\n", + "\n", + "# Run post-proccess - apply sigmoid to output probabilities, then apply hard\n", + "# threshold of 0.5 for binary mask prediction.\n", + "mask = torch.sigmoid(mask).gt(0.5).squeeze()\n", + "mask = ToPILImage()(mask.float())\n", + "display(mask)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-k6ZLKHL1hIM" + }, + "source": [ + "# 7. Convert to ONNX/TensorRT" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "br7n55Szm4Nq" + }, + "source": [ + "SG is a production ready library. All the models implemented in SG can be compiled to ONNX and TensorRT. Deci also offers the [Infery](https://docs.deci.ai/docs/installing-infery) library that allows to do inference on models saved in various frameworks with the same API regardless of a framework.\n", + "\n", + "Let's compile our model to ONNX." + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "id": "dsIPbyX6GVKs", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "a2c3f05e-cb84-4fc0-db8a-e59bc72faf80" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\u001B[31mERROR: Could not find a version that satisfies the requirement infery (from versions: none)\u001B[0m\u001B[31m\n", + "\u001B[0m\u001B[31mERROR: No matching distribution found for infery\u001B[0m\u001B[31m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m776.3/776.3 MB\u001B[0m \u001B[31m1.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", + "torchaudio 2.1.0+cu118 requires torch==2.1.0, but you have torch 1.12.0 which is incompatible.\n", + "torchdata 0.7.0 requires torch==2.1.0, but you have torch 1.12.0 which is incompatible.\n", + "torchtext 0.16.0 requires torch==2.1.0, but you have torch 1.12.0 which is incompatible.\n", + "torchvision 0.16.0+cu118 requires torch==2.1.0, but you have torch 1.12.0 which is incompatible.\u001B[0m\u001B[31m\n", + "\u001B[0m" + ] + } + ], + "source": [ + "! pip install -qq onnx-simplifier\n", + "! pip install -qq infery\n", + "! pip install -qq torch==1.12" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "q0AGQvEf11PT" + }, + "outputs": [], + "source": [ + "from onnxsim import simplify\n", + "import onnx\n", + "\n", + "onnx_path = \"/home/data/model.onnx\"\n", + "\n", + "input_size = [1, 3, 480, 320]\n", + "model.prep_model_for_conversion(input_size=input_size)\n", + "\n", + "torch.onnx.export(model,\n", + " torch.randn(*input_size).cuda(),\n", + " onnx_path,\n", + " opset_version=11)\n", + "\n", + "# onnx simplifier\n", + "model_sim, check = simplify(onnx_path)\n", + "assert check, \"Simplified ONNX model could not be validated\"\n", + "onnx.save_model(model_sim, onnx_path)\n", + "\n", + "print(\"ONNX successfully created at: \", onnx_path)\n", + "\n" + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/notebooks/segmentation_connect_custom_dataset.ipynb b/notebooks/segmentation_connect_custom_dataset.ipynb new file mode 100644 index 0000000000..2e3b465b5b --- /dev/null +++ b/notebooks/segmentation_connect_custom_dataset.ipynb @@ -0,0 +1,980 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "sh6t_y7KzqBH" + }, + "source": [ + "![SG - Horizontal.png](data:image/png;base64,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)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "5aISf1B-AGDQ" + }, + "source": [ + "# SuperGradients Semantic Segmentation How to Connect Custom Dataset\n", + "\n", + "In this tutorial we will explore how you can connect your custom Semantic Segmentation dataset to SG.\n", + "\n", + "Since SG trainer is fully compatible with PyTorch data loaders, we will demonstrate how to build one and use it.\n", + "\n", + "The notebook is divided into 5 sections:\n", + "1. Experiment setup\n", + "2. Dataset definition: create a proxy dataset and create a dataloader\n", + "3. Architecture definition: pre-trained PPLiteSeg on Cityscapes \n", + "4. Training setup\n", + "5. Training and Evaluation\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-1nPOPmc1lGp" + }, + "source": [ + "#Install SG" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "VAssbjJw7Yt1" + }, + "source": [ + "The cell below will install **super_gradients** which will automatically get all its dependencies. Let's import all the installed libraries to make sure they installed succesfully." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "JKce1SM6voVH", + "outputId": "a6397510-a140-443f-f13c-eec1272cc1a8" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Requirement already satisfied: torch in /usr/local/lib/python3.10/dist-packages (2.1.0+cu118)\n", + "Requirement already satisfied: torchvision in /usr/local/lib/python3.10/dist-packages (0.16.0+cu118)\n", + "Requirement already satisfied: torchaudio in /usr/local/lib/python3.10/dist-packages (2.1.0+cu118)\n", + "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch) (3.13.1)\n", + "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.10/dist-packages (from torch) (4.5.0)\n", + "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch) (1.12)\n", + "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch) (3.2.1)\n", + "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch) (3.1.2)\n", + "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from torch) (2023.6.0)\n", + "Requirement already satisfied: triton==2.1.0 in /usr/local/lib/python3.10/dist-packages (from torch) (2.1.0)\n", + "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from torchvision) (1.23.5)\n", + "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from torchvision) (2.31.0)\n", + "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision) (9.4.0)\n", + "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch) (2.1.3)\n", + "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.3.2)\n", + "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.4)\n", + "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2.0.7)\n", + "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2023.7.22)\n", + "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch) (1.3.0)\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n", + " Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for pycocotools (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for termcolor (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for treelib (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for coverage (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for xhtml2pdf (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for antlr4-python3-runtime (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for stringcase (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for svglib (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", + "lida 0.0.10 requires fastapi, which is not installed.\n", + "lida 0.0.10 requires kaleido, which is not installed.\n", + "lida 0.0.10 requires python-multipart, which is not installed.\n", + "lida 0.0.10 requires uvicorn, which is not installed.\n", + "tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.23.0 which is incompatible.\u001b[0m\u001b[31m\n", + "\u001b[0m" + ] + } + ], + "source": [ + "! pip install torch torchvision torchaudio\n", + "! pip install -qq super-gradients==3.4.1\n", + "! pip install -qq prettyformatter" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "njthhNJR1pJm" + }, + "source": [ + "# 1. Experiment setup" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "YPym4wvpOcOJ" + }, + "source": [ + "We will first initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", + "\n", + "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", + "\n", + "```\n", + "ckpt_root_dir\n", + "|─── experiment_name_1\n", + "│ ckpt_best.pth # Model checkpoint on best epoch\n", + "│ ckpt_latest.pth # Model checkpoint on last epoch\n", + "│ average_model.pth # Model checkpoint averaged over epochs\n", + "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", + "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", + "└─── experiment_name_2\n", + " ...\n", + "```\n", + "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", + " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`.\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "A2PlnTWpimnH" + }, + "source": [ + "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "id": "_v1N3kXs3wo1" + }, + "outputs": [], + "source": [ + "from super_gradients.training import Trainer, MultiGPUMode\n", + "\n", + "\n", + "CHECKPOINT_DIR = '/home/notebook_ckpts/'\n", + "trainer = Trainer(experiment_name='transfer_learning_semantic_segementation_ppLite', ckpt_root_dir=CHECKPOINT_DIR)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "J9ZaMulSvwhr" + }, + "source": [ + "# 2. Dataset definition\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "_1TXuJKkKzFJ" + }, + "source": [ + "## 2.A Generate Proxy Dataset" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Y7us7VHRig7M" + }, + "source": [ + "\n", + "A proxy dataset generation is available merely to demonstrate an end-to-end training pipeline in this notebook.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "wbdVYnIyjgv-" + }, + "outputs": [], + "source": [ + "from PIL import Image\n", + "import os\n", + "import numpy as np\n", + "\n", + "\n", + "# creation of proxy dataset to demonstrate usage\n", + "def generate_proxy_dataset(write_path: str, num_samples: int, num_classes: int, img_size: int = 256):\n", + " # Create training files and text\n", + " os.makedirs(os.path.join(write_path, 'images', 'train'), exist_ok=True)\n", + " os.makedirs(os.path.join(write_path, 'images', 'val'), exist_ok=True)\n", + " os.makedirs(os.path.join(write_path, 'labels', 'train'), exist_ok=True)\n", + " os.makedirs(os.path.join(write_path, 'labels', 'val'), exist_ok=True)\n", + "\n", + " train_fp = open(os.path.join(write_path, 'train.txt'), 'w')\n", + " val_fp = open(os.path.join(write_path, 'val.txt'), 'w')\n", + "\n", + " # Create random samples\n", + " for n in range(num_samples):\n", + " img = np.random.rand(img_size, img_size, 3) * 255\n", + " img = Image.fromarray(img.astype('uint8')).convert('RGB')\n", + "\n", + " lbl = np.random.randint(0, num_classes, size=(img_size, img_size))\n", + " lbl = Image.fromarray(lbl.astype('uint8')).convert('L')\n", + "\n", + " im_string = '%000d.jpg' % n\n", + " lbl_string = '%000d.png' % n\n", + "\n", + " img_train_fn = os.path.join(write_path, 'images', 'train', im_string)\n", + " img_val_fn = img_train_fn.replace(\"train\", \"val\")\n", + " img.save(img_train_fn)\n", + " img.save(img_val_fn)\n", + "\n", + " lbl_train_fn = os.path.join(write_path, 'labels', 'train', lbl_string)\n", + " lbl_val_fn = lbl_train_fn.replace(\"train\", \"val\")\n", + " lbl.save(lbl_train_fn)\n", + " lbl.save(lbl_val_fn)\n", + "\n", + " train_fp.write(f\"{img_train_fn} {lbl_train_fn}\\n\")\n", + " val_fp.write(f\"{img_val_fn} {lbl_val_fn}\\n\")\n", + "\n", + " train_fp.close()\n", + " val_fp.close()" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "id": "DXu4yfuZoiv0", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "dccaf4ba-159f-4a47-d13d-ba4b60eaac80" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Train file `train.txt` content: \n", + "/content/example_data/images/train/0.jpg /content/example_data/labels/train/0.png\n", + "/content/example_data/images/train/1.jpg /content/example_data/labels/train/1.png\n", + "/content/example_data/images/train/2.jpg /content/example_data/labels/train/2.png\n", + "/content/example_data/images/train/3.jpg /content/example_data/labels/train/3.png\n", + "/content/example_data/images/train/4.jpg /content/example_data/labels/train/4.png\n", + "/content/example_data/images/train/5.jpg /content/example_data/labels/train/5.png\n", + "/content/example_data/images/train/6.jpg /content/example_data/labels/train/6.png\n", + "/content/example_data/images/train/7.jpg /content/example_data/labels/train/7.png\n", + "/content/example_data/images/train/8.jpg /content/example_data/labels/train/8.png\n", + "/content/example_data/images/train/9.jpg /content/example_data/labels/train/9.png\n" + ] + } + ], + "source": [ + "num_classes = 10\n", + "generate_proxy_dataset('/content/example_data', num_samples=10, num_classes=num_classes)\n", + "\n", + "print(\"Train file `train.txt` content: \")\n", + "! cat /content/example_data/train.txt" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "MDksFYrIqClt" + }, + "source": [ + "## 2.B Create Torch Dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "id": "AGziBKSIqaUu" + }, + "outputs": [], + "source": [ + "import torch\n", + "from torch.utils.data import Dataset\n", + "from torchvision import transforms, utils\n", + "\n", + "\n", + "class CustomDataset(Dataset):\n", + " \"\"\"\n", + " A PyTorch Dataset class to be used in a PyTorch DataLoader to create batches.\n", + " \"\"\"\n", + "\n", + " def __init__(self, data_folder, split):\n", + " \"\"\"\n", + " :param data_folder: folder where data files are stored\n", + " :param split: split, one of 'TRAIN' or 'TEST'\n", + " \"\"\"\n", + " self.data_folder = data_folder\n", + " self.split = split.lower()\n", + " assert self.split in {'train', 'val'}\n", + "\n", + " # Read data files\n", + " with open(os.path.join(data_folder, self.split + '.txt'), 'r') as f:\n", + " data_lines = f.readlines()\n", + " self.samples_fn = [line.strip().split(\" \") for line in data_lines]\n", + "\n", + " self.transforms = transforms.Compose([transforms.ToTensor()])\n", + "\n", + " def __getitem__(self, i):\n", + " # Read image and label\n", + " image = Image.open(self.samples_fn[i][0]).convert('RGB')\n", + " label = Image.open(self.samples_fn[i][1])\n", + "\n", + " image_tensor = self.transforms(image)\n", + " label_tensor = torch.from_numpy(np.array(label)).long()\n", + "\n", + " return image_tensor, label_tensor\n", + "\n", + "\n", + " def __len__(self):\n", + " return len(self.samples_fn)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "id": "2B0hlas_1Rh-" + }, + "outputs": [], + "source": [ + "train_dataset = CustomDataset(\"/content/example_data\", split=\"train\")\n", + "val_dataset = CustomDataset(\"/content/example_data\", split=\"val\")\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "eIG5tsiuor9E" + }, + "source": [ + "Let's have a look at the first sample:" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "id": "ZsHqcq1jpN0F", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "e9c1182a-5359-45b6-c0f3-ad430a4fc67d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "torch.Size([3, 256, 256]) torch.Size([256, 256])\n", + "tensor([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])\n" + ] + } + ], + "source": [ + "img, lbl = train_dataset[0]\n", + "print(img.shape, lbl.shape)\n", + "print(torch.unique(lbl))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "aWfFrYLzo9j8" + }, + "source": [ + "## 2.C Create Torch Dataloader" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "D3ThxDIopDDB" + }, + "source": [] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": { + "id": "XrWjWfjXnw_r" + }, + "outputs": [], + "source": [ + "from torch.utils.data import Dataset, DataLoader\n", + "\n", + "train_dataloader = DataLoader(train_dataset, batch_size=4, shuffle=True, num_workers=2)\n", + "val_dataloader = DataLoader(val_dataset, batch_size=4, shuffle=False, num_workers=2)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vB1sGPO8qwZJ" + }, + "source": [ + "Lets' have a look at the first batch:\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "id": "O-KuZQ3XBduM", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "489fc05e-f972-464d-c150-360e4f2dbd7e" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "torch.Size([4, 3, 256, 256])" + ] + }, + "metadata": {}, + "execution_count": 10 + } + ], + "source": [ + "next(iter(train_dataloader))[0].shape" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fFfvyMHU32QF" + }, + "source": [ + "\n", + "# 3. Architecture definition" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "EpqgjQjl4awr" + }, + "source": [ + "SG includes implementations of many different architectures for object detection tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "GNM64JAa4sbF" + }, + "source": [ + "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", + "and extra Auxiliary heads aren't used for training.\n", + "\n", + "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "id": "YDK4btf04Gbu", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "cc7f1ab1-3a01-49c1-c9a2-6ca434dcc192" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Downloading: \"https://sghub.deci.ai/models/pp_lite_t_seg75_cityscapes.pth\" to /root/.cache/torch/hub/checkpoints/pp_lite_t_seg75_cityscapes.pth\n", + "100%|██████████| 31.4M/31.4M [00:01<00:00, 32.4MB/s]\n", + "[2023-11-12 14:41:45] INFO - checkpoint_utils.py - Successfully loaded pretrained weights for architecture pp_lite_t_seg75\n" + ] + } + ], + "source": [ + "from super_gradients.training import models\n", + "from super_gradients.common.object_names import Models\n", + "\n", + "model = models.get(model_name=Models.PP_LITE_T_SEG75,\n", + " arch_params={\"use_aux_heads\": False},\n", + " num_classes=num_classes,\n", + " pretrained_weights=\"cityscapes\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "40UcYJ3u5JyF" + }, + "source": [ + "That being said, SG allows you to use one of SG implemented architectures or your custom architecture, as long as it inherits torch.nn.Module." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "LYPVR-XM4GsZ" + }, + "source": [ + "# 4. Training setup\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6K_56lDV8azX" + }, + "source": [ + "\n", + "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "id": "3eRe0hBz4G1n" + }, + "outputs": [], + "source": [ + "from super_gradients.training.metrics.segmentation_metrics import IoU\n", + "from super_gradients.training.utils.callbacks import BinarySegmentationVisualizationCallback, Phase\n", + "\n", + "\n", + "train_params = {\"max_epochs\": 10,\n", + " \"lr_mode\": \"cosine\",\n", + " \"initial_lr\": 0.005,\n", + " \"optimizer\": \"SGD\",\n", + " \"loss\": \"cross_entropy\",\n", + " \"average_best_models\": False,\n", + " \"metric_to_watch\": \"IoU\",\n", + " \"greater_metric_to_watch_is_better\": True,\n", + " \"train_metrics_list\": [IoU(num_classes=10)],\n", + " \"valid_metrics_list\": [IoU(num_classes=10)],\n", + " \"loss_logging_items_names\": [\"loss\"],\n", + " \"phase_callbacks\": [BinarySegmentationVisualizationCallback(phase=Phase.VALIDATION_BATCH_END,\n", + " freq=1,\n", + " last_img_idx_in_batch=4)],\n", + "\n", + " }" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "D3tVVUhy4OqP" + }, + "source": [ + "# 5. Training and evaluation\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "8tKUuxbe9NlQ" + }, + "source": [ + "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", + "\n", + "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders." + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "id": "-Ojnc1bk9L3s", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "b36b8b9b-b554-444e-d440-02bf623c3efa" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-12 14:41:51] WARNING - sg_trainer.py - Train dataset size % batch_size != 0 and drop_last=False, this might result in smaller last batch.\n", + "[2023-11-12 14:41:58] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231112_144158_892860`\n", + "[2023-11-12 14:41:58] INFO - sg_trainer.py - Checkpoints directory: /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860\n", + "/usr/local/lib/python3.10/dist-packages/super_gradients/common/registry/registry.py:72: DeprecationWarning: Object name `cross_entropy` is now deprecated. Please replace it with `CrossEntropyLoss`.\n", + " warnings.warn(f\"Object name `{name}` is now deprecated. Please replace it with `{deprecated_names[name]}`.\", DeprecationWarning)\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is now moved to /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860/console_Nov12_14_41_58.txt\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-12 14:41:59] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", + " - Mode: Single GPU\n", + " - Number of GPUs: 1 (1 available on the machine)\n", + " - Full dataset size: 10 (len(train_set))\n", + " - Batch size per GPU: 4 (batch_size)\n", + " - Batch Accumulate: 1 (batch_accumulate)\n", + " - Total batch size: 4 (num_gpus * batch_size)\n", + " - Effective Batch size: 4 (num_gpus * batch_size * batch_accumulate)\n", + " - Iterations per epoch: 3 (len(train_loader))\n", + " - Gradient updates per epoch: 3 (len(train_loader) / batch_accumulate)\n", + "\n", + "[2023-11-12 14:41:59] INFO - sg_trainer.py - Started training for 10 epochs (0/9)\n", + "\n", + "Train epoch 0: 100%|██████████| 3/3 [00:08<00:00, 2.91s/it, CrossEntropyLoss=3.49, IoU=0.0319, gpu_mem=0.686]\n", + "Validating: 100%|██████████| 3/3 [00:00<00:00, 3.83it/s]\n", + "[2023-11-12 14:42:09] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860/ckpt_best.pth\n", + "[2023-11-12 14:42:09] INFO - sg_trainer.py - Best checkpoint overriden: validation IoU: 0.013753225095570087\n", + "Train epoch 1: 0%| | 0/3 [00:00=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision) (9.4.0)\n", + "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch) (2.1.3)\n", + "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.3.2)\n", + "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.4)\n", + "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2.0.7)\n", + "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2023.7.22)\n", + "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch) (1.3.0)\n", + " Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", + " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m135.8/135.8 kB\u001B[0m \u001B[31m3.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m684.5/684.5 kB\u001B[0m \u001B[31m21.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.9/2.9 MB\u001B[0m \u001B[31m71.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m102.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m408.6/408.6 kB\u001B[0m \u001B[31m48.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m154.5/154.5 kB\u001B[0m \u001B[31m23.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.5/79.5 kB\u001B[0m \u001B[31m12.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m4.5/4.5 MB\u001B[0m \u001B[31m108.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m13.5/13.5 MB\u001B[0m \u001B[31m96.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m68.0/68.0 kB\u001B[0m \u001B[31m9.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", + " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m17.0/17.0 MB\u001B[0m \u001B[31m83.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m3.3/3.3 MB\u001B[0m \u001B[31m54.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.2/2.2 MB\u001B[0m \u001B[31m81.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m458.9/458.9 kB\u001B[0m \u001B[31m41.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.0/46.0 kB\u001B[0m \u001B[31m4.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m11.3/11.3 MB\u001B[0m \u001B[31m76.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.8/79.8 kB\u001B[0m \u001B[31m6.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m108.3/108.3 kB\u001B[0m \u001B[31m13.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m176.0/176.0 kB\u001B[0m \u001B[31m21.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m407.7/407.7 kB\u001B[0m \u001B[31m35.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m107.7/107.7 kB\u001B[0m \u001B[31m13.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m277.4/277.4 kB\u001B[0m \u001B[31m25.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m68.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m913.9/913.9 kB\u001B[0m \u001B[31m59.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m117.0/117.0 kB\u001B[0m \u001B[31m14.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m575.5/575.5 kB\u001B[0m \u001B[31m52.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m121.1/121.1 kB\u001B[0m \u001B[31m18.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m86.8/86.8 kB\u001B[0m \u001B[31m12.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m13.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m15.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.6/120.6 kB\u001B[0m \u001B[31m17.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m11.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m12.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m84.7/84.7 kB\u001B[0m \u001B[31m12.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m13.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m13.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.8/99.8 kB\u001B[0m \u001B[31m12.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m12.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m11.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m90.6/90.6 kB\u001B[0m \u001B[31m11.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m13.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m12.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m13.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m94.0/94.0 kB\u001B[0m \u001B[31m7.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m105.0/105.0 kB\u001B[0m \u001B[31m12.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.2/46.2 kB\u001B[0m \u001B[31m6.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m106.8/106.8 kB\u001B[0m \u001B[31m15.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m194.6/194.6 kB\u001B[0m \u001B[31m24.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m58.1/58.1 kB\u001B[0m \u001B[31m8.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Building wheel for super-gradients (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for pycocotools (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for termcolor (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for treelib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for coverage (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for xhtml2pdf (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for antlr4-python3-runtime (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for stringcase (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for svglib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", + "lida 0.0.10 requires fastapi, which is not installed.\n", + "lida 0.0.10 requires kaleido, which is not installed.\n", + "lida 0.0.10 requires python-multipart, which is not installed.\n", + "lida 0.0.10 requires uvicorn, which is not installed.\n", + "tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.23.0 which is incompatible.\u001B[0m\u001B[31m\n", + "\u001B[0m" + ] + } + ], + "source": [ + "! pip install torch torchvision torchaudio\n", + "! pip install -qq super-gradients==3.4.1\n", + "! pip install -qq prettyformatter" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "892xArqDsGsQ" + }, + "source": [ + "# 1. Experiment setup\n", + "We will initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", + "\n", + "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", + "\n", + "```\n", + "ckpt_root_dir\n", + "|─── experiment_name_1\n", + "│ ckpt_best.pth # Model checkpoint on best epoch\n", + "│ ckpt_latest.pth # Model checkpoint on last epoch\n", + "│ average_model.pth # Model checkpoint averaged over epochs\n", + "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", + "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", + "└─── experiment_name_2\n", + " ...\n", + "```\n", + "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", + " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "pl0WPz1HisFz" + }, + "source": [ + "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "HAff--HysJmP", + "outputId": "4d3b9778-480b-4b72-ad8a-b3c257962771" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is logged into /root/sg_logs/console.log\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-12 13:59:13] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n", + "[2023-11-12 13:59:13] WARNING - __init__.py - Failed to import pytorch_quantization\n", + "[2023-11-12 13:59:14] INFO - utils.py - NumExpr defaulting to 2 threads.\n", + "[2023-11-12 13:59:28] WARNING - calibrator.py - Failed to import pytorch_quantization\n", + "[2023-11-12 13:59:28] WARNING - export.py - Failed to import pytorch_quantization\n", + "[2023-11-12 13:59:28] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n", + "[2023-11-12 13:59:29] INFO - env_sanity_check.py - Library check is not supported when super_gradients installed through \"git+https://github.com/...\" command\n" + ] + } + ], + "source": [ + "from super_gradients import Trainer\n", + "\n", + "CHECKPOINT_DIR = '/home/notebook_ckpts/'\n", + "trainer = Trainer(experiment_name=\"segmentation_transfer_learning\", ckpt_root_dir=CHECKPOINT_DIR)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "dwVMY4gMjQSL" + }, + "source": [ + "# 2. Dataset definition\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fpIWhnR9j2rm" + }, + "source": [ + "\n", + "For the sake of this presentation, we'll use **Supervisely** semantic segmentation dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ZACgRb-qjzDJ" + }, + "source": [ + "SG trainer is fully compatible with PyTorch data loaders, so you can definitely use your own data for the experiment below if you prefer." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6ulV6Hpao3IN" + }, + "source": [ + "## 2.A. Download data\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mVwslNv-j-2C" + }, + "source": [ + "Feel free to change the download path by editing SUPERVISELY_DATASET_DOWNLOAD_PATH" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "id": "dfR18Rmbo00y", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "72323bbd-b94f-4488-a14e-6a5dee578d62" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Downloading and extracting supervisely dataset to: /home/data\n", + "/home/data\n", + "--2023-11-12 13:59:29-- https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", + "Resolving deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)... 52.217.166.25, 52.217.204.241, 3.5.25.206, ...\n", + "Connecting to deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)|52.217.166.25|:443... connected.\n", + "HTTP request sent, awaiting response... 200 OK\n", + "Length: 3564001012 (3.3G) [application/zip]\n", + "Saving to: ‘supervisely-persons.zip’\n", + "\n", + "supervisely-persons 100%[===================>] 3.32G 61.8MB/s in 62s \n", + "\n", + "2023-11-12 14:00:31 (55.2 MB/s) - ‘supervisely-persons.zip’ saved [3564001012/3564001012]\n", + "\n" + ] + } + ], + "source": [ + "import os\n", + "\n", + "SUPERVISELY_DATASET_DOWNLOAD_PATH=\"/home/data\"\n", + "\n", + "supervisely_dataset_dir_path = SUPERVISELY_DATASET_DOWNLOAD_PATH + os.path.sep + 'supervisely-persons'\n", + "\n", + "if os.path.isdir(supervisely_dataset_dir_path):\n", + " print('supervisely dataset already downloaded...')\n", + "else:\n", + " print('Downloading and extracting supervisely dataset to: ' + SUPERVISELY_DATASET_DOWNLOAD_PATH)\n", + " ! mkdir $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " %cd $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " ! wget https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", + " ! unzip --qq supervisely-persons.zip" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": false, + "id": "V9ZcklupX8Qx" + }, + "source": [ + "## 2.B. Create data loaders\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3Mk_YixjlEhj" + }, + "source": [ + "The dataloaders are initiated with the default parameters defined in the [yaml](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/recipes/dataset_params/supervisely_persons_dataset_params.yaml)\n", + "file. Parameters as batch_size, transforms, root_dir and others can be overridden by passing as `dataset_params` and\n", + "`dataloader_params`, as implemented bellow." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "S3BzMRhSX8Qx" + }, + "outputs": [], + "source": [ + "from super_gradients.training import dataloaders\n", + "\n", + "root_dir = supervisely_dataset_dir_path\n", + "batch_size = 8\n", + "\n", + "train_loader = dataloaders.supervisely_persons_train(\n", + " dataset_params={\"root_dir\": root_dir},\n", + " dataloader_params={\"batch_size\": batch_size, \"num_workers\": 2}\n", + ")\n", + "valid_loader = dataloaders.supervisely_persons_val(\n", + " dataset_params={\"root_dir\": root_dir},\n", + " dataloader_params={\"batch_size\": batch_size, \"num_workers\": 2}\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6dHIwvs46-dk" + }, + "source": [ + "As you can see, we didn't have to pass many parameters into the dataloaders construction. That's because defaults are pre-defined for your convenience, and you might be curious to know what they are. Let's print them and see which resolution and transformations are defined." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "id": "76tzhKxi6aS-" + }, + "outputs": [], + "source": [ + "from prettyformatter import pprint\n", + "\n", + "print('Dataloader parameters:')\n", + "pprint(train_loader.dataloader_params)\n", + "print('Dataset parameters')\n", + "pprint(train_loader.dataset.dataset_params)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "I4QEOkKyy93R" + }, + "source": [ + "We can take a look at some images from the dataset." + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "id": "xXPMJQCJzmb4", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 937 + }, + "outputId": "f7b11090-410d-4d86-8730-b27ebfe385b2" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Dataloader parameters:\n", + "{\"batch_size\": 8, \"num_workers\": 2, \"shuffle\": True, \"drop_last\": True}\n", + "Dataset parameters\n", + "{'root_dir': '/home/data/supervisely-persons', 'list_file': 'train.csv', 'cache_labels': False, 'cache_images': False, 'transforms': [{'SegRandomRescale': {'scales': [0.25, 1.0]}}, {'SegColorJitter': {'brightness': 0.5, 'contrast': 0.5, 'saturation': 0.5}}, {'SegRandomFlip': {'prob': 0.5}}, {'SegPadShortToCropSize': {'crop_size': [320, 480], 'fill_mask': 0}}, {'SegCropImageAndMask': {'crop_size': [320, 480], 'mode': 'random'}}]}\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.10/dist-packages/torchvision/transforms/functional.py:1603: UserWarning: The default value of the antialias parameter of all the resizing transforms (Resize(), RandomResizedCrop(), etc.) will change from None to True in v0.17, in order to be consistent across the PIL and Tensor backends. To suppress this warning, directly pass antialias=True (recommended, future default), antialias=None (current default, which means False for Tensors and True for PIL), or antialias=False (only works on Tensors - PIL will still use antialiasing). This also applies if you are using the inference transforms from the models weights: update the call to weights.transforms(antialias=True).\n", + " warnings.warn(\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "from PIL import Image\n", + "from torchvision.utils import draw_segmentation_masks\n", + "from torchvision.transforms import ToTensor, ToPILImage, Resize\n", + "import numpy as np\n", + "import torch\n", + "\n", + "def plot_seg_data(img_path: str, target_path: str):\n", + " image = (ToTensor()(Image.open(img_path).convert('RGB')) * 255).type(torch.uint8)\n", + " target = torch.from_numpy(np.array(Image.open(target_path))).bool()\n", + " image = draw_segmentation_masks(image, target, colors=\"red\", alpha=0.4)\n", + " image = Resize(size=200)(image)\n", + " display(ToPILImage()(image))\n", + "\n", + "for i in range(4, 7):\n", + " img_path, target_path = train_loader.dataset.samples_targets_tuples_list[i]\n", + " plot_seg_data(img_path, target_path)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "l5GcDAg_pUGJ" + }, + "source": [ + "# 3. Architecture definition\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fU8orO7wlwIK" + }, + "source": [ + "SG includes implementations of many different architectures for semantic segmentation tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-oGSU3V8lqcm" + }, + "source": [ + "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", + "and extra Auxiliary heads aren't used for training.\n", + "\n", + "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "id": "f6ZTsO0nrdje", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "dc6dc47a-402a-4bc1-b5a6-2af9728270dd" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Downloading: \"https://sghub.deci.ai/models/pp_lite_t_seg75_cityscapes.pth\" to /root/.cache/torch/hub/checkpoints/pp_lite_t_seg75_cityscapes.pth\n", + "100%|██████████| 31.4M/31.4M [00:01<00:00, 22.4MB/s]\n", + "[2023-11-12 14:01:13] INFO - checkpoint_utils.py - Successfully loaded pretrained weights for architecture pp_lite_t_seg75\n" + ] + } + ], + "source": [ + "from super_gradients.training import models\n", + "from super_gradients.common.object_names import Models\n", + "\n", + "model = models.get(model_name=Models.PP_LITE_T_SEG75,\n", + " arch_params={\"use_aux_heads\": False},\n", + " num_classes=1,\n", + " pretrained_weights=\"cityscapes\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "X-_dBewgr1dG" + }, + "source": [ + "# 4. Training setup\n", + "\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "H1Rll8Orl-Dy" + }, + "source": [ + "\n", + "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n", + "\n", + "We will be using an average of BCE and Dice loss for segmentation, with different learning rates for the replaced segmentation head layer, and the rest of the network- this is controlled by the `multiply_head_lr` parameter which is the multiplication factor of the learning rate for the newly replaced layer.\n", + "\n", + "As our `metric_to_watch`, we will be monitoring the `target_IOU` which is one of the components of `BinaryIOU` torchmetrics object (the other components are `mean_IOU` which is the mean of the background and target IOUs, and `background_IOU`)." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "id": "NShu3zLgr5qD" + }, + "outputs": [], + "source": [ + "from super_gradients.training.metrics.segmentation_metrics import BinaryIOU\n", + "from super_gradients.training.utils.callbacks import BinarySegmentationVisualizationCallback, Phase\n", + "\n", + "train_params = {\"max_epochs\": 15,\n", + " \"lr_mode\": \"cosine\",\n", + " \"initial_lr\": 0.005,\n", + " \"lr_warmup_epochs\": 5,\n", + " \"multiply_head_lr\": 10,\n", + " \"optimizer\": \"SGD\",\n", + " \"loss\": \"bce_dice_loss\",\n", + " \"ema\": True,\n", + " \"zero_weight_decay_on_bias_and_bn\": True,\n", + " \"average_best_models\": True,\n", + " \"metric_to_watch\": \"target_IOU\",\n", + " \"greater_metric_to_watch_is_better\": True,\n", + " \"train_metrics_list\": [BinaryIOU()],\n", + " \"valid_metrics_list\": [BinaryIOU()],\n", + " \"loss_logging_items_names\": [\"loss\"],\n", + " \"phase_callbacks\": [BinarySegmentationVisualizationCallback(phase=Phase.VALIDATION_BATCH_END,\n", + " freq=1,\n", + " last_img_idx_in_batch=4)],\n", + " }" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qTECVyhcs506" + }, + "source": [ + "# 5. Training and evaluation\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "S1K5MU2kmmDb" + }, + "source": [ + "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", + "\n", + "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders.\n", + "\n", + "**Note:** While training, don't forget to refresh the tensorboard with the arrow on the top right." + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": { + "id": "u6roEj9ktFTi", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "2e9c3992-0566-4cba-e7aa-ae4853cc442e" + }, + "outputs": [ + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "[2023-11-12 14:01:21] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231112_140121_753664`\n", + "[2023-11-12 14:01:21] INFO - sg_trainer.py - Checkpoints directory: /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664\n", + "/usr/local/lib/python3.10/dist-packages/super_gradients/common/registry/registry.py:72: DeprecationWarning: Object name `bce_dice_loss` is now deprecated. Please replace it with `BCEDiceLoss`.\n", + " warnings.warn(f\"Object name `{name}` is now deprecated. Please replace it with `{deprecated_names[name]}`.\", DeprecationWarning)\n", + "[2023-11-12 14:01:21] INFO - sg_trainer.py - Using EMA with params {}\n", + "[2023-11-12 14:01:21] WARNING - ema.py - Parameter `decay` is not specified for EMA params. Please specify `decay` parameter explicitly in your config:\n", + "ema: True\n", + "ema_params: \n", + " decay: 0.9999\n", + " decay_type: exp\n", + " beta: 15\n", + "Will default to decay: 0.9999\n", + "In the next major release of SG this warning will become an error.\n", + "[2023-11-12 14:01:21] WARNING - ema.py - Parameter decay_type is not specified for EMA model. Please specify decay_type parameter explicitly in your config:\n", + "ema: True\n", + "ema_params: \n", + " decay: 0.9999\n", + " decay_type: constant|exp|threshold\n", + "Will default to `exp` decay with beta = 15\n", + "In the next major release of SG this warning will become an error.\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "The console stream is now moved to /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/console_Nov12_14_01_21.txt\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "[2023-11-12 14:01:24] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", + " - Mode: Single GPU\n", + " - Number of GPUs: 1 (1 available on the machine)\n", + " - Full dataset size: 2477 (len(train_set))\n", + " - Batch size per GPU: 8 (batch_size)\n", + " - Batch Accumulate: 1 (batch_accumulate)\n", + " - Total batch size: 8 (num_gpus * batch_size)\n", + " - Effective Batch size: 8 (num_gpus * batch_size * batch_accumulate)\n", + " - Iterations per epoch: 309 (len(train_loader))\n", + " - Gradient updates per epoch: 309 (len(train_loader) / batch_accumulate)\n", + "\n", + "[2023-11-12 14:01:24] INFO - sg_trainer.py - Started training for 15 epochs (0/14)\n", + "\n", + "Train epoch 0: 100%|██████████| 309/309 [01:56<00:00, 2.66it/s, BCEDiceLoss=0.223, background_IOU=0.749, gpu_mem=1.14, mean_IOU=0.779, target_IOU=0.809]\n", + "Validating: 100%|██████████| 65/65 [00:15<00:00, 4.20it/s]\n", + "[2023-11-12 14:03:36] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:03:36] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.857506275177002\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 0\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2225\n", + "│ ├── Target_iou = 0.8094\n", + "│ ├── Background_iou = 0.7486\n", + "│ └── Mean_iou = 0.779\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1821\n", + " ├── Target_iou = 0.8575\n", + " ├── Background_iou = 0.7387\n", + " └── Mean_iou = 0.7981\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 1: 100%|██████████| 309/309 [01:42<00:00, 3.02it/s, BCEDiceLoss=0.167, background_IOU=0.815, gpu_mem=1.14, mean_IOU=0.837, target_IOU=0.86]\n", + "Validating epoch 1: 100%|██████████| 65/65 [00:16<00:00, 3.96it/s]\n", + "[2023-11-12 14:05:38] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:05:38] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.8748109340667725\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 1\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1665\n", + "│ │ ├── Epoch N-1 = 0.2225 (\u001B[32m↘ -0.056\u001B[0m)\n", + "│ │ └── Best until now = 0.2225 (\u001B[32m↘ -0.056\u001B[0m)\n", + "│ ├── Target_iou = 0.8598\n", + "│ │ ├── Epoch N-1 = 0.8094 (\u001B[32m↗ 0.0503\u001B[0m)\n", + "│ │ └── Best until now = 0.8094 (\u001B[32m↗ 0.0503\u001B[0m)\n", + "│ ├── Background_iou = 0.815\n", + "│ │ ├── Epoch N-1 = 0.7486 (\u001B[32m↗ 0.0663\u001B[0m)\n", + "│ │ └── Best until now = 0.7486 (\u001B[32m↗ 0.0663\u001B[0m)\n", + "│ └── Mean_iou = 0.8374\n", + "│ ├── Epoch N-1 = 0.779 (\u001B[32m↗ 0.0583\u001B[0m)\n", + "│ └── Best until now = 0.779 (\u001B[32m↗ 0.0583\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1594\n", + " │ ├── Epoch N-1 = 0.1821 (\u001B[32m↘ -0.0227\u001B[0m)\n", + " │ └── Best until now = 0.1821 (\u001B[32m↘ -0.0227\u001B[0m)\n", + " ├── Target_iou = 0.8748\n", + " │ ├── Epoch N-1 = 0.8575 (\u001B[32m↗ 0.0173\u001B[0m)\n", + " │ └── Best until now = 0.8575 (\u001B[32m↗ 0.0173\u001B[0m)\n", + " ├── Background_iou = 0.7766\n", + " │ ├── Epoch N-1 = 0.7387 (\u001B[32m↗ 0.0379\u001B[0m)\n", + " │ └── Best until now = 0.7387 (\u001B[32m↗ 0.0379\u001B[0m)\n", + " └── Mean_iou = 0.8257\n", + " ├── Epoch N-1 = 0.7981 (\u001B[32m↗ 0.0276\u001B[0m)\n", + " └── Best until now = 0.7981 (\u001B[32m↗ 0.0276\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 2: 100%|██████████| 309/309 [01:40<00:00, 3.08it/s, BCEDiceLoss=0.145, background_IOU=0.844, gpu_mem=1.14, mean_IOU=0.86, target_IOU=0.877]\n", + "Validating epoch 2: 100%|██████████| 65/65 [00:19<00:00, 3.39it/s]\n", + "[2023-11-12 14:07:39] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:07:39] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.8893157839775085\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 2\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1449\n", + "│ │ ├── Epoch N-1 = 0.1665 (\u001B[32m↘ -0.0216\u001B[0m)\n", + "│ │ └── Best until now = 0.1665 (\u001B[32m↘ -0.0216\u001B[0m)\n", + "│ ├── Target_iou = 0.8768\n", + "│ │ ├── Epoch N-1 = 0.8598 (\u001B[32m↗ 0.0171\u001B[0m)\n", + "│ │ └── Best until now = 0.8598 (\u001B[32m↗ 0.0171\u001B[0m)\n", + "│ ├── Background_iou = 0.8437\n", + "│ │ ├── Epoch N-1 = 0.815 (\u001B[32m↗ 0.0287\u001B[0m)\n", + "│ │ └── Best until now = 0.815 (\u001B[32m↗ 0.0287\u001B[0m)\n", + "│ └── Mean_iou = 0.8603\n", + "│ ├── Epoch N-1 = 0.8374 (\u001B[32m↗ 0.0229\u001B[0m)\n", + "│ └── Best until now = 0.8374 (\u001B[32m↗ 0.0229\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1405\n", + " │ ├── Epoch N-1 = 0.1594 (\u001B[32m↘ -0.0189\u001B[0m)\n", + " │ └── Best until now = 0.1594 (\u001B[32m↘ -0.0189\u001B[0m)\n", + " ├── Target_iou = 0.8893\n", + " │ ├── Epoch N-1 = 0.8748 (\u001B[32m↗ 0.0145\u001B[0m)\n", + " │ └── Best until now = 0.8748 (\u001B[32m↗ 0.0145\u001B[0m)\n", + " ├── Background_iou = 0.8025\n", + " │ ├── Epoch N-1 = 0.7766 (\u001B[32m↗ 0.0259\u001B[0m)\n", + " │ └── Best until now = 0.7766 (\u001B[32m↗ 0.0259\u001B[0m)\n", + " └── Mean_iou = 0.8459\n", + " ├── Epoch N-1 = 0.8257 (\u001B[32m↗ 0.0202\u001B[0m)\n", + " └── Best until now = 0.8257 (\u001B[32m↗ 0.0202\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 3: 100%|██████████| 309/309 [01:50<00:00, 2.79it/s, BCEDiceLoss=0.13, background_IOU=0.858, gpu_mem=1.14, mean_IOU=0.873, target_IOU=0.887]\n", + "Validating epoch 3: 100%|██████████| 65/65 [00:16<00:00, 4.00it/s]\n", + "[2023-11-12 14:09:48] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:09:48] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.8972753882408142\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 3\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1296\n", + "│ │ ├── Epoch N-1 = 0.1449 (\u001B[32m↘ -0.0153\u001B[0m)\n", + "│ │ └── Best until now = 0.1449 (\u001B[32m↘ -0.0153\u001B[0m)\n", + "│ ├── Target_iou = 0.8869\n", + "│ │ ├── Epoch N-1 = 0.8768 (\u001B[32m↗ 0.0101\u001B[0m)\n", + "│ │ └── Best until now = 0.8768 (\u001B[32m↗ 0.0101\u001B[0m)\n", + "│ ├── Background_iou = 0.8581\n", + "│ │ ├── Epoch N-1 = 0.8437 (\u001B[32m↗ 0.0144\u001B[0m)\n", + "│ │ └── Best until now = 0.8437 (\u001B[32m↗ 0.0144\u001B[0m)\n", + "│ └── Mean_iou = 0.8725\n", + "│ ├── Epoch N-1 = 0.8603 (\u001B[32m↗ 0.0123\u001B[0m)\n", + "│ └── Best until now = 0.8603 (\u001B[32m↗ 0.0123\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1304\n", + " │ ├── Epoch N-1 = 0.1405 (\u001B[32m↘ -0.0101\u001B[0m)\n", + " │ └── Best until now = 0.1405 (\u001B[32m↘ -0.0101\u001B[0m)\n", + " ├── Target_iou = 0.8973\n", + " │ ├── Epoch N-1 = 0.8893 (\u001B[32m↗ 0.008\u001B[0m)\n", + " │ └── Best until now = 0.8893 (\u001B[32m↗ 0.008\u001B[0m)\n", + " ├── Background_iou = 0.8195\n", + " │ ├── Epoch N-1 = 0.8025 (\u001B[32m↗ 0.017\u001B[0m)\n", + " │ └── Best until now = 0.8025 (\u001B[32m↗ 0.017\u001B[0m)\n", + " └── Mean_iou = 0.8584\n", + " ├── Epoch N-1 = 0.8459 (\u001B[32m↗ 0.0125\u001B[0m)\n", + " └── Best until now = 0.8459 (\u001B[32m↗ 0.0125\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 4: 100%|██████████| 309/309 [01:46<00:00, 2.91it/s, BCEDiceLoss=0.12, background_IOU=0.867, gpu_mem=1.14, mean_IOU=0.882, target_IOU=0.897]\n", + "Validating epoch 4: 100%|██████████| 65/65 [00:14<00:00, 4.37it/s]\n", + "[2023-11-12 14:11:52] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:11:52] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9037814736366272\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 4\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1202\n", + "│ │ ├── Epoch N-1 = 0.1296 (\u001B[32m↘ -0.0094\u001B[0m)\n", + "│ │ └── Best until now = 0.1296 (\u001B[32m↘ -0.0094\u001B[0m)\n", + "│ ├── Target_iou = 0.8972\n", + "│ │ ├── Epoch N-1 = 0.8869 (\u001B[32m↗ 0.0103\u001B[0m)\n", + "│ │ └── Best until now = 0.8869 (\u001B[32m↗ 0.0103\u001B[0m)\n", + "│ ├── Background_iou = 0.8672\n", + "│ │ ├── Epoch N-1 = 0.8581 (\u001B[32m↗ 0.0091\u001B[0m)\n", + "│ │ └── Best until now = 0.8581 (\u001B[32m↗ 0.0091\u001B[0m)\n", + "│ └── Mean_iou = 0.8822\n", + "│ ├── Epoch N-1 = 0.8725 (\u001B[32m↗ 0.0097\u001B[0m)\n", + "│ └── Best until now = 0.8725 (\u001B[32m↗ 0.0097\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1215\n", + " │ ├── Epoch N-1 = 0.1304 (\u001B[32m↘ -0.0089\u001B[0m)\n", + " │ └── Best until now = 0.1304 (\u001B[32m↘ -0.0089\u001B[0m)\n", + " ├── Target_iou = 0.9038\n", + " │ ├── Epoch N-1 = 0.8973 (\u001B[32m↗ 0.0065\u001B[0m)\n", + " │ └── Best until now = 0.8973 (\u001B[32m↗ 0.0065\u001B[0m)\n", + " ├── Background_iou = 0.8312\n", + " │ ├── Epoch N-1 = 0.8195 (\u001B[32m↗ 0.0117\u001B[0m)\n", + " │ └── Best until now = 0.8195 (\u001B[32m↗ 0.0117\u001B[0m)\n", + " └── Mean_iou = 0.8675\n", + " ├── Epoch N-1 = 0.8584 (\u001B[32m↗ 0.0091\u001B[0m)\n", + " └── Best until now = 0.8584 (\u001B[32m↗ 0.0091\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 5: 100%|██████████| 309/309 [01:44<00:00, 2.97it/s, BCEDiceLoss=0.116, background_IOU=0.871, gpu_mem=1.14, mean_IOU=0.885, target_IOU=0.9]\n", + "Validating epoch 5: 100%|██████████| 65/65 [00:16<00:00, 3.94it/s]\n", + "[2023-11-12 14:13:57] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:13:57] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9058071374893188\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 5\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1162\n", + "│ │ ├── Epoch N-1 = 0.1202 (\u001B[32m↘ -0.004\u001B[0m)\n", + "│ │ └── Best until now = 0.1202 (\u001B[32m↘ -0.004\u001B[0m)\n", + "│ ├── Target_iou = 0.8997\n", + "│ │ ├── Epoch N-1 = 0.8972 (\u001B[32m↗ 0.0024\u001B[0m)\n", + "│ │ └── Best until now = 0.8972 (\u001B[32m↗ 0.0024\u001B[0m)\n", + "│ ├── Background_iou = 0.8707\n", + "│ │ ├── Epoch N-1 = 0.8672 (\u001B[32m↗ 0.0035\u001B[0m)\n", + "│ │ └── Best until now = 0.8672 (\u001B[32m↗ 0.0035\u001B[0m)\n", + "│ └── Mean_iou = 0.8852\n", + "│ ├── Epoch N-1 = 0.8822 (\u001B[32m↗ 0.003\u001B[0m)\n", + "│ └── Best until now = 0.8822 (\u001B[32m↗ 0.003\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1191\n", + " │ ├── Epoch N-1 = 0.1215 (\u001B[32m↘ -0.0025\u001B[0m)\n", + " │ └── Best until now = 0.1215 (\u001B[32m↘ -0.0025\u001B[0m)\n", + " ├── Target_iou = 0.9058\n", + " │ ├── Epoch N-1 = 0.9038 (\u001B[32m↗ 0.002\u001B[0m)\n", + " │ └── Best until now = 0.9038 (\u001B[32m↗ 0.002\u001B[0m)\n", + " ├── Background_iou = 0.8351\n", + " │ ├── Epoch N-1 = 0.8312 (\u001B[32m↗ 0.0039\u001B[0m)\n", + " │ └── Best until now = 0.8312 (\u001B[32m↗ 0.0039\u001B[0m)\n", + " └── Mean_iou = 0.8705\n", + " ├── Epoch N-1 = 0.8675 (\u001B[32m↗ 0.003\u001B[0m)\n", + " └── Best until now = 0.8675 (\u001B[32m↗ 0.003\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 6: 100%|██████████| 309/309 [01:44<00:00, 2.96it/s, BCEDiceLoss=0.117, background_IOU=0.871, gpu_mem=1.14, mean_IOU=0.885, target_IOU=0.899]\n", + "Validating epoch 6: 100%|██████████| 65/65 [00:14<00:00, 4.57it/s]\n", + "[2023-11-12 14:15:58] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:15:58] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9092355966567993\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 6\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1167\n", + "│ │ ├── Epoch N-1 = 0.1162 (\u001B[31m↗ 0.0005\u001B[0m)\n", + "│ │ └── Best until now = 0.1162 (\u001B[31m↗ 0.0005\u001B[0m)\n", + "│ ├── Target_iou = 0.8993\n", + "│ │ ├── Epoch N-1 = 0.8997 (\u001B[31m↘ -0.0003\u001B[0m)\n", + "│ │ └── Best until now = 0.8997 (\u001B[31m↘ -0.0003\u001B[0m)\n", + "│ ├── Background_iou = 0.8713\n", + "│ │ ├── Epoch N-1 = 0.8707 (\u001B[32m↗ 0.0005\u001B[0m)\n", + "│ │ └── Best until now = 0.8707 (\u001B[32m↗ 0.0005\u001B[0m)\n", + "│ └── Mean_iou = 0.8853\n", + "│ ├── Epoch N-1 = 0.8852 (\u001B[32m↗ 0.0001\u001B[0m)\n", + "│ └── Best until now = 0.8852 (\u001B[32m↗ 0.0001\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.115\n", + " │ ├── Epoch N-1 = 0.1191 (\u001B[32m↘ -0.004\u001B[0m)\n", + " │ └── Best until now = 0.1191 (\u001B[32m↘ -0.004\u001B[0m)\n", + " ├── Target_iou = 0.9092\n", + " │ ├── Epoch N-1 = 0.9058 (\u001B[32m↗ 0.0034\u001B[0m)\n", + " │ └── Best until now = 0.9058 (\u001B[32m↗ 0.0034\u001B[0m)\n", + " ├── Background_iou = 0.8415\n", + " │ ├── Epoch N-1 = 0.8351 (\u001B[32m↗ 0.0064\u001B[0m)\n", + " │ └── Best until now = 0.8351 (\u001B[32m↗ 0.0064\u001B[0m)\n", + " └── Mean_iou = 0.8754\n", + " ├── Epoch N-1 = 0.8705 (\u001B[32m↗ 0.0049\u001B[0m)\n", + " └── Best until now = 0.8705 (\u001B[32m↗ 0.0049\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 7: 100%|██████████| 309/309 [01:42<00:00, 3.02it/s, BCEDiceLoss=0.108, background_IOU=0.879, gpu_mem=1.14, mean_IOU=0.892, target_IOU=0.906]\n", + "Validating epoch 7: 100%|██████████| 65/65 [00:19<00:00, 3.29it/s]\n", + "[2023-11-12 14:18:04] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:18:04] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.911444902420044\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 7\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1079\n", + "│ │ ├── Epoch N-1 = 0.1167 (\u001B[32m↘ -0.0088\u001B[0m)\n", + "│ │ └── Best until now = 0.1162 (\u001B[32m↘ -0.0083\u001B[0m)\n", + "│ ├── Target_iou = 0.906\n", + "│ │ ├── Epoch N-1 = 0.8993 (\u001B[32m↗ 0.0067\u001B[0m)\n", + "│ │ └── Best until now = 0.8997 (\u001B[32m↗ 0.0063\u001B[0m)\n", + "│ ├── Background_iou = 0.8786\n", + "│ │ ├── Epoch N-1 = 0.8713 (\u001B[32m↗ 0.0073\u001B[0m)\n", + "│ │ └── Best until now = 0.8713 (\u001B[32m↗ 0.0073\u001B[0m)\n", + "│ └── Mean_iou = 0.8923\n", + "│ ├── Epoch N-1 = 0.8853 (\u001B[32m↗ 0.007\u001B[0m)\n", + "│ └── Best until now = 0.8853 (\u001B[32m↗ 0.007\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1122\n", + " │ ├── Epoch N-1 = 0.115 (\u001B[32m↘ -0.0028\u001B[0m)\n", + " │ └── Best until now = 0.115 (\u001B[32m↘ -0.0028\u001B[0m)\n", + " ├── Target_iou = 0.9114\n", + " │ ├── Epoch N-1 = 0.9092 (\u001B[32m↗ 0.0022\u001B[0m)\n", + " │ └── Best until now = 0.9092 (\u001B[32m↗ 0.0022\u001B[0m)\n", + " ├── Background_iou = 0.8456\n", + " │ ├── Epoch N-1 = 0.8415 (\u001B[32m↗ 0.0041\u001B[0m)\n", + " │ └── Best until now = 0.8415 (\u001B[32m↗ 0.0041\u001B[0m)\n", + " └── Mean_iou = 0.8785\n", + " ├── Epoch N-1 = 0.8754 (\u001B[32m↗ 0.0032\u001B[0m)\n", + " └── Best until now = 0.8754 (\u001B[32m↗ 0.0032\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 8: 100%|██████████| 309/309 [01:44<00:00, 2.96it/s, BCEDiceLoss=0.0988, background_IOU=0.889, gpu_mem=1.14, mean_IOU=0.902, target_IOU=0.915]\n", + "Validating epoch 8: 100%|██████████| 65/65 [00:15<00:00, 4.31it/s]\n", + "[2023-11-12 14:20:06] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:20:06] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9122714996337891\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 8\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0988\n", + "│ │ ├── Epoch N-1 = 0.1079 (\u001B[32m↘ -0.0091\u001B[0m)\n", + "│ │ └── Best until now = 0.1079 (\u001B[32m↘ -0.0091\u001B[0m)\n", + "│ ├── Target_iou = 0.9146\n", + "│ │ ├── Epoch N-1 = 0.906 (\u001B[32m↗ 0.0086\u001B[0m)\n", + "│ │ └── Best until now = 0.906 (\u001B[32m↗ 0.0086\u001B[0m)\n", + "│ ├── Background_iou = 0.8893\n", + "│ │ ├── Epoch N-1 = 0.8786 (\u001B[32m↗ 0.0107\u001B[0m)\n", + "│ │ └── Best until now = 0.8786 (\u001B[32m↗ 0.0107\u001B[0m)\n", + "│ └── Mean_iou = 0.9019\n", + "│ ├── Epoch N-1 = 0.8923 (\u001B[32m↗ 0.0096\u001B[0m)\n", + "│ └── Best until now = 0.8923 (\u001B[32m↗ 0.0096\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1112\n", + " │ ├── Epoch N-1 = 0.1122 (\u001B[32m↘ -0.001\u001B[0m)\n", + " │ └── Best until now = 0.1122 (\u001B[32m↘ -0.001\u001B[0m)\n", + " ├── Target_iou = 0.9123\n", + " │ ├── Epoch N-1 = 0.9114 (\u001B[32m↗ 0.0008\u001B[0m)\n", + " │ └── Best until now = 0.9114 (\u001B[32m↗ 0.0008\u001B[0m)\n", + " ├── Background_iou = 0.8472\n", + " │ ├── Epoch N-1 = 0.8456 (\u001B[32m↗ 0.0016\u001B[0m)\n", + " │ └── Best until now = 0.8456 (\u001B[32m↗ 0.0016\u001B[0m)\n", + " └── Mean_iou = 0.8797\n", + " ├── Epoch N-1 = 0.8785 (\u001B[32m↗ 0.0012\u001B[0m)\n", + " └── Best until now = 0.8785 (\u001B[32m↗ 0.0012\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stderr", + "output_type": "stream", + "text": [ + "Train epoch 9: 100%|██████████| 309/309 [01:41<00:00, 3.05it/s, BCEDiceLoss=0.0923, background_IOU=0.897, gpu_mem=1.14, mean_IOU=0.908, target_IOU=0.919]\n", + "Validating epoch 9: 100%|██████████| 65/65 [00:15<00:00, 4.11it/s]\n", + "[2023-11-12 14:22:09] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:22:09] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9127917289733887\n" + ] + }, + { + "metadata": { + "tags": null + }, + "name": "stdout", + "output_type": "stream", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 9\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0923\n", + "│ │ ├── Epoch N-1 = 0.0988 (\u001B[32m↘ -0.0065\u001B[0m)\n", + "│ │ └── Best until now = 0.0988 (\u001B[32m↘ -0.0065\u001B[0m)\n", + "│ ├── Target_iou = 0.9194\n", + "│ │ ├── Epoch N-1 = 0.9146 (\u001B[32m↗ 0.0048\u001B[0m)\n", + "│ │ └── Best until now = 0.9146 (\u001B[32m↗ 0.0048\u001B[0m)\n", + "│ ├── Background_iou = 0.8968\n", + "│ │ ├── Epoch N-1 = 0.8893 (\u001B[32m↗ 0.0075\u001B[0m)\n", + "│ │ └── Best until now = 0.8893 (\u001B[32m↗ 0.0075\u001B[0m)\n", + "│ └── Mean_iou = 0.9081\n", + "│ ├── Epoch N-1 = 0.9019 (\u001B[32m↗ 0.0062\u001B[0m)\n", + "│ └── Best until now = 0.9019 (\u001B[32m↗ 0.0062\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1106\n", + " │ ├── Epoch N-1 = 0.1112 (\u001B[32m↘ -0.0006\u001B[0m)\n", + " │ └── Best until now = 0.1112 (\u001B[32m↘ -0.0006\u001B[0m)\n", + " ├── Target_iou = 0.9128\n", + " │ ├── Epoch N-1 = 0.9123 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " │ └── Best until now = 0.9123 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " ├── Background_iou = 0.8482\n", + " │ ├── Epoch N-1 = 0.8472 (\u001B[32m↗ 0.001\u001B[0m)\n", + " │ └── Best until now = 0.8472 (\u001B[32m↗ 0.001\u001B[0m)\n", + " └── Mean_iou = 0.8805\n", + " ├── Epoch N-1 = 0.8797 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " └── Best until now = 0.8797 (\u001B[32m↗ 0.0007\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 10: 100%|██████████| 309/309 [01:43<00:00, 3.00it/s, BCEDiceLoss=0.0851, background_IOU=0.904, gpu_mem=1.14, mean_IOU=0.914, target_IOU=0.925]\n", + "Validating epoch 10: 100%|██████████| 65/65 [00:15<00:00, 4.22it/s]\n", + "[2023-11-12 14:24:11] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:24:11] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9131874442100525\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 10\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0851\n", + "│ │ ├── Epoch N-1 = 0.0923 (\u001B[32m↘ -0.0072\u001B[0m)\n", + "│ │ └── Best until now = 0.0923 (\u001B[32m↘ -0.0072\u001B[0m)\n", + "│ ├── Target_iou = 0.9251\n", + "│ │ ├── Epoch N-1 = 0.9194 (\u001B[32m↗ 0.0057\u001B[0m)\n", + "│ │ └── Best until now = 0.9194 (\u001B[32m↗ 0.0057\u001B[0m)\n", + "│ ├── Background_iou = 0.9037\n", + "│ │ ├── Epoch N-1 = 0.8968 (\u001B[32m↗ 0.0068\u001B[0m)\n", + "│ │ └── Best until now = 0.8968 (\u001B[32m↗ 0.0068\u001B[0m)\n", + "│ └── Mean_iou = 0.9144\n", + "│ ├── Epoch N-1 = 0.9081 (\u001B[32m↗ 0.0063\u001B[0m)\n", + "│ └── Best until now = 0.9081 (\u001B[32m↗ 0.0063\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1101\n", + " │ ├── Epoch N-1 = 0.1106 (\u001B[32m↘ -0.0005\u001B[0m)\n", + " │ └── Best until now = 0.1106 (\u001B[32m↘ -0.0005\u001B[0m)\n", + " ├── Target_iou = 0.9132\n", + " │ ├── Epoch N-1 = 0.9128 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " │ └── Best until now = 0.9128 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " ├── Background_iou = 0.8489\n", + " │ ├── Epoch N-1 = 0.8482 (\u001B[32m↗ 0.0008\u001B[0m)\n", + " │ └── Best until now = 0.8482 (\u001B[32m↗ 0.0008\u001B[0m)\n", + " └── Mean_iou = 0.881\n", + " ├── Epoch N-1 = 0.8805 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " └── Best until now = 0.8805 (\u001B[32m↗ 0.0006\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 11: 100%|██████████| 309/309 [01:40<00:00, 3.06it/s, BCEDiceLoss=0.0809, background_IOU=0.907, gpu_mem=1.14, mean_IOU=0.918, target_IOU=0.93]\n", + "Validating epoch 11: 100%|██████████| 65/65 [00:15<00:00, 4.15it/s]\n", + "[2023-11-12 14:26:10] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:26:10] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9135512709617615\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 11\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0809\n", + "│ │ ├── Epoch N-1 = 0.0851 (\u001B[32m↘ -0.0042\u001B[0m)\n", + "│ │ └── Best until now = 0.0851 (\u001B[32m↘ -0.0042\u001B[0m)\n", + "│ ├── Target_iou = 0.9295\n", + "│ │ ├── Epoch N-1 = 0.9251 (\u001B[32m↗ 0.0044\u001B[0m)\n", + "│ │ └── Best until now = 0.9251 (\u001B[32m↗ 0.0044\u001B[0m)\n", + "│ ├── Background_iou = 0.9068\n", + "│ │ ├── Epoch N-1 = 0.9037 (\u001B[32m↗ 0.0032\u001B[0m)\n", + "│ │ └── Best until now = 0.9037 (\u001B[32m↗ 0.0032\u001B[0m)\n", + "│ └── Mean_iou = 0.9182\n", + "│ ├── Epoch N-1 = 0.9144 (\u001B[32m↗ 0.0038\u001B[0m)\n", + "│ └── Best until now = 0.9144 (\u001B[32m↗ 0.0038\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1095\n", + " │ ├── Epoch N-1 = 0.1101 (\u001B[32m↘ -0.0005\u001B[0m)\n", + " │ └── Best until now = 0.1101 (\u001B[32m↘ -0.0005\u001B[0m)\n", + " ├── Target_iou = 0.9136\n", + " │ ├── Epoch N-1 = 0.9132 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " │ └── Best until now = 0.9132 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " ├── Background_iou = 0.8496\n", + " │ ├── Epoch N-1 = 0.8489 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " │ └── Best until now = 0.8489 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " └── Mean_iou = 0.8816\n", + " ├── Epoch N-1 = 0.881 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " └── Best until now = 0.881 (\u001B[32m↗ 0.0005\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 12: 100%|██████████| 309/309 [01:39<00:00, 3.10it/s, BCEDiceLoss=0.0766, background_IOU=0.913, gpu_mem=1.14, mean_IOU=0.923, target_IOU=0.933]\n", + "Validating epoch 12: 100%|██████████| 65/65 [00:15<00:00, 4.22it/s]\n", + "[2023-11-12 14:28:08] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:28:08] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9138703346252441\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 12\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0766\n", + "│ │ ├── Epoch N-1 = 0.0809 (\u001B[32m↘ -0.0042\u001B[0m)\n", + "│ │ └── Best until now = 0.0809 (\u001B[32m↘ -0.0042\u001B[0m)\n", + "│ ├── Target_iou = 0.9327\n", + "│ │ ├── Epoch N-1 = 0.9295 (\u001B[32m↗ 0.0032\u001B[0m)\n", + "│ │ └── Best until now = 0.9295 (\u001B[32m↗ 0.0032\u001B[0m)\n", + "│ ├── Background_iou = 0.9127\n", + "│ │ ├── Epoch N-1 = 0.9068 (\u001B[32m↗ 0.0059\u001B[0m)\n", + "│ │ └── Best until now = 0.9068 (\u001B[32m↗ 0.0059\u001B[0m)\n", + "│ └── Mean_iou = 0.9227\n", + "│ ├── Epoch N-1 = 0.9182 (\u001B[32m↗ 0.0045\u001B[0m)\n", + "│ └── Best until now = 0.9182 (\u001B[32m↗ 0.0045\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1091\n", + " │ ├── Epoch N-1 = 0.1095 (\u001B[32m↘ -0.0005\u001B[0m)\n", + " │ └── Best until now = 0.1095 (\u001B[32m↘ -0.0005\u001B[0m)\n", + " ├── Target_iou = 0.9139\n", + " │ ├── Epoch N-1 = 0.9136 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " │ └── Best until now = 0.9136 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " ├── Background_iou = 0.8503\n", + " │ ├── Epoch N-1 = 0.8496 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.8496 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " └── Mean_iou = 0.8821\n", + " ├── Epoch N-1 = 0.8816 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " └── Best until now = 0.8816 (\u001B[32m↗ 0.0005\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 13: 100%|██████████| 309/309 [01:42<00:00, 3.02it/s, BCEDiceLoss=0.0756, background_IOU=0.915, gpu_mem=1.14, mean_IOU=0.924, target_IOU=0.934]\n", + "Validating epoch 13: 100%|██████████| 65/65 [00:15<00:00, 4.18it/s]\n", + "[2023-11-12 14:30:11] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:30:11] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9141221642494202\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 13\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0756\n", + "│ │ ├── Epoch N-1 = 0.0766 (\u001B[32m↘ -0.0011\u001B[0m)\n", + "│ │ └── Best until now = 0.0766 (\u001B[32m↘ -0.0011\u001B[0m)\n", + "│ ├── Target_iou = 0.9336\n", + "│ │ ├── Epoch N-1 = 0.9327 (\u001B[32m↗ 0.0009\u001B[0m)\n", + "│ │ └── Best until now = 0.9327 (\u001B[32m↗ 0.0009\u001B[0m)\n", + "│ ├── Background_iou = 0.9148\n", + "│ │ ├── Epoch N-1 = 0.9127 (\u001B[32m↗ 0.0021\u001B[0m)\n", + "│ │ └── Best until now = 0.9127 (\u001B[32m↗ 0.0021\u001B[0m)\n", + "│ └── Mean_iou = 0.9242\n", + "│ ├── Epoch N-1 = 0.9227 (\u001B[32m↗ 0.0015\u001B[0m)\n", + "│ └── Best until now = 0.9227 (\u001B[32m↗ 0.0015\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1086\n", + " │ ├── Epoch N-1 = 0.1091 (\u001B[32m↘ -0.0004\u001B[0m)\n", + " │ └── Best until now = 0.1091 (\u001B[32m↘ -0.0004\u001B[0m)\n", + " ├── Target_iou = 0.9141\n", + " │ ├── Epoch N-1 = 0.9139 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " │ └── Best until now = 0.9139 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " ├── Background_iou = 0.8508\n", + " │ ├── Epoch N-1 = 0.8503 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " │ └── Best until now = 0.8503 (\u001B[32m↗ 0.0005\u001B[0m)\n", + " └── Mean_iou = 0.8825\n", + " ├── Epoch N-1 = 0.8821 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " └── Best until now = 0.8821 (\u001B[32m↗ 0.0004\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 14: 100%|██████████| 309/309 [01:39<00:00, 3.12it/s, BCEDiceLoss=0.0731, background_IOU=0.916, gpu_mem=1.14, mean_IOU=0.925, target_IOU=0.935]\n", + "Validating epoch 14: 100%|██████████| 65/65 [00:14<00:00, 4.53it/s]\n", + "[2023-11-12 14:32:07] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", + "[2023-11-12 14:32:07] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9143829941749573\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 14\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0731\n", + "│ │ ├── Epoch N-1 = 0.0756 (\u001B[32m↘ -0.0025\u001B[0m)\n", + "│ │ └── Best until now = 0.0756 (\u001B[32m↘ -0.0025\u001B[0m)\n", + "│ ├── Target_iou = 0.9352\n", + "│ │ ├── Epoch N-1 = 0.9336 (\u001B[32m↗ 0.0016\u001B[0m)\n", + "│ │ └── Best until now = 0.9336 (\u001B[32m↗ 0.0016\u001B[0m)\n", + "│ ├── Background_iou = 0.9158\n", + "│ │ ├── Epoch N-1 = 0.9148 (\u001B[32m↗ 0.001\u001B[0m)\n", + "│ │ └── Best until now = 0.9148 (\u001B[32m↗ 0.001\u001B[0m)\n", + "│ └── Mean_iou = 0.9255\n", + "│ ├── Epoch N-1 = 0.9242 (\u001B[32m↗ 0.0013\u001B[0m)\n", + "│ └── Best until now = 0.9242 (\u001B[32m↗ 0.0013\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1082\n", + " │ ├── Epoch N-1 = 0.1086 (\u001B[32m↘ -0.0004\u001B[0m)\n", + " │ └── Best until now = 0.1086 (\u001B[32m↘ -0.0004\u001B[0m)\n", + " ├── Target_iou = 0.9144\n", + " │ ├── Epoch N-1 = 0.9141 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " │ └── Best until now = 0.9141 (\u001B[32m↗ 0.0003\u001B[0m)\n", + " ├── Background_iou = 0.8514\n", + " │ ├── Epoch N-1 = 0.8508 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.8508 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " └── Mean_iou = 0.8829\n", + " ├── Epoch N-1 = 0.8825 (\u001B[32m↗ 0.0004\u001B[0m)\n", + " └── Best until now = 0.8825 (\u001B[32m↗ 0.0004\u001B[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-12 14:32:09] INFO - sg_trainer.py - RUNNING ADDITIONAL TEST ON THE AVERAGED MODEL...\n", + "Validating epoch 15: 98%|█████████▊| 64/65 [00:13<00:00, 3.64it/s]" + ] + } + ], + "source": [ + "trainer.train(model=model, training_params=train_params, train_loader=train_loader, valid_loader=valid_loader)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "id": "X8BJq1crcbjl", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "3baac32f-83fe-4e0a-d3d8-d75ac230caa5" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Best Checkpoint mIoU is: 0.9143829941749573\n" + ] + } + ], + "source": [ + "print(\"Best Checkpoint mIoU is: \"+ str(trainer.best_metric))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3Nybj15cchxd" + }, + "source": [ + "Now you can download your trained weights from this directory" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "id": "_iHsFgPSciQh", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "919b2142-4b98-4724-dcac-c5e6744d80d2" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "/home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664\n" + ] + } + ], + "source": [ + "print(trainer.checkpoints_dir_path)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "yuhYeXLA18q5" + }, + "source": [ + "# 6. Predict\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "VjRA1tu1mvXQ" + }, + "source": [ + "When the training is complete you can use the trained model to get predictions on the validation set, your data or some other image. Let's load some image and\n", + "run a model inference to create a binary segmentation mask." + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "id": "Ads7RyGN2JwQ", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "outputId": "a40cb318-010e-4a52-dba8-2c931fd773fb" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUAAAAHgCAAAAADx5+uYAAAGiklEQVR4nO3d2Y4UOxAAUTf//8/wMIwYumuxK9JLVUY8Ia7EOE+ne+GqRSlmZmZmZmZmZmZmZmZmZvbkXrMP8Nnvz99a8JTfLXa0Dbz/Wuy4Za0Tnel9t9KZVzpMrV8pKx17nZO0+JV1Dr7KOVr9Slnk7Esc4hJfKWWF488/QSnX/UqZPsEKgITvq4lTzAfkfF9NmmQ6YJRfmTTLZMBAvlKmTPNr/I/8UbBf+J9X0cwN7DLu6IGmAXZblsETTQLsedfGjjQFsPNT1dCZJgD2f6YfOdRwwCEvlAOnGv02ZswbjYFvZwYDjppsnODQKzzyfe6owUYCDv6cMGa0cYDjP2YNmW0U4IRPqWOGGwM4ha+UEeMNeRWe5jfgJ48AnOc3oAGAU/26//D+gJP3r/eP7w44/f52PkBvwOl+vY/QGXABv86H6Au4hF/fY3QFXMSv60F6Ai7j1/Moc/+/8Li6CXYEXGgBS7/T9ANcy6/bebJc4dJLMBFgH8FugKvd4FL6nCnTBnYR7AW44gKWHsfKtYEdygYYvoKdABe9wR3KtoHhD20fwDwLmG8Do+sCuPQCBh/ODYT1AFx6AaNLuIGxj28HwFQLmHEDYx/hjIChxQPe4AZHHtENhOUEDFzBcMAb3ODQcm5gYALCBIRFA2Z7Csy6gXGPc1LAuASECQgTEBYMmO5F2A2kCQgTECYgTECYgDABYQLCBIQJCBMQFguY76OwG0jLChh2V7IChiUgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsKyAob909dZAcMSEBYKmPCrcm4gLRIw4wJGAqb08wrT4gBzLqAbSAsDTLqAbiAtCvBuC+hX/lcpCPBuCxiXGwiLAcy7gG4gLQTwjgsYdeYIwDv6hRUAmNov8XNg0OPOAXMvYOINDHrkMWDyBcy8gTGPfWbAkCjgrW9wxOFzb2CAIAS89QKGxABv78cHyH2FAwQR4O0XMCACqF9BgM/wo1NcB3yGH57jMuBT/GhXAR/kx0a5CPggPzjMNcBH+bFxsr+R/goIXvnCycP2r5QCvnhzYQOf6Hd9qHbAR/pdrxnwqX5X52q9+0/1K+Xi82DjBj7Z79pwvo350RXBNsBHL2C5NF8T4NP9rkzYAvh8vws1AKbwax7SFxFYPWCKBWwf0w2EVQMmWcDmQd1AWC1gmgVsHdUNhAn4WdMKVgImusGlbdo6wFx+TXmFt2pYmCrAfAtYP3ENYD6/hrzC21UvTQWgC3iUG7hT7dqcA2ZdwMq53UDYKWDWBayd3A3cr0rwDDDvApa64d3AoyoETwBTL2Cpmd8NPO5U8Bgw+wKWcwI38KwTwUNAF7CUM4UjQP2+OnQ4ANTvuyOJfUD9/nVg4YtIVfuCu4Au4H/tcuwB6vfWHsgOoH4f7ZBsf7lEv822sDY3UL/ttly2APXba0NmYyv1O+od7ANQvpPexN6vsH5nvQm9jv6j7fTa+KV2jf2Vewl3vVcREOffxsAEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiBMQJiAMAFhAsIEhAkIExAmIExAmIAwAWECwgSECQgTECYgTECYgDABYQLCBIQJCBMQJiDsD0JPxM/Z23MkAAAAAElFTkSuQmCC\n" + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-12 14:33:33] INFO - checkpoint_utils.py - Successfully loaded model weights from /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth EMA checkpoint.\n" + ] + } + ], + "source": [ + "from torchvision.transforms import Compose, ToTensor, Resize, Normalize, ToPILImage\n", + "\n", + "# Initiate a model with best checkpoint.\n", + "model = models.get(model_name=Models.PP_LITE_T_SEG75,\n", + " arch_params={\"use_aux_heads\": False},\n", + " num_classes=1,\n", + " checkpoint_path=os.path.join(trainer.checkpoints_dir_path, \"ckpt_best.pth\")).cuda().eval()\n", + "\n", + "pre_proccess = Compose([\n", + " ToTensor(),\n", + " Normalize([.485, .456, .406], [.229, .224, .225])\n", + "])\n", + "\n", + "demo_img_path = \"/home/data/supervisely-persons/images/ache-adult-depression-expression-41253.png\"\n", + "img = Image.open(demo_img_path)\n", + "# Resize the image and display\n", + "img = Resize(size=(480, 320))(img)\n", + "display(img)\n", + "\n", + "# Run pre-proccess - transforms to tensor and apply normalizations.\n", + "img_inp = pre_proccess(img).unsqueeze(0).cuda()\n", + "\n", + "# Run inference\n", + "mask = model(img_inp)\n", + "\n", + "# Run post-proccess - apply sigmoid to output probabilities, then apply hard\n", + "# threshold of 0.5 for binary mask prediction.\n", + "mask = torch.sigmoid(mask).gt(0.5).squeeze()\n", + "mask = ToPILImage()(mask.float())\n", + "display(mask)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-k6ZLKHL1hIM" + }, + "source": [ + "# 7. Convert to ONNX/TensorRT" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "br7n55Szm4Nq" + }, + "source": [ + "SG is a production ready library. All the models implemented in SG can be compiled to ONNX and TensorRT. Deci also offers the [Infery](https://docs.deci.ai/docs/installing-infery) library that allows to do inference on models saved in various frameworks with the same API regardless of a framework.\n", + "\n", + "Let's compile our model to ONNX." + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "id": "q0AGQvEf11PT", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "8916ce5d-c1a4-447f-d7b7-472e6e18fe71" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "ONNX successfully created at: /home/data/model.onnx\n" + ] + } + ], + "source": [ + "from onnxsim import simplify\n", + "import onnx\n", + "\n", + "input_size = [1, 3, 480, 320]\n", + "onnx_path = \"/home/data/model.onnx\"\n", + "\n", + "model.prep_model_for_conversion(input_size=input_size)\n", + "\n", + "torch.onnx.export(model,\n", + " torch.randn(*input_size).cuda(),\n", + " onnx_path,\n", + " opset_version=11)\n", + "\n", + "# onnx simplifier\n", + "model_sim, check = simplify(onnx_path)\n", + "assert check, \"Simplified ONNX model could not be validated\"\n", + "onnx.save_model(model_sim, onnx_path)\n", + "\n", + "print(\"ONNX successfully created at: \", onnx_path)\n" + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} From 4fc55958b1360e893a5a4073c6f4b4589af9b2b6 Mon Sep 17 00:00:00 2001 From: shayaharon Date: Mon, 13 Nov 2023 11:03:18 +0200 Subject: [PATCH 2/5] final touches --- notebooks/quickstart_segmentation.ipynb | 1 - .../segmentation_connect_custom_dataset.ipynb | 1941 ++++++++--------- ...nsfer_learning_semantic_segmentation.ipynb | 1 - 3 files changed, 970 insertions(+), 973 deletions(-) diff --git a/notebooks/quickstart_segmentation.ipynb b/notebooks/quickstart_segmentation.ipynb index 63150b51c5..13798e9e3a 100644 --- a/notebooks/quickstart_segmentation.ipynb +++ b/notebooks/quickstart_segmentation.ipynb @@ -145,7 +145,6 @@ ], "source": [ "! pip install -qq super-gradients==3.4.1\n", - "\n", "! pip install -qq prettyformatter\n" ] }, diff --git a/notebooks/segmentation_connect_custom_dataset.ipynb b/notebooks/segmentation_connect_custom_dataset.ipynb index 2e3b465b5b..1e2d55dad4 100644 --- a/notebooks/segmentation_connect_custom_dataset.ipynb +++ b/notebooks/segmentation_connect_custom_dataset.ipynb @@ -1,980 +1,979 @@ { - "cells": [ - { - "cell_type": "markdown", - "metadata": { - "id": "sh6t_y7KzqBH" - }, - "source": [ - "![SG - Horizontal.png](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAABGcAAAD5CAYAAACK9FhIAAAACXBIWXMAAAsSAAALEgHS3X78AAAgAElEQVR4nOydCZgcV3W2z62ZkUbryPIu20gywWhFEsZJzGLJkBCWEMsYsFk1hjiGgLEclgQIWCIsgQQ8Zov/sFgibGazxB8ghhiPCJD8EViSZUuyvEnYlm3ZljQaLaNZqv7ndp/quX373urqrrrV1d3f+zw13VNVXcut6uq6X33nHBEEAQHQZMwgoqW8yXN4MLGViA7x+H4c5DLUNlTf6+zhgbT2BAAAAAAAAACQEp1oSJBzpGiw4ukDY2+Y1O2dN3my6JGbe/x4EEyaJES1TR8bIxoZCfzubuGNjdHo0WP+U8PDwZZTTu74Lgs2e3K+/2kg23Dp0wfG/rqrU8ydPt07RS5zaKjYLnGWf/SYPzZlstch3x87FgwcH/J3T+gSP502zbuVRRsAAAAAAAAAAHUC5wzIIysHBvzVU6aKF1FAorOzughTL1JoGBkJftbT432ixUSGQhtOnCgu7OoSEzo63KxkdDQIhBD+oYGxO0+e2SHbcIObNQEAAAAAAABA6wJxBuSFOQMD/qenTBGXCSE8V2KCDd8vOEkOnBgOPnHSDO+rTRq+M+fAgbGPTJ7svdGlIGNDtmEQ0Njx4/7Ppk71/rpNXEkAAAAAAAAAkBiIM6DRzHnqqbFvnnJKx/PzcCRkGJQ0hAyd8L8wZbJ3XcM3KB5zjhzxvzR1qvfyvGyQ71Nw8NDYb0+e2fE6iDQAAAAAAAAAEA3EGdAwDhwY+9r06V6vy7ClepEizehYMDRxgnh9nkN1jh7zb5g4QVybxzYMGRjwv9fT4/0VkgkDAAAAAAAAgBmIM6ARLD182P/11KneZC9WOtrGIUN1jh7zfzttqvenORMXmqYNJSeGC0LXy1E1CwAAAAAAAAAqaYJuHWglDh7y3zM6GmyZPr05RAW5jdOmes87MRw8GlFuOlNkGxJR07ShZOIE0e37dId0S+VgcwAAAAAAAAAgV8A5AzLj0X2jPz7zjM5XNIugoDM2Rn5HB72kke6PJ54Y+59TT+34o2ZtQ+lEGhz0N/X0eCsR5gQAAAAAAAAAReCcAZlw3/0ju5pZmJF0dBS+L3ewcyVzdu4afqCZhRkKnUjTvOXHjgUPENGMHGwSAAAAAAAAADQcOGeAc7bddeLRxYsmzmpmUcHAxVk6aO7ceuKJpc+ZeForteHx48H+SZPE6TnYFAAAAAAAAABoKHDOAKf8y5cPbW9BYUaGON2eVQ6a7/1w8NFWE2YkEyeK0wYGfCQIBgAAAAAAALQ9EGeAM6665vH/eMdVMxa1mqhAHOJ07FjQ7zo050Nrn9p12cppLSduEYc4TZ3qLR8c9NfmYHMAAAAAAAAAoGEgrAk44cV/8fC3N37rrCumTW1t/e+pp8d+esrJHa9wseyLXv7w/97+o7Mv6OoSLhafN5YR0dZ22FEAAAAAAAAA0IFzBqTOc16455/efdVJLS/MSGae1PFyIlqR9nKf84I9v7zxH09rF2GG2IUEAAAAAAAAAG0JxBmQKs950Z4Pzpsz8b2vesXUtmhYGZpzYjj4cZrLfM6L9tx+wXMmvWjJcyamudhc090teo4e829omx0GAAAAAAAAAAWENYHUWLz8oevECfHZH37zLPqDZ3a1TcOOjUnnh//RadO865Mua/Hyh24TJ8RLf/WzZ9D06e2lnY6Nkd/RQScT0aEcbA4AAAAAAAAAZAacMyAVFl/80HViRHz2gqWT6Nxz20eYoWJyYJo82ftg0uUsfvFD3xEj4qWXv3o6TZvWfl9NmWT5wIGxz+ZgUwAAAAAAAAAgUyDOgMQsfnFRmKExog+8byZ57ZEmRafz4CH/PfV+ePFLHvq6GBaX05iga94+g0R7tiH19HSscl0BCwAAAAAAAADyBsQZkIjFf/LQdWJUfJZ8olmnd9HcOe3lmgkpumfER+v57OI/feifxbB4sxS3XnzRZJoypa2/lt7goH9dDrYDAAAAAAAAADID4gyom8UvfejqgjAzRiTFmbdf1VMQKdqVzg4xmYiW1rL7i1/60HU0It4j2498QW/rnd7WbSj3fcJE8d4cbAoAAAAAAAAAZAbEGVAXi1/24NU0SjcVhJmgKCy85OLJbd2YUlgYGPD74s6/+GUPXkej9FnB4pYsPb54YftUaLLR1Vm7yAUAAAAAAAAAzQzEGVAzi18hhRlxkxgTVHR8EM171gTqntiYRCmyWpIcqo3LgqlTvRfGWc3iVzwoHTOfLbVhUAxpGh1tXPW04eHydcv208dlgSxPPjDgJ06wDAAAAAAAAADNQieOFKiFxa988M00Im6iUSqJChQIesMV06irKztxRlaAP3LUp9vvOEb9vzxO+/aN0q7dw6Xp886bQLPO7KSX/slketELJ9HUKV4mSXaDIPCIhHR9bLXNs/jPH7xaCjOhsBU6j173mqnU2ZldG0rxxfcD+s1/D9F/3nGM9j02Spt/N1SaLtvvrFmd9IILJ9ErXjaZTju1syCcZEH3JPHKzBoCAAAAAAAAABqMCILGPakHzcXiVz34ZhoWX5fCjBgVVAhpKgyC/vPHZ9Npp2aTLGVkJKBP33CQbvn+YPSMyql9+Wum0fv/5qRMBKQDB8Zunjmz462maYv/4sGraVjcRMNEBdeM0oa/+/UzMhO4RkYD+pf/M0BfWT8Q7wMB0QXP7aYPvH8mnTu3y7lI4/sUeB6dS0R73K4JAAAAAAAAABoPwppALBZf8uCbaVR8PUz+W3R7FF9lrpSshJnHnxilFS9/hG754WEiL7APgs9u1jqkkLPizx4pfN41nV3i5aZVLF4pHTNUdB0F5a4ZKXxkhWyDV122j77yb4eqt2GoFQmizVuG6NVX7KMf/fiIFE+c4nmFNa/IrFEAAAAAAAAAoIFAnAFVWXzpg6+mUfp6mPxX+JwnhYohTTKESLpZXDN4xKfL3rSPBo+NFc/cyEFuaCg0FMWFwUGfLnv9Y87zqEyf5p2hj1t86QNvLggznGOm0IbBuLtnyeKJmbhmwjbc9+TIePt18FDRnkobKiLNRz76dCYCzdMHxv7a7RoAAAAAAAAAIB9AnAGRLL7sASnM/ICU5L8FgvFB5iRxLSz4AdHqv9tPg8f8cVeMOgiqcHqMjx93gUiB5h3X7C8szxUjI4VYwVK1ocWXPcCuI1Hefko7yrw4rpFiSqENj/vmttPHeYY2pHGBZv+Tbl1Ik7q985w3CgAAAAAAAADkAIgzwMri10phRvxADWUSgSgTZiTLlrov//z734/Q5m1DES4PTVRQRRpNXNh851Bhea7o6iqkHi6IM4tf98DVpXCwYDyMSW/DZ5ztPjf3trtPFNvQJG55BnFLaMKNJtB86CNPOxW5Jk8WPe6WDgAAAAAAAAD5AeIMMLL48gdeSiOhMCOUykzliXbl+9NPc5tvRjo+vvrtgUqHTIV4QHYXjSbQfHXdYadhOQcOjF206PIHri5UttLaUKjtGBQ36OST3bahrMz0ha8epKCDioNXPhidM2Rq03KRa/9+d+4ZmRRYdSABAAAAAAAAQKsCcQZUsPj1D7yURsVtZaWeyfRa7KXLkssukZWBbv/1UQq8wDhYBRtdoFFef7HpmNOKQ0dPBH8qRummsgTKJnFLlv1+1oSCeOKSY8d9+t/tx63Jf0vtqoc4GZ00QWncd79/xNlWB0FhbTPctgwAAAAAAAAANB6IM6CMxW+Qjhm6rVBRyGcBJiKkKQse+v1IMdeMJRSnKCpoIo3VQROUcs/s2esutImC4GwaDfP0iAqnTHEernY1zf3X8K4dJ6qLV4LGhZqoeWj89Ve/Oe5MWOoomolQsQkAAAAAAADQ8kCcASUWv+mB59Mo3VZIXBtYHB9aSNMF53fTseNulZqd90UIC4oIUybSUHX3zP33uxNnZkzvKG8v23t2HnU4rkT+698di65uZRCxAps4o7hndu0edrrtjz8xutDd0gEAAAAAAAAgH0CcAQUWveV+Kcz8uqwqU+iUoUpRRmXiBLeVmu6W4owpCbAlL0qFi4YMr0R03wPuxJlpU73K5L+k5Juh8VfXYWFSPNv10LA5mXJFVSb+kE2gIe09ET38iLu8M11dYrazhQMAAAAAAABAToA4A2jRqvtni1Hx64r8KGQIXzKMd+36KAgLtqpMtrwoFDGO+e3vhtxueFSuGeX9WbPcijOTJwl6VCbureI8soaBkVncCuc9dsxdZuXuiWKms4UDAAAAAAAAQE6AONPmLHrr/TPEmNhTqihkCmMK880YePZ5E5w24OARv7xKUJSoQAb3DBmEGWVZTkOybCFNDWDfkyPF5L9CSQSsO48sDhpreBNz7253DqQxn6Y0psUAAAAAAAAAIDsgzrQ7o3RwPPmv6vYwJP81CAuF8B3XKMlqSwIDRTg+SBFoqHycLtSMjbpRS2SS3FmndWluGW3l/P8557h1zhQw5pgxiDRGwUtzz2RLd0PWCgAAAAAAAAAZAnGmjVm06v5AjIry3ChkEWNi5J5xhsnVIYJKN02EUJO1sCBLV591OosuVdpMhh05R3cc6SJNVGiYQVcCAAAAAAAAAJAeGTyyB3lk0VvvD4SvJP8lQ46UPKGLLaVtDMonBtp8ooH7EtWOWW+Tp2yPUWjhCep0oU0GAAAAAAAAAOAEiDNtyKK33R8UQpnGIhLWksEt04AO+uRJXnXnS6CIMPprHlArNTUKoYlaRpEmKFpkbG3YAAcSAAAAAAAAALQDCGtqMxZddX8gRRlRlvxXC2myldBWySjORVaCKuS1MZZ6Dt8bctDkhbwIRFUSKEeVHC/+b8jfAwAAAAAAAAAgFSDOtBGLrr4vqCiXnQN3TDWec97E2oQFGv8/aGwy23EakatHJcwro+fvIV20UTYOThkAAAAAAAAAyASENbUJi95xX0CjxRwzQq/EZMuNEiUiiGLcy727h5034LzZE80iQVl+FCUkJy8hTXkSvArtFKj/jLedOjowjAcAAAAAAAAA4BQ4Z9qARe+UjpmwKpOlZDbZRRphC2ESRIODvnlaisw7d4K5MhMld3ZMm+bmK5BJifFa0MPA4oYzAQAAAAAAAABwDpwzLc6id903nvzXt4Qy6e9N/9tggcaVyCHpmeYVxI7BwaDc1aE6Pepwy2z+3RA954K96W9wB1HQ4160qglT5SVVkDElAAYAAAAAAAAAkAlwzrQwi969m3PMjIcxCRJ2IaaOjrkUOFwKM5LOTkEv/sMplflQkqC7cNIaJhIFM8fyFxpUsX+B2S2j5p1BzhkAAAAAAAAAyASIMy3Kout2F0OZ9OS/1fLMUD7dE5f/SU+5qKALB7rgEEU4r5fyMDGg4KSx7BsnBoGxzTSBBkIMAAAAAAAAADQEiDMtyKK/2R3QSCjMRAg0lI4Qs++xUeeNuPi8ieY8LvWICnrVojSGTqLgJD+/YUFCE2h0YQsAAAAAAAAAQMOAONNiLHrv7srkv2QJaQoT/ZrGxYFn3bNnJJNG/OvLZsZzekSNT1uU8TiU6WRDjpm8CTW2MCYAAAAAAAAAAA0F4kwLsej9u8uT/1KMEKYYBKYcJEqn/rd3nsikES+5eBqvu86dcSLMBBTM8PMvcoQ6HKoyAQAAAAAAAEDugDjTIiz8u3uLwoyS/LesTDZZwpmiKjbFZNv2EzQy4t4mMrnboyv+bHp9H3aR/FdWZWoGYYYikvyqOXwAAAAAAAAAADQEiDMtwMIP3huI0RqS/6alo3DHXlZs6upy37vv6CB631tOoWlTajhtpT7lBekPEzj5b96qMkVRkUA5f5sIAAAAAAAAAO0IxJkmZ+GHVGFGVOSZMb4mxdCx37M3m7wzEyYI+uvXziz+U01cCLcz7VCmCUQkc8x0ZLHHKaG0VS1phQAAAAAAAAAAuAfiTBOz8MP3BiIsl62LMlpIkyglHUmwvya3Bf9/28+PZdaQr395D82bOyF6Jj1xcNqhTM3glNGpN1dPFQ4PGpIhp0SHR087WzgAAAAAAAAA5AQRBM3YywQL19wb0DCRkCWzwwTAoYNmjHPPhKKNjL4JlGlB+KoJO4XpSqWnsM8t06r42vgwVIrXM++8CfTNdWdkEt4k2f/0KL3kqt+P709puzT3UJrIctnT/YKkKcbUdhOV26G2P88n9HHcpt/7xpn07POqiE0JWfSW+8vWKXxt+8u2S8ldpJwzpe3XzxV3l5CL79o8u99pwwAAAAAAAABADoBzpglZuHaXTyNSIFDcMKZYFb3THKcTrSaIjVuqWhDt2j1Mx49nJ/SdPKOTPvXu08wTvaLDJdVhAlEwc6z4HnpmOS6SLQu6DsIMAAAAAAAAoF2AONNkLPyHXT6NCjHuXLEk/nWZb0Z9r4TKfPu7g5k1pkwO/LKLptIVL9OqNxXEpZSHDlkue8wuVjUD1QS3epBt46U+DJMXXH/X/87ua+LWBgAAAAAAAICagDjTRCz8OAszapiRKxFGRa5KzVeidvKV97f+6AiNjWXXnp5H9P6rTqZ5505wk/jX41CmmU2W/LcaaQg0woE7qYOGyaO33PU/cz7qtgEAAAAAAAAAIF9AnGkSFn5yZ0mYqSyTrQs1Wu+7nvAmsnTiI0Kd9j02SrvvH860Qbs6BX3tY7No2lTPjTBzql/+LUFIUxE3oUwfv+s3c27Jw+4BAAAAAAAAQJZAnGkCFn5KEWZ8SzhTSBwHTRznhKkqk8Eto4c2/fNnD2bqnpFIYeaHN55dFGjSrMo0010VoqbFlUPJo0/d9Ss4ZgAAAAAAAADtCcSZnLPw0yzMKJVxROiMMYkycbG5YqJEGdP8VC5qbL5ziAaPZC9qnHFaJ33vhrNp2rQUHDSdSP5romDQ8pwMF9/1yzl/l789BgAAAAAAAIBsgDiTYxb+886xslCmUPOwhSmZXDPVxIW4VZnUvDNGEWd8RV+86RA1okL7mad20idkBaekwswpY62VYyYNCo4ZJwmAP7W9fw6qMgEAAAAAAADaGogzOWXhDTvHaEx4xVAmYcgz49DVEZbTNk5TXoVhHBHd8v1BGjicvXtGJgi+6HmT6fN/e0Z9QkFnUHTMhPsD10wRN/ll5PCp7b+YC8cMAAAAAAAAoO2BOJNDFvTtGKMR4dFYpRgjqiX/TYLuiLHlmbFOHxd13vP+JzPPPUOqQPP+M1GVKQ3cVGWSbX7F9tshzAAAAAAAAAAAQZzJHws+t2NMjIaOGYNLptbkv3GICm2i8mlloU0RrzL3zEN7RhrSvgWB5vzJ9LUPz4qXJNjjHDP4NpTjMPnv9p/PRVUmAAAAAAAAAGDQHc0RC74QCjOi3BkTtypTLSKN5nQpH294T9UFGX2Z77puf8MaVwo05y+YRLd8/OzoMtu2HDPtHtLkMpTpP+CYAQAAAAAAAAAViDM5YcGXFMeMXipbpZ6qTKbBNF/Z/0HlNGFJDGwUaYj2PTZK3/j2YEPCm4gFmnPO6KIffvIcs0DTSeSXqjKlGB7W7LhL/vvF7T+FMAMAAAAAAAAAOhBncsCCm9RQJlEZzhRVQjsNbHlmyCK8VHtV3DOf/uwBevpAg9QZ5oxTO+mHHz+Hpk4RBWGpMHQE48IMGMdlKNOPz30XWhoAAAAAAAAAKulEmzSWBf96z6gY8YrJf3W3jEmMsQkz9Qg2Qvuc+n/4XsSYT61upL8PiN517X767rfObGg7S4Hmjs/PpTd97BHa9cgJCk7ialLZF5XKL4ooN22KR/POnZjKpp59eue3P/ruU/+DiFbU8fFDRLQ15y0HAAAAAAAAAImAONNAFnz5nlEaER3jooylZLaJNJIA6wJLhTATFLfJkDBYuk8EiXJBpkzQGZ+w675h+twXD9E73z6DOhroVOmeIOjbHzmbrvjcXtr12PC4MNPu+WWo0gklhZkvf/RMOnY8mXolpFlJ0OuOHvNfV+tnOzwhhkeC/dOneY1V9gAAAAAAAADAMRBnGsSCr90zSqOiQ0SEMZVIuypTmQsmGHfomAQaivG/sLhmwjcB0VfWDdCFf9xNF5zf3dB27+oU9J13z6arvvIobb7vWEO3JTfo4ptPJA6LgpBWyNWTjLAYd10MjwSNPWEAAAAAAAAAIAOQc6YBLLj5npFyx0wVUUbHNN6WzzZM4qt/KCq3TKy8MpwYWP9cxHJWv+9JevyJ0Ya3vxRo1r39bLrij2c0fFtyQeE48jBGJB7vIBpBgmQAAAAAAAAAyAqIMxmzYP09wzQqOgvuhNA1Q4ZqQXq+mTiuGVsJbH0ekyhjel8KTzKIL2Sp3FSxHUFp2uCgT+/+mydpZDQfcUQfuuw0uuLCNhZoDMl/xZNeQaABAAAAAAAAAJAdEGcyZMG/3X2CRkVXRfLfODlmkmISbEoiSmAWVmyvttLc1nHjAs2u3cP09nfuz8XxkKW2P/Sa0+iK57ehQGMUZjogzAAAAAAAAABAA4A4kxELvnn3EI2ICYXOb1gW2ybM6AJNGrlmbP/HKpttcMco02OFNykCzeY7h+jjnzpAfg4qJUmB5oOvPY0+/ZYzGr8xGRJ4yiBdXI91EJ1om90HAAAAAAAAgFwBcSYDFnz77uM0IiaSKfkvaSFNaSX/pUoxpZB3plpuGZsTRv/flH+myjyqQ+eW7w/SJ/8pPwLNy543jb5w1azGb4xrFJEsTOTsPQXHDAAAAAAAAAA0EogzjlnwnbuP0ojoLokyFBHCZBJmahVpNBFkfJz2v/5eFVeMr0H5Z2IINnEEmtv7j1GQgxQ0niBavmQKfeGvWligCUOZOoLiIMuhH/CIRnKwbQAAAAAAoAzP616hDmgdAFobiDMOWfDduw/TqJhcFGbGXTNCd8rUI8CY/q9WYCfKLWNbvjAIKybhJSq8KWLce/72Sfrlfx3PhYNGhALN1S0q0KhtL6sy7ZehTNVOGgAAAAAAkBWe1z3D87r7PK/7EBHdoQ6e1x14Xvc6z+uegwMCQOsBccYR83+w/UkaFdMKgowfIwGwSaCJEm2qCTKx3DLavFHii+3zpvCmGhIES655z376r1/lSKBZOoW+8I4WE2j05L9PdxA1vqo5AAAAAABg2B2zh4iuJaIeS7usIqKHPK97NdoNgNYC4owD5v9w+8NiWJwyXpVJVIowSUKXojC5YBRhJCAtfEmdL+JzZQJOxbjyITCJPOryPLtAk4cQJynQXLRsCn34jac2fmPSQq/KhFAmAAAAAIDcwMLMHRGijM4Nnte9BkcQgNYB4kzKzN+w/WExKs4mSzUmoQo1OkmEiTjhSfp8RgdNYJ7H9FmTo0Z10OjzxXDQyBCnvOSgee3yHrpiRdzfx5zjFYXCYihTa+wSAAAAAEArIEOZiGhDHbtyPXLRANA6dOJYpsf8H23fI0bE2eOOGUsIE1kqNdVDQeDQBB/BK5DLFsq6+L10zwghysaVRBN1XmvIlLJs0j6rizX6uAqUGYKiQPOpj51Cf/anUwpVlBpJocz2m04tbNd3fjHQ2I1JShCQeKqTaLi5dwMkw/O6lxKRvImTr3P4tZoCuYlf+4loqxx8f2gPDgUAAACQGqtrcMzo9PHvOQCgyYE4kxLz/337g2JEzNaT/1aIL2m5ZgSVCy/qeE2MqRBPqr2SJr6oyyPDPOouqlqLrAZEomaB5m///im67/4Retc7ZuRCoPn7VafS1vuHaNee5rWcoCpT++J53SuJKBzqufFbrr3KZW5jsWad7w9tbfc2BgAAABKSJH/MEvnwBb/HADQ/EGdSYP5P7npQjHhzy5L/kiG/TFTeGRNRYURxRBnb+6hXUpZvcsao/+sijv5ZznFTj0DzlXUDNHjEpw+8b2bDBRrJtz9yNv35+39Pjz7RpFl0UZWprWB7dC/f7M12sO9LeLiWhZo+3x9a1+7tDkA7wdVi5rAbj5RXG7LjeIhf96AjCZoJz+vur3NzV1c71zksKWkc/Ur+bgEAmhiIMwmZ/9O7HqQRMbci+a8tnKkeTA4ZFatbRgttUsSdIODQJlNIUqCELlENoopldCyBRo73ywWaW74/WHj7ob+dmdbhqpuuLkHrP3gWXfqBh2nwaA7KSgFgwfO6pSizxpEoY0KKNDdzUsI1EGkAaE1YjFnBncB6OpPL1X88r5s4bFJ2ejdArAE5Z3mdmzcDBxYAEBeIMwmY/7O77qdRMVf4FkHGVKWJLP/Xih7WJAyOGKrBLWMQcIxqiklgqeqKiSHQBJy0NgiK77lNpUDz4IMjdNMXT6Ouzsa6P844tZM++fbT6V2feayh2wGACSWZYJwbyDAs6RC/2pij5KZZWkXwmc0ijXTr9KKjBUDzo7jwelmITZvlPMikpnv5GtaHvFYAAADaEYgzdTL/P+/aTSPimcbkv2RO/iuSJv8lg4tGdcjY5jN+RnHPmPLSkPZ5sog6ZEgQbCGWQFMqtx2UxK3Ndw7RG9/yOH3z62c0XKC56PzJdMWfTKdbfna4odsBgAon+u2v8iR7I3d85BPqQ/U0IHfUVipPz03rkx24LZ7Xvdb3h1DiE4AmhF0y8vu7KsOtlwLvtRwuuYmdePWGkgAAAIgBh9XN8P2heqqFgZRBKe06mH/7XbtoWDxLCjNCT/5rc8iYKjXVSKk8dUhUmevSOKVctaGcdUCBcfz4YCmJHXc+bQjC7bHNS9oy+f2u+4bpjasepxMnGltnW5bY/rtVp9CsU6FrgnxQRZiRZcbWEtFc3x9aKUOO6hVmJPKzvIxedtRcSUR7LbPLJ+H9LOgAAJoAKcp4XrcMTXwoY2FGR7pp7uBrCMoEg1YnDacp3KqgKnyNX+l53X18fZUdqzsSJqQGKQJxpkbm9991N42IZ1uT/5LmoEmKXg2JIspcmwQaivhfF0OoRuEl7nxpCTS7h+nFL3+EHm9wUt7OTkFf+/BZDd0GAKi6MCOdMrJ6wxoXIQKKUBMl0sgOFgQaAJoAzhvVaFFGJxRpNrCbB4CWgx+abEywXwNwPQAT8j5RhptL0d3zuvfwNf5WdinWm0cJOEXrj7cAACAASURBVATiTA3M37TtbhqmhaEoUwrRsSUANgk0NtHGJpbY5qn4jCKMqNP1aSaxRHfP2ISXqkKKe4FmcNCny17/WMMFmjNP7aS/XIn+JmgcSo4ZkzBzHTtlMsnboIg0aw2TDyVx6wAA3MI37/JacX2Om/oS6QzgnFYAtCJJwoD7cEYAGr+er1FcMVuI6AYW3bMqFAESAHEmJvP/a9udNCIWFhLVhmFMZblmtJCmNPLLqNgWV9X1YpjPNK/+uWrLiRyyEWge3TdKfoMKJ8ny3u+8fCbNOg3hTaBh2CoyXen7Qw25UeMcM8sUF80AJxIFAOQQFju2NMlNuxSib2AXDZ6OgJaCk+ibHnBUYxPyuwGFlSy0wxXTpKBnGYP5v962mUbEslIYUxyHDMWYFodQnODlSEGjLImvvnx9XPi/4A+EooeWlLeUHLjic+qy4iX+La0gal5RbMuakgQr/0iB5nVveoy+dfMZdM45XQWxJGtkeNM/vut0esvf78t+5Y6RIWRvu/qJWCt5W+90ev6Fk5p4b5sPtvdfa9jwKxtdylreYCrhVutQdQWAfMK5ZfIUwhSXSzhcElXhQEshRRYWHk2/7yY2cWccANAiQJypwvzfFISZ5xWrMo2HMYlAlAs0FFOsiUvBKaK4cVTRRWjiiSpehIKIqUw2kVmc4fdVqzcJsosuRnElQqAplc4urrfQpn7EvBaB5lWv2Uef/8xp9KIXTmqIQLNkXjddsGgSbd52PPuVO2TwiE+btwzFWsElr5rapHvZ1Jis/RsbLcyEcBjT0jxsCwCgHO789TsqjZ0VyD8DWhLfH5L5Qfo5VMnmaBvgamYIZwKgxUBYUwTz/2fbb0rCjB7GpGITY+oRaUz5ZkzhSFHTbblp9Lwx+vy15J/Rx3nm5UWGOGnjAy8oX16MECfJNe/ZT//1q+MNCXGSgtCHrzol+xW7RB6HOEPKkXugJkyhQsjFAACIpEWEGQlcM6Blkcl9OY/bMg51upGI1vP7S31/aAaEGQBaEzhnLMz/322/oFG6cLwqU8zkv/UKMuGrKSSJFGEisIQ2qeFLNmcLGabbwol0KuYzrMfjtqrVQaOGV6lhTp7Sxr7dQSPfSoHmO18/k+Y9e0LmDppnzOoqume2N7l7RtQo14453BZghUOG9CTA6xE+BACIooWEmStRmQa0AyxAQoQEoI2Ac8bAvN9u/QWN0MU0JirdMgXDgCX5bxxhxuYsMc1neq8vhyyvUW6aetwzxsGRg8aUKNgzLbvcQXPVO5+g/U9mX8Wpo4Pow29vAfeM2s7VBtBIVhjWjY4KAKAa61pEmMlF+CYAAACQNuhmacy7c+ttYkRcXFaVKTC4ZlxRJqoYRAqyvJZND8yfEdpyE4c3RQg0ccWcWgQao/iTjzLbBffM4iZNihtLiDOIfqBRmKqU9ONoAABsyNKqnEg3CwY4UWk4bEtpnTdCmAEAANDKIKxJYd7WrT8SI+KlxeS/ESFMlFI4k44exmRaphrapCf/1V/VbRXacqOqN4VGoDgVnMIPp1HFKSrESZ2/InyqPEmwFGj6bzuburqyUxKke+Ztl85ozsTAtYguDSpdDsqocM5wAl4AAKjA87pXcGlVV2xkgbg/Kg8Mh1Ut5WvYyhpdPDJ0E3m1AAAAZAo/3KiFdfw7V0hcX63UPacrCKuu9UOcYeZt37pejIhXFZP/GsKZQuKW0I6LMIgyqgiivyeLgKNPD8UXveS2aRlkH1cm0Him3C/Kh6uVzg5zx2Qg0Lxx1eP0ra+fUSh3nRUveN4kmjbFK6y/aRBBKSqvGoVqWgCAXMM/8mXuKt8fajlnVbvsJ42Xzg+rE22tRYxlQcRF2ONeWS1GLjvu9vB8/Tys4f3q5WTmeh4tFSnMmJKgNyXa8dzjOl+YIoqV0arfFxWtrUMOIZl0utjOsVqvV82M5Vxri+9ZFvBDBhPOr6F1PNzo59+25cTfD9vDBSUXXOk3EOKMFGbu3vplGqa3lBwzYd/ahTuGDIKMOj7CLaM7VwrChSj9Y3bHiKgOdbxswKkINOH8pSS/Ae+rYeYUBJpd9w3Tpz5zkD7wvpmZJgi+/OXT6SvfbYLfIT1ELgrXoXwAWDDd8DXyRiet7TEsp67OAgsUK3hYaiu76nnd4dtNnFxyay2d6kajPFUK99PYkef9HOD96+d9zKwTZhCMar5pVPZ1qeWYXlxjGGNfFeGjVmT7rk4jvIjbRoo0fSzQmG6At7kUZiJu+KsRq9PJ33X13NXdQmtZ5EoNz+tW17c8Ytuo1u9LRCe8Gs4FkRhtrc5LLDCq+97wpPquz8c0YBFiRZx2pvG23qYIs/1ZbKtNLIlB1XO1jnMttd+lGN/BevZZMiMv51/ca5gyPyltHA79OSqUca3ndW+w3C+u03+jhex4tzPz7tnyQRoWHxcjgmiUO/uj7JwZ4ypNimgjfGVaoFRy8hXRwjQuUD4fkGGaKJuvJBCZKkUpnxOBsjwiQziWtlwiw/8RlaiUoSSQqNtApnmFZXyV+fVt1+Yt7au+DxXzitK+ff4zp9Hyi7LLBfP4/lH6097flx9/3kahj+Pj+r1vnEnPPm+C0+1a/MoHqZDgup68SWPa9oftzsv72EdOob/48ylOtvvwoH9o+jTvJCcLb1LYXql3Yk5qxadTfKNwhzrO94caZuFKa3sMy9nk+0Oxbor4xmw1P5UxijE1IMNR1uWx8k2K+7mXb376XH9HPK+7X7uRXFvNzkzjnYjemPt6cVxB0HS+JmQjl7B20o4sTKlJi2WHboXL4+Z53fXeBEceB257eTxXVVlOrHOkGnwOreEOYxIxbi8LeutM7Z7gnIp9jauVGtq66jbyfhuFR1fnStbrqAdFjFidUlLx9dzWLrfZdK8UB+u5muK5VnAe1iNyO7iup0Hi809p26TXMJW97By1CSO1bJuKeg5s4++FylZer3o/IMWjOep11fO65edu0D67tq0TAs/bseU6GhEfF2qn1VQyO6TWS2a1XB5R0/VkuKZxYeJcy7Ti/0Hk58vmsSTnLa2LlPm8qHkD6zIi51crBNn21bYM0pbH4z645ik6fDi7MKNTZnbQrNNzbEhTjx+qMjUrpo6Kk5tekB/kzTE7DA7yDWdSYYY4Qeytnte9x/O6cxE2wvu5LsX9nM3LOSjbjzsZuUDZ14dSPKYqaToyrvP9oZUuhRL5NNn3h5Zyx825MOMCKZKwSHdHCh24qvD6wnNoVQqdmtncWdhTR56FTJFiXsptLTtSN/P1cGWM+VsevkbJ80A6EG5OsdqbPF53yOOXwK2RGfw925DiuTZbOdfa+v5N7r/2PU7T6Snb+Vo+12Rbr67nHkAKO+rA34eQQ/p0y+9WDz98KKAI6hW0bfdr3q4tr6Yx8Vnd3WAMYTLlnUmCLsqYBAYVW9Wmap+PK77EmScrgabKsmMJNMo0mf/luvc9SWNjCY9ZTGSOm4ueNzmbldVKxHG1HieQR0zqf8vkYwCV8NOVPXyT4YLwRnEruxcagrKfcW5+NxmGalzLnc6Gd7xq3Nd6lr8ijh08JrJ8dZ+L7TTBYUxOhSAX8DF9KMV2j4TXt9XROSQ7Etdzh6Zh1wQTilC9xVFbz2bRekOexNysUUSZ61PuMKss547zBu6s5rEdevl75qLa3Wze/7ZLds7f41DwyuKaWSY8N+i7fYnyIGyDNUQ7223KB/Pu3fJ8GhU/kGFMQnXKUIQQYxNkahFq4jhp9PmiXnXBIparxIFAk0Lp7NQFGmVfNt85RLf9/GgNByoZf/GSaZmtqyZkm3gxB5BbOFZ5r7Z9l7T705dWhG9e+vmGwnZzLM+FG4noUiKaK8Os9IGIlvH0Gw3njop8Krol6yfmMfdThtRcqezjCsMg93Uuz7fRspwe7nita8TNmXJDGrWvxBbojZyT5ErONXMxdxLikNYxvLIR5atzlCsgFuxe0e3prtYV9xxKg9l8TciLs24OP6BwJVSryM54QwXrRsCOpK2ORRmdsK1z5Vji7/XNGbTDDbyutoC/U3scCV7V6OFzO0uRRr0f6eNjHbrQBtgpWqLtxJl5926ZTWP067KcMbZcKSbq6bfGcrsE0dP1V6HNa1xmjcJLOJ9XZT5doNGHMrHGsUATOYy36cc/fYBGRrMRHeY9023+mLpQhTSEMrUCph/xXIVsgGTwzcvWiCdKMvRjme8PyRhmmaDVmtCSw0U28HxzWKxZzzcFJq7PSryosp8DLE6cxCE166p12uV0nm8lCzVrLfsp3Qb9WX5nlKoMthvSUGiTx3UG7/Ma3p8ou7S+nlhJFGOwthHCTLPBN9rOQ5go3jnkipsbLdAo14q0QmviMJuvE20h0PAx3pJxG4eEwnlmLr0osvxeM6vyIoK6hPexP0Phz0Yo0mTh2OpTBJge7bxaracraL+uWED3FxKZViTONSWjjRHbUc0Jo00PTCKM/t4oxkS7ReK5ZzISaEjv8CcUaCzzBqb59LbnQYY33fK9IxEHK0VEDgUaCDOtRp+hw7kk684mcAN3BPotOUjWs3ukt96KDyzW9PJNyXrLbM7Fiyr7uZaT562pN7yFhZo1Efu5JKsn40qn2tTp2csOlVBoS1rVJg2L/KY0EtW2Og0SZhrRcSZ2EDSk86hcKxrRoeupsUJaU6K4RBrNtY12kbBAlKUwE3JzKwuB7PDOwokUl00ZVnRcabhv32h6ANJW3bF59275DY2JzlL1H19xypAlz0zchMA2gUGfR59fnxZXoNGXETu8KYh0pNQk5FQTaNT/vaBymXEFmmpCkW2cuh9E9C9fPpSJe6arU9Czz21CcQY0DdxZNXWcQoEml7HboDoRnZABrojQm1bIhzyPWKS52BLutMRVpyRiP7exc6RuUUZH2c9LDTdHUhjKwiVk61RLEWppyg6VpKEBA8hjVR1+AtwuwkxI5h3WBgszIXnpTDqhAS6RaqxqoECzNKOwORu5cA6lDd+X5q0yZGbnGN+3qfft1t/ZHJeVSZd5u7e8iHy6sKxMtamcswuEJuyo/4fvS4KEKJ9GyvRAEx705enzBAaxIlDWo89H2nxSUFHKUpvm++6ac6jDE5WfjSKGPjI2FtD1Nz5Jux4YsewDvxXBeIlvfT5t/tA986bXu88Jc8GiSbTxtoycOnGoJr5kV9AKpIRM0Mk3rfoNVegGWJNlEk+QHO6AmZLEOa1cI8NllDLGerjEEnmTzOJGmvtpE2Zc7ucG3s8NWic3FKGcPLHkp7B6p3qA9zXVp3acsyFpR7Kv2XK+NIAZGXeiTOdQyxNxTQQpkUNhJkQKNJTmb09MGn2uLZcOE5dlxhvEupx9j/dmHbbL9+0r+D7LmvS+bcQZCujnapnsikTAZHDLpGGy0IUYmzBjE2AqxgfjQpJBjCmJFfrnSV+mbISY83kWgWbAowXnTkzUPDaOHQ9o2tQOIjFc3F/TttG4e6dMoCG9zcdFr298+zC9/nXTqKPDyWaXeM58N+1SN1FJfn2TogWahNXcodRv2ns4wZy8qZEiTd6eVgAz6wwhPjKRXK/ryjW8/JWWG3V5k7w1RbHPJMysz+ImXIoOfHOkuxCkCCVFiVSrZvC69KewLkWopInBB1r1yW3KrM6qo5GlQyeHmK6JICU4+Xuez620f3uahTUpXMtzQ8rVA9OiUeeUvJ6vjhLf2iKYYd7uLTeQTxMrymWHRFVrqgeTg0PVQfRcKWXzVUkMHCfUqFp4U63hQWEOGiUUShwT5D3iTuGQzpmyECfTPuj7HLVP/LrvsVF6+JERZ9sdcspMx+pPrdjaDjQ13LlbEZE3ZAkn2JMd0l7ko8kv3AHTXSvbOClsZiWFWSAxnU9r0giX486ALiZuSylPSiyU740e4nStg6pn+pM5p+6gFG7o+5qthHWDyKpcdtYOndzALrBGVHNpC/had30T7OsN7VYxi90zrRSenrf8ZQNZhjSpcJh1ZHu0R6YJn66xumLSTEFiCzkShnl0McQ0zjSNyJ4cWFlHLIHGJHpY1z+eq0YcFyQeysB0VYuYRJVtUL6s8QP9o393X1Z72tScfbWQY6ZlUfJprI3Yx9mchO2gLMEKoSZ3mDpgexv15IzPp23a6J6kNzN8s6l3BgYcixVGFIFGJ80btl7tqf9AlJU5KfydThr6gupM+aKvHUN62lmUygIlXCxN5PVtEw/670dS2vG6lKuy4vXCv/u5c83k+SFEy4c1zdu95b1E1KGGNFVUZQqpu0y2litGHa+HIOnvSREYbKFLRocDhyUFBnGitDkBiTB0yZavxjQhIsRJHPVIPJSRK0QXhwJDnhwl10xhf9V5TO6lgOgntx2ld/xVD3V1ubOOjI0VBZrBw/lI5hJoYoxAjpmWQyrxUnjhG9qoH8JLeJBVATZyeIe1DDPIBFOHOlPHjGn9XLZW7RjKp3krE4TJmW6wG7afMt+L53Wv1QSj2ZyvKY0nfXo4xkrH37OkT5e34TqQH7hT067hTKsRzuSUtES/TXxd7zddO9jx0su/J0mO55IUr8vNwkqLQLmnysO4ekOI9iYQwaJ+N9IQmTaymLgnDAfic2sGD0uVIc55llTsW6cUS7Ddv6zmbSO+l4pCnXdP6+ecCehjuiAj1OS/tQgyJreG6fMm8cWUGyYUEkz5ZnQRgjRxxzSPqjVEjasgpkBzyCPxcMbhOmX7qgg0tv2KFMGKgpYMbXLNseN+oZz25i1DGTRSDFRxxn3BKtAgOLnoCraD98X4kQqFGmkb3ss/fv18o4XQhsaxNsPyjkY4N8tqQ2nVvnqeuFpiztc3Oukhi5orNYFsNeefSfM7cGMG+5rUaeU0NxWfA67dYHuyTvLokLQ7ouE1PuzkFDpULAIt5WPT22inDrs60g5zDB9EyOvq1vC7zefkHO5AtkUIFe9zUtFvW7W8GTR+T7Kar6mr+Zyu9/xycV2ul23K9bKfzyH1e5TGd8gosBiq/pTBocP1iDN7HIlfSR4abOOcexX3Q9q40m8XX89W8rXM9OBrfdKHEPz5yGXUcg+nz9vS4sy83VveSIHMNaO4R0KiXDP1dl5FhDigvyeLIFOaR3HjmMQYdR6yiClSy/CC8ZLhtvlsE7Tkv96jHY3p2JeJMUFlgmBF3IoUvZRt3/v7UfqDZ3Y52+T8hTXxziP5b1vAzoYN3OlcHfOHejYnLi0kL/W87k2Kq6ahQkGbsTcvTwdlJ5dz4ajnz+w6K0no+zSQozh0+R25Q/m/h2/s0gqryGpfk4YqZiEeuc5xsakVQiBSds1EdqSVjsaGlDrRSUmj4ljIek6Kb+xIKW2yjts87wly0yDptaiu5O1cqWaDoVpeXHr4Wt3I343I8ymEfzcTu5NapGpTvblzBupxm/L8su37+Dvdq4UY5z5cstUzT1xjHJt2vhlbrhky5IdRXsuS2OqvtjwqcfLGaHlZAn2eWpYlOdIgYca2vaZpUfMYctHs2DmcxR7kB+SYaUukSOP7Q7JDNFc+uecnp3FZzh2pLZ7XfUhW8UGumkzIm23btD01PdW2xJznJpSOb343aaPTfHKf1dPeRGFNLVi6NU+E+Thu5JAEOVwcISSllW9CuvCW1nJsuTLOHAd5Q+KSxndPtvfFUkSIe52R87HocKkhWXhLkELVnCuTVNXjY7Eiwbm1ukH3IHJ7l8U9n9i9t7TGey4T7ZYIWWVrGg4X+bDL94fk9exKFhZz/7CxZbts83ZvmUEB/VFkjpkQNcypVkx5TaLeR71WCAsRwkvZZ4LydRqGmgQaT5lXVmV6OAeOGX0b9XE0/lohehFV/P/oPvehTQDkBf6BWs0/UMu4g1DLzVEPP01EUmG37M1bSAZ36vRz5ZIaj72ps5U3EUpv99kpVghphsSmLdkZbTDye3OdFMd9f2iGFMr5OryGB2OeDiaNsvJX1uvC40TzSyMqATqBv3NJk1rLdq9JkFJh56mpmlsrkOS8Wp/G75OSjL0egaanAYlywwp7NXXq+buddFvb+R4r1WpV8txNIixmSSs/T19hFGaclco2VFAii7ig/G90z+gCjWm9NoHFNE5dV5x5icNgZFWmPR1EjUoeaxFfKgSaaq8G7e23v3OfC2bW6a2f0gk0H/IGgzsIS9lRcyXH49dyI3oJCzV72FHTzk930iSv4RgmcaGWm0593k15SzzLnQ79O5DGjdz6JsnfhNDF9NjErg0pEPTVeq6z0yypQHFlSkLv6owdNEk7s3WFQuiE+dsSbkuuYEG93pCtbWl2bPmaWO/y0s5HFMXeJNUE+TxKInC21DlYI/IBSVtWD2xlcebdhb+m5L/V8suYjDShwKF/wBI2U/a+ViGhFuGFFLEixnyxBZqjIruqTNUwbStp/9um6+jt5JDBIyiJBPINO2rk0wR5MzuDXTXXsVgTh9BRI0Of+tkyDeonrzcipkSxsY41dzT1xNROE88mQN+uNM7nLPc1b+VK2w0pDFzKDpkkIWJJBYqNaTnwuFO6MkMXSdLvXGoV0bhjHVUVp9lIcl6lLogkEC6W8O9KFvSmIK6jJHz9rPK87q2cP7FtaF1xJqALErtkojrxcUQZ/b1pmu6esblEkgo03vi4qgLNEY/E3py4PiKFJLvoFdhCvTJk8CjEGdBcsKumj8UawTkR1sZ8cio7hndApKmbvXktY8w3p3pOlrjH2DRfXsUZvUO9JIXQPeRxaQ8G+Al7Gud2kuvnQEqOrxJKgs0sSCIwbkw7bxKHhSXNG5IX6u3gbnKYj6re8NYs7jFSqSbIIlS94mYrhDUldWVKF+GtSu7Dla0eUt+S4sy83Vukojq18I/mlhFJ8suEVBNcSBcRyJ5stx5nTOSQUKA5zOWyG6wrTJ7k2duMLO4ZnQYIMvVw7DhqW4N8wjkR1hhCoKIIRZp1yElTE3kVLEL0m9TZMY+vHvKWWxHKIqQkKgOKkvRtwUA9OSkiSHLOuUo+3efaPZOCqO8q3CVv+bHqpd72dSbM8W9BXKeuShbiTJrHvd5rQ9LwxjyQlrAXOrVv5dyHYVj96lYLrW9V58zSWHlmau0TVxUIqieiNX5OrahEBjFCd4ro80QJQfq8nubY0R0zjSqXrdGhRlRZBS+qTA6sf6YJePjh5k5OPHmSu8b2BB12tnBQE2oIFBGdxOFPUY4a+SO6FfloYpP3nB+m7YtzbPV58irMhB0FvQOa5PzNel91dxPIht6UK4DoYYC14CQ0kgUf1wJyku/aRleiryUfVVPBv8P1lHUeSMkNFkU9nXfX9xXbUj6f2tZByeePi+/PbL7PvIFD60OxpulDoFo1Y2m6imrY9wzCl4CEEOMihuBpguzj1Fd9mUJZrnT26CJDlc+VKBtXtmCrWFFYp5w4Ikg84uVCmAmZN3cCbb7neDFvkHYMCgiDwGaaz8DzntvtdNuHh/PlhjnrlC56dP+Ik2Wfc467y4gQ5GajQSL4Rl0+Tevjp51rLHb02fyjmVZyylYmt6IFY9q+OHH/+k103m9St2rnchL3VzMl2YWIWh+b0uy8JnSPpN2h1NmQIKFsHJJ811wLCK733TX15miRoSSunUP1bJtrR0nenazNhrxfvN7xNodijcxTM8DHsK8ZSmfrtL44E9VHjnrgL/jDqghDBgGmQnxRlJAKYSUYT1BsElmixJzwvYgQXaQrxq9doKFjROKB+kKZpAgxYYIb58SsU7rs7UGG8SaxxsIpJ7tNdjw6Fn9e12W95TGadUqnM3Fm8mR3BrwjR4NHp0xxtniQAhyTvYI7FX2Wm6abPa+bINDYcRjTnwryBkceQ404N9X1PK3NE+0iWvTIMDWEYdVMlpVjquG0QylFKMM1IE2SCFOur5/9TS7O1Hsdm51Bp7ouZFJgh2IkcoWlSx9fK7O6H+hRhBrpKF2T93ssldYUZwI6I/a8qgijulNsVZziuGX0+cuEBU1h0MSFkpMl0hkTIbpUiBhVBJphWZWp/tPgqafHaNaZbk6jhc+cWHwTilq6K0YVxPRcQhy2ZcwxFBAtWzrRyTaHyFCfzdvileve95hbcUaKZ/NmT6TNO46nt1DlPD/nbHeXkSAI9jtbOEgV/uFbyk/ZTDdz0mWztRmfYoBU6c15wmi9E9NMeZP6EyZUXeq4U1LrsnPZKVTY5OB6lqQKTRadj205zIMxkEEeq7w7G6vRivnf5jg8Ls1+vHOFFP053OiOBmxXmAdRijSrm+EetFWdM6cX/prKaNeK7sgwumWUdVQLZzKJO6RrKCzQRGIRXUL3TByBZliQ92Adp4DSBoODPtGZdbZtFRb+wcTqbWs7NlWY/YwGnvpafhwpzrh0IEleuGAy/dtP038gOu9ZE2hkJKCuLjfbfsbpnb92smDgDJlAWFZs4qe46lOSHs6HgPCJ9mZ2wpwawE7Si/wKlx18FnBjL9/zuvMuzrhwAmZVIrhe8uisyqKz1ewPFfC7WwMOxL62fyglr/8yxF06qRu0Ccs5zH4tV2HLLa1cSjsZel9TS7AblMo/Geaxlca2TTfMU1Fe2zifJfGvF2Pewx5593cS1RB+Y8JlMtvODkHz5k6sbDP1VX9fjUDQvPMmONvmkKcPxGhY3icpcHV0uBNmJH+4cFJ6C1MSbb/w+ZOcCTNjxSZs+x+0ZoQ7YaakbLI0caplXgEAJZJeL5s+kWLG5Momn5Ftvy3D3hDuBxKC82c8ufaVDU6wfb18gJjnaqIt55yZt3tLcru0LTzJ9j7uKylCQrVwJFuelbh5ZaIcNCOCvEfSqcp03wMj9CcvSb4cE55H9IY/66GP3LQ/2i0Tdz94vqve1uNMUAg5eDh+Ap9du4fLq1M5wPMEveT8KXT75mOpLrz3zdOdbbMQhSOGH7QmhZ+SrDWEJqxxVVEENAV7m8wy3jQCMX/nkixiieM8Dq1EnkvCuwTnBgCgbqRAI0Pc+T6wUSGS0kUjfy9X5FF4bdWwptoxhcSE+V0M+UyqqRQSfgAAIABJREFUCzJanhQdfXrxdSsFtIEE7ZFDQMEcQWIOBbSShGJJjJtXxuP/VTFnhMh7MIYwE1P0+O2dQzQ8PN1ZSM6fXzSV/ukbTxXDp/SEynWIS9OmerTiohRdJBa23B0v30zI00+P0ckOkxRL8Wf1a06h2zf/PtmCFNfMBed305Qp7sx3nlc4ynDONDEc4tSrhbHMlmU9kXumbVmXd0txk5M0J0gvC6ggmtyJFBkJawiPAQAkgu//whyFWSYKVlnC4fe5y4HXmmFNtXTareKJ9l4PpwlDmyJCkyo+S/r40ob2k6A5O65esGzH2xes3fH2Bet3/PWCTTveuWD9Pe+av/aea+YvK8QhC1pXc4iTGuY0wjlm0ohE4vVJ14fLXCldnYI+8Y7T6w9jCgmKQ9+nTy0s0zXb7o0pzvAxunf3sPNtesaZXfTKC6cmX1AgRa4O6vunU506fg4f9p9yt3SQIaaOHsInNHKeJLfQ8TOMjvPEqZH25XYkaWjL6jzbvXNEHit/5D1fjSuc77d8oNCwvUsHuJBB7uAHNfL7u5ZdtVmzPINS8TXTujlndEJjSShikKWTbxJibOOjXmPmliER9O74ywUX77hqQeRJec818/fec838K0nQSvLoUM0CjRRmHqhRmDHtvyY2SUfL4JE6anDXwEXnT6YrXjq98ngJw76rk0OXDR+Gy189nc5/rtMykAVkrpTNdw1Fi0haO/72zhPOt0sKKR97+2mFyk11obhmPnH9KQUXkkuODwU7nDcKyAJTJybXQkSDyHuH2NQBiuN+0ufBsXdL0pDBnpyVh243kog+WQgIeRQpskgw3uzCF5yqIJfIsCIp0vj+0BzOR7Mx4+283vLwqWG0ojhT342f3rkX2rSKccWXssTAJrHGlti3OPTueOvC9bVs5j3XzN9IglbUJNCEjpl6kv/qzh8D27e7FRZk7pkPvPVUWv2Gk2NvUwl2zLz/3TPpA++dSZ570wz5fkD7nmAVTDneQcR2/+o3xwtVj1wjXUPf/oez6IIFMUO7SoJMccOnTfFo/U1n0ote4D407PTTOhqV0R2kCNvs4Z6oTt5FC9P2xQmh0OdBWIRD2C6e9Ank6rzdrOaQPDoRnH63+JxwGX5Qt4CQgfMQ162ckVECbJAhMh+N7w9JZ/VJGQs1uXoggZwzUQlmbbll1M+q2OYzl9Lu3bGqNmEm5J5r5m9b+Pmda8ijPgpNK7YcNCdEMcdMwqpMFSht89OfHaPnX+i2sy4FmisvmUGvfOFU+uTXnqZf/M/RypmCyveXvHwqvfOtM+i0UzsLy8iCXQ/UEKLEeXR2ZRDWFCIFmq98eBb999bj1Petp2nXg9XXLV0yf3lFD73pde7yCxnAD2/rsJUTsIU0642uS3dL3sUZ/ZgNxMxvIY/9KuX/HuQcco6Mo782wUrCsvdwOdlxdf4mWa7rcFHXy08ieK10fM/Q7KG4/Ybk/HG5OPvNBe0MJ+ldFzpBPa97Jf8erXCURLg3TwJN+4ozulhCESKNabryv3TPCBLlyyx7ryT/LY7v3fHm+oSZkHuumX/jws/vlCFOK6wCzYhHoiDMpNiZNrTL5t/Vlvy2XqS4csapnfTZ955ecKfcvfsE3blziB56ZIT2PT5aSHw867TOQvntJfMn0rxnTShUKXJdCUnnlp8erj6TQRSUAs3ihXWGHNWIbMsXPHcSXbj0bDpy1Kftu0/QXbuHCsJSIUzNJ5p37gSae9YEet6Sbpr7jK5CuFZWbXnkqD84dYqHqhCtSyOSv6WBS1Ep75Vy9I563I6QLawN4ow7+hKKMxTG4iN5c7bITonndQ/UeY2UwudK3x/a4Gijex0tNyTJNcFZ54odQ42qKpMWSX5XZjg8pwCoCp9/hXOQc6KFYs3KlO4ne7hyUy4eCrefOKN3iqmKW0YYBBabQ0ZfntDEkuLne3e8KZkwo7COQ5zGqzKF6x0RLMyktCYytVtQqHe877FR2v/kGJ12ajY9dykQdHQIWrawuzDkjV/8xuDqKbWZ/fz77veP0ML5EzNz+BCLNNOnefSC8ycVhiiyFLnGRuk/slsbyAA9RGJTkza6a8fPSu5Y5wp+aqXfAMW6WZcOGc/r3qvlhejN4362ClLg87xuaQe/JOEuyVj8PdJqnnXTtEAC1iToTsNaWB33u1kLHDbkWqBIIs7IzlWvo3O16QVKvibo1+G4rHRxTgFQDwZXzQq+p1iVsEFX5MWx34o5Zyov7oZEtmXTSu8NOT9sCXa1aaXKTdGf6d3xxtSEGbnMDWXr8HiQVadTq8rESZS1cXpbfue7gymsrPnZed8JGjxqSZBcpR1/selYpsJMXvF9op4e7xPt3g4tRhYJG7MgaWe3GnlNxGp6Yl7LTYx+Y7+kzTvfWZCW+HUzl8PPDD432jmsNcm+L3d0vJwLFOwaTJIvqS/tSmN8Libt9OWFes+rVchBBfKKdLv4/lAvh9+1RH7DVuwKjifKNVTwKWGqRGR6X/YaVI5XE75Gl9bu3fGGFIUZGdr0rvkDJFiMCtcjHTMP1SHM2NrANJ/hMz+5LcIt0iZIUeGbP1ZCmqJEwbI2LZ5XsvLVb/77eLs3Ix077h9CyEPrYEnUmCTxY0NuEtk94prZWXeEq8HtrYtSm2oMvzIJBagI5BC2Z6flUJMCTSbHi8///iYOfUyDpMJUX5rXST729Tp5aiXJvvekUK1MJ3PXmEOSuF9aqR1Ag3B5f8O/eUnu03KTY60tn9OXOUFsbhqrQGPocNucMuPjendcvihVYSYkEMGh0nrCUKYTDhK2Wp1HRUFKhjZtv8d9Oeg8MzYW0MbbByMFmcAmgvH7r647XMjt0r5tSDQyQh/LwabkkiZ9emX6sYwrzpgEgEa1QVZiQt4s9KbtqelGnYUcXShYBfeMc9K8Eb7B87o3pO1MCJHL9bxueV7d3ObCTNjJSPIEWLZfKseKxfUbki6nBpIKU5fweZQYXk6z55opwXk76j2vCjmoHGwWyA9O761YmLnZ5e9Iq1Twah9xJo6gQlXcMfp0w2AIb+rd8Vo3wky4joLYVMoxk7IwU63dFPo+d6ithYVbfhKRCNjUjob3m+8cKibkbVOCIPBPmuF9tW0bIALuyG5N68YzC/gHuO6QGIs7I/OnG9xByerJ8eysXArVsFj699aZ18F0Y4+8Mw7h78+NKa5BOqj2pP30k5e3p4XCR9IgaY4PKSr0JxH02S2Yda6RJAJCyKqkv5P8+VY8H5O0y/VZODszcqmCSma7Ek2Y8B4g/B1J/T4n4fYnqRaXKi0nzuw6b1n5TX+1PDOm8XFeTZ3ryvCm3h2vcSjMhOscFST2eunkmFGXG6XzGMQrKSw8/MhIihvRPIyMBvSl7xys3o56qJ1B9PtM38G2FbkOH/bX5+kCmRf4B2cdPxFNfOOZIasNT8FrDYnRHReNuHHL+onhmpy4pEznWV1tYQmzWZ4XIaqFWZMwj4dODz/9LIg09d4Ms1NmtVwO3DJG0rjGL2FBv6bvGB8bKZzemvVx4WSfaQhC8ndya63uPDm//FwLC4VJBfGbXTlopCjD14NbkeOmYTi5v2JRT8092MNuzLTF/iTnd27SKbS2c0YXF9QEvsKSwDfGZ0sd7KhwJinMvNqxMEOFUKYVnkz+O5ySY8aUp0dNCmxsq/EwsX/4xAHyDXmVWx3pmilLBGxKLq21aUV78uvGfz9CTz6VptLWHIyOBv7MmR1/03Y7Ho8+zV6de4GGb4qvN0yqdbv1H8wlWd648VO8LFwzqnDR0+jqGNw50y392xJWQzF1EtcgvMkd3Nl1ccM9m0WVg2xTX23JL1VCTuf55Ll9kMNl0k4WntdS9DXBYmYaopraCVodde2U07jjvSeFUuxJSMtRJ69fW/j8jPwOsDAgr21bWimUSYcfjCTtm0gHTX8a120WAntZELtVuR4ghKp+koT2rKkmuPN1YkONoq9t3tmK2N9X7znFvy39CUVVOGecImhb2eLjJLgtE1YCk9hSPr++3PL5enesdC/MLPjSjku8WqoyRbVBXIGKLO2huGfuu2845ga1BsPDAX3ploOVbaO3L4VCl6U9FQHsQx95upBguF0IAqLjx4OPwTVjpc9g9V7FN0gubah1wdtkEhfq6dybbjQyuXHjjoy6vS4rAehPj5Y0SoDjJ1mmzlmiJ1yyrDYRrdVGp5YfA0S2+3UOm+cSFlru8LzugId+ZSiMk9N5PldVz/a2WKLpNK9zs7ntH+KOkDwu63joZ8fCQyyoN9TFxOdrWsmsic+3W7Xzsk89P1kYaJewutUp/JYtZ+Grv9YwJEWQWcdC4M0GQWxVNbEXWElyHz3bFg6pCJgP8Xcq1rU2Zhn+2XzPIc+pPXxdWs2iS4VgwwKRnLaGhb07UniIlpt8NZ052AYX7C8tU3Z4w95w2AkOtPeleSP+r+h0B+MjRNm/vTsvWezeMSNPzoc7V9ddlUnwBuv7rM6n77veVmX7Pd7O1773Sfr3H86izs6U89/kEBl+9JEvPkmDx3xzG5HWjqb3pLVlUBS5frdliC44v7vl21AyeMQ/OH2aZ3JZAL5Z5RugO7T2WM629d68JELjTna/5Yl4zZ17mcTQ87oHtA6DvHFbU2N4VE0oApO63j6LGygxcl88r3uttny5n8RlIjMhTNpnWNda7jQlwveH1vC5rN6shTeEK9jpkSnsElrq+0Mt2xnw/aG+jMsCZ5WjSWVlI84fV0ghm50sabuLZvPQiGMUlzWG37s0CPc5z/vuFPkd4fMqjUTPyzk8lVhQ28riwB4e5DUnFN5XcNLZuOfzmjxV0GkW+H4xydYuYRF3Lx/DGRZxReaoWRrjvqBWkXk2/06VfqsS7k8c9qZxf5MWrRrW9Js4bpmK0Kay14jQJbLOs27nn2cjzCx67+5L6JhI/6IV4ZYJTMmS9Veu3PSl/zPQFnlT7to1RD/+1ZHKCTaHFZE56XRpvvFpq9/3ZMGV0+qMjgY0fZr34pbf0YSw+HKp4YnXbH5i3ddo9wE/Idlj+SFP0rk32dydhf0oApO6HwOuE9hK4cJS1WhdFseWbcomYWYTb1tarDCcx2EC08xCnJQKQde2Q/4bFvkyuUdpAFfm6eY6RXJVWj8r+PduY3vsbfZIsTZldxKxUHMtP2C4WXHKXc/D8hqFxuVwz9RNGsc2FHCjXC+Rv5kZF1NIQq7SBLSqODP+BDlKXLFNjxO6VDnPup2vXHylk73RWPS+3Uto1MGJFCUo6ONswgL//5V1A3Tn1qHUNzFPyKpK7/z0Y5VbFCOELtDPI8O5JZf/jmv2t7TIJUO3Bo8E781TIq48w6UwTR1b4pui1CupxEFJIHmHxRK/PmHn3hTW5STsh+28ujAj6c3oqbx0lWzTxq1KWnklCj5+GyxPUrelnbOE2zFKoHGe9JlvGvXEnze0Qf6b1Ybzq9m5MmEupNzS5iJFr+NQ0nZnZcrJwl2A3DP1kVXOupVVHhw1g7g8AHEmGyo7etXypZjmMTlrSO9IB0Vh5uUZCTPX37uExgodh3SeohqFJkPyX9VtRJVCgmle6fx4/InWTGwrqzO99aP7xpMA19iO1pLtVH7eyfCmf/znAy2Zf0bu02OPj/7upBneZ3KwOU0DPx1eaulgqZVUVrt2W3CnvloCyfVJw3K4M29yraSad4dFga0GYeZGFsacw/tq6pSElVdSvVlVyhib8oDIc8xJqBGfxyaBpofzQ2xwIUZxrPo6FhL1p7h576gkRh5L3x9a2kIOmpYVZhR62+Hc1FGuhcBd+67MuQAG90x9ZCXO9Nge3vDvdzPkcepzGSJfDy0pzuw6b9khErS78I/aSQ4RFrGBTJ1sS3JgNZTpzzISZj6wu4eOiuTCjC3kK4bjo+K1QqAZb6/BQZ8ue/1jNHSi9UJz3nvjE7Rrzwn7+aW/N7R5tHtm/Ly85fuD9F+/Ot5SAo3cl533Du85a1bn83KwOU2H/CHhDtaNlm0Pkz8e5JCYak83YqMk8wsrrkQlkFybVr4US8gPsWV2DyeGq7esb5jp31Q6VopLmYa7RAhwPVwpI5H4phzDqDLGzoSZEEWgMQmNl3Dc+7o0RBo+xmEyQ9MN4ybOO9MWLj7+XtquH82A7FBe2gbCTLN0op3Aongzn6e5JkIkzxNwz9QIiw1ph63ZsN3jNcNx25ZyyHYqtHIp7Z9aO8n6q81NYxVsFGHmpc/JRpj5yL09NJrAMWMTZPRpMcLAIgUtg0Dzlrc+3lICzce++iTd/tuj1cO+bNMNAkxUeJN8veY9+wsCTdACzSiFmTu3Dj29cP6EuTnYnKaGRYOLq4QprGLR4aBSpaI3ztMo7sSv4Pn7OCv+Qe7QR1VckU95L3bwo2cK+aFQtGCRZh1vb2SHXoavsMARlek/seunXvjmaoUlpEEV3zbUsL96hQxb/L/c76VZhHEpnQNb6MYqFmn6q5UCVuFzdyWft3v4GNue4kkRsSHJiBsJXz+ubMJOfygcNrTcfJbw9yQPOZEyP1f4PM2D06slxbEmEGiWZxHq2oJkJTost/wu5z1EeCDtkO20EEEr9PYMzNu9ZSkFtKVQAUcOvqDS+8L/VPa/0Ker8xHbHNTPSGHm4myEmYUf29UjBr1+8mlpaZt83qYxZTt9ZTv98vHCND4wzK9PM7RVaXnWttI+HxBNm+bRD759Jp1xevMWCJOhTH/18X20+d4hPmditqPPKove9lScJmzt5yttyONXv+sk6n3zdPKaVFaV+XM2/25o8I//sHt6DjanpeDwlCSVPbZxlYVaqinoDLBF1NlNgSVZbxTbtNKStsoDKnI/VpueynPZ1RK+P2STvaP2YYVeiSRqOZyodk3MErf6/tZyPAc4t05DOr18A74uxn4OKFVBVKfLHGWIs8+beH9TsTSz+0oV+dbm8amcjlIyvhkSN0oXxZq0hDT9+1wDFzeiQl5ENbUsCEuV31rHujYlrYbGwnKjwiQG2CFQz77HPlcaeT7W8duaJbKaTk0OSg4Brqu6Yj2/61W2peI3P4ttYYdz1IO0tLhRdxgrFS/z+LsywAJ/Lp2yLeuc2XXesq0k6FjhnzihJ1Xyp2jhTdkJMx/d1UPHCqFM6SqQZieQ3fGh/W/MPVPh/qgMcdr8u6GmS3Art1fmznn93z9Cm+87Xj4xTjtGurKCUqV3a3iTMr7vCwfpk/90oCAUNRuyHb+y7tBBCDNukEIC37xcWWfCzyV1VFMIkTftsgz0HNedUSVnxtqYHwn3a3mMygOkhLfkJlyCK2vMifn0WN/fOMdzQDl+DXMj8Lrn8LZEPcXt4X27RKkEcj133OLs8ybuzKzIW6x5I+AwSdl5uC7HT8+38TFb3W4OJxW+Lpmq9rkmTAzesLZvYLWxbewsaenzro7f1qzYlFeHQxPQm1EC+Irjw+dTHn9XtvG9Tm5DmFs5rEnyZWu4iCYoBKr4Ypp/fNy6nSsyEmY+vquHTqQszFQLb9LnqdZ+pmll48oFmre9/Qn64k2HmkZckCE4t/38KF36t7+nXftOjE+oCHULyqeR1gYU1U6Wsu2eWaCROWhe9ep9dP8DI00R5iRFmcOHfXrzVY89cfXbZszMwSa1NCzSyGvGMn7K7CqR5ADfKMu8DwVRJstOE4tAc1O8Wc91h51vdHp5n29M6WanTFTLQ6eX93MNizTXpXj+hufrMj7Gmbse8o4iAuapc7aXk/4uxTErolTty6rq1qa8PGXma2CWoXgb8/yE3QV8/V2WYc4SG+F3v63aP00iCgykzWxbpUP+XclLEvq1WYVsJ6Flw5rIFNpkCm9SwkmsoU3j86zb+cIl2Qgzn97VQ4OiFMokxkRFqFJNYU1kCbtRQ24M4UtqaFJ5OA6Hg6lhN/rnyRziJJl1Zid98h9OoSWLJ+YyREd+LZ7YP0ofWvM0/b8Dg0TdleeQiBMiZmpHPVSMtPAmY/tp7cjT/rK3h97xVz3U1ZWqAzM1pAh3y/eO0E3rDh781W3PgDDTIDhsYSX/QC6t07a8jUNI5NCfp5slts+uVPYxrvtnE9tuN8QVZPRcPfV0GGsNa4pYjrrPcY/pJraub2iWG16+6VvJHdKlMcO7iPd1K++r8449b6eaF25Pszpz+JrRy2Escds7TbZxiKRTB1uzhTXpcOiGq2M0wCFkpUp5CcIzEoc1qWQQilcR2ppg35sirMkE7/OajENTUvnuI6ypbP1z+F7HRcjaJv6uVL2f4O1Y04DwxPV8LWuK3+OWFmeoKNDsIJ/mF/6p1klWxYbK/B/rdr4gI2HmH3f10HEWZkIRpB5xhhRxplpuGXXZpEwLxRtT7hm9zSiGQEPl0y54bjd9+IMz6Zyzu6ijI4vWjUa6PI4d9+lzXzpEt/zwMPlzRzVhplzgq8g7ox4LvT2qtmV9Ao3M5/OOq2bQ5a+dSp4QuWhHyc9/cYw+87mD9OhTw0e2/2rutBxsElDgH0k1hju8cdbzeGxtxjACRUSZobgPwzwlhxopTKQlzliWS0reFfVYNq1QoKOdu6Eg0pL7mgc4z8nKDHIXDHAHoi+r72ezizM0Lk6vTlGkKeQO4+NQdu3PizijbU+a4sFeFn3S3PemFWdCWHhezdcBV0Jgqt99iDPG7UhTzN3Ix6ueB1QzWPzvdZjjaED5LjfV/UA7iDO9FNDNcd0zpCcHJhZmLsxImPnMzh6SyX8DFmaqiTOmJL9+5bzCJhz4ZE/8a3N9RAk0piHEIi4UjtN5E+idV8+g5z+/mzo80RA3zZ69I9T3hUP0i18eI+oIyJ89QjQloo0Djlgqa1N7EmVrO5YJgTUINFQ+XYo0l182rZAweOpUL/M2DIWtjf/3KH3jO4dp3xMjFEyg49t/OWdytlsCQL5xJc4A4BLFobYigQNPp6FOLk7iXA+xnhRnieYgrFVICzvH/VGOBe6k99mmR7BVTxiaJizYruZzs9bzcq9yDlpzbiXY99jnSjOcj+zYXFFnW6uE3/1+R66f3ogyz5GkLSQmOHdcbEsSYaRmt3GM7ZnD51KtzmcT25TvctOGwba8OENFgeYI+TSl8I+tk6yE7GjhTet2/lFGwsxnd/bQUdFPY0KrypRQnAk094y+z6ZqQqQLDJbwJpuDxibSlL2vFGmkwPDi5ZPpda+ZWhBsJC5CdkZGgsJypSDzo38/Sj+57Sjte2KUyAsKuV78OSNEkw1trLpnKCK0SRdddHeN5dyLJdCUta8m0vB72XZvuGIaPf+PJ9Fpp3YUhBMXjhq53KNHffrVb47Tz24/Rr/YdKzYhh1EQWcwtH3T3EnprxWA5gbiDGgF+CZ/qVYlK6qiSnizvJXdTcgj4RC+zoTHZIaWvzA8FntYOGmZY6Gcl2GnVg03VB12/XDZJUc7z0hrb1Icq8Rt3lDnKhiHhZGlyrVBFYL28HCIrxGZiB2G3xUy/Lao3+OtfE61TE6ydhFn1lBA1xvdM2R2jLDQsG7nBRkJM307e+iIKDpmxjSXSxJxRnXP2PLOqKJBlHuG9M/Q+LhaBBrSP1/pAgmRIsPzzu+mF17YTeec00XnnF0sxX3seECdHUQTJtj7MwUnxzG/IPhInn56jH7/yCht2XqCfv3fxwuVo0pJeD1OuttJRcfMZIOrSA9RCvfZ19qpom0t551pvtJ4i0Cjt6syv9524f9y/2X42AsunETLlk6k00/voGlTvUL7nBgOaPKk6D6hbOuJE4rhUsPDAe17bJR27Bym//5/Q7T5zqHC/6XExV4xkXHQFZzYfsfc7sgFA9CmQJwBAAAAAAA67SLOSBXukZJ7RutcVwgRxXnW7Tp/aTbCzOd3liX/TSzOWMYLk6tDD22qbIfy5ZJBdKDy/40CDRk+Q9o0ihZqVC44v7s0XYoPz2aXjUSKBfv2jZYqHO3aPVyoFFXAViXJY2HmnBGiqZa2LGszRZypEGKEWZjRBRzSx5sFl6oiTcV7i1CjtblMynzWrM7S/89+9oSCaBPy21C8omKlLdmOlaXDtTYMhZnbIcwAYAPiDAAAAAAA0GkLcYaKAs1qCuiG8s6y0PJ9lDrJ/bueu/TiLLZr4RfYMaMk/61bnAkM40lbjkk4iOOeCbR5TIKBTaAhy7zqq3WcqJyuoo8zdW9Kgkww/r86SEGhI6DgnFGiaZY21tulWmiT3rZkmUdvG4srRlSMi2g3m+Bla8M4VGtDuTsybKozGNn+n3Mn1LkWANoCiDMAAAAAAEAnh0WM3bDrvGV9JOjR8g5loDkpCsNWEoWkRM5Z+CXOMVMeB2wm6rZdVBElot6bpkUOgXm69tnACwpD2TR2VpTeq6/q58vmCzjkKBh/rw4d2mCaFr4Pl9uhDJ0FpwcFMpTJmLs8sLdtaFCp2H9DG1W0r/Hcs7RpUNRyPK2NTO1mfLW0XdxBb8Ow/ZT3sg1pgg9hBgAAAAAAAADqoLPNGu1NpaeVgl0EYQe3aAiQwsyKXUuXDrjekIU37eiho148YSZE74wXhANhcI9o4/V9NY0LlydVAHU+/XP2kZXLLrwURYiSk0Zo6w7K5y97tYlLtWIRPgrikQxlOnuExHTFaWTYD/N2K+1Vfh6VT7PCC4tqzlKbBcXDQ4LbU9kOU5vp2xPjsFUQJe6VxCIWbLqCke3/cS6EGQAAAAAAAACog7ZxzlDRPdNPgn5iyTtSFGaWuBdmChyL6ZipBZNLQ3sfkBaWok6vEH/MrpiydvM0p4vFzREIg5PG5vYwDVUdPYbPd2iDp4QwSadHF+eYUXPK29omql1C90yEk6hqO5qmeQaHDLuICs6kUCDRB9v6bG1brc21Ng2KlZiKThop704MRiHMAAAAAAAAAED9tJtzRvJGEvQIEScHLhBsJRIrdi3KRphZeMPOm1MVZoQhl4jJGWRzrJTNY3CDUISDpqo7pHx7AnaACDVprcntUXXdlnVY3geqmCSKIoN0zNB0ze1TvLj7AAAgAElEQVSiO46iXD6628jUlupG6fPoO6g7i9Rjqo4rtVdQWn24DULfBn01dbRhoK5PF2w6g7HtPz63K8aSAQAAAAAAAABYaCvnDBXdM4c4vKlI6JhZmJEwc2NBmOlNbYF6R7+Kc8boqND/tyV9NY3T59VdF7pzg1/DnDRluWniuGHiOj3CvDcdxaHMRdMVFIWZGYb2MbZXUDnO4KIxu2eCyv23taPNWaS7ikzHg/PHlNqUXTXqUK39Co6YcP6OoNR+4/l7qJSnpzBMKAgz7SjwAgAAAAAAAECqtGXHatd5yzbM273lRiJaTgGt2DV/WTbCzOdTEmZUd4U6jlRnhcHREYoIQUBCCIPTQ12Jxa5idX5Q5QR9G3WXjOJECUKRx7B8ERi2g5TP6G1gei2JEAH5s0aIThKVThh9m03vje0+3talzbc5aDzNFaM6YgKlAaJyBEVVZ1JFIb29bM4ZmyvJJpYVwpkCf/vGZ0KYAQAAAAAAAIAUaJtS2ibm7d4yddd5y45ksa6FX9h5MwXUW1b62lgem0s0+1RRJlv4hvlJmV8v3xxZNltJ0hsoyyqbV1SO0xPQVkxXevr6dH2ciWrjo4QE/X/VrSKKzo+iMEPl5cZ534WvtCWRVkK7SplxfzwcSpjmIYpdhnz8f6X9TW1qa8+or3RUO1ZzV3nsnpHCzK3P7IhYCwAgApTSBgAAAAAAOm395DsrYWbBl3bcLEjU7pgxOWT06ZEuD0sOFTI4VIRhnOyM+zFzypSWE4wLC7pLhqqICpHLjTHeGH41Xv7ZP4tDmSKrMiltVuZW4fFq+5HW1ux4Mabh0R00FdM0Fw0pbVmyPFmcR1SDOGMjynUklDLkXYG//QcQZgBIiAyv3YRGBAAAAAAAIW3tnMmCBTftuJl86hW+4ryI65zRxgvd1aE7ZHTHh809ozhABBmmk8kdUsVFQ5bPk+amIcNnTO/jYBUUtDwvqmPGNzhhwrZQXTI2J5LBgVTRToGyvDgOGr29TG1JyvKrtVct7Whz0OgJlGUemq7Av/t7EGYAAAAAAAAAIG2QM8IhC/51R/rJf6PcMlF5USqcHsV/S/lnVIwGD4slRHV7mLYhdIwE+ocq3yZy0QhLUuFOIv+MUaKZehsoOxnRPuXvNVeNPl1rJ1szlhw0ulOmbB5DW5a2W98uUfnZWtsxqqR3oXS279/93T+AMAMAAAAAAAAADoA444gFX0lRmDGFruihOGryX/UzpAgBVEPnvZpAUxGCo4k0+jaqr2ryX+OORqAmurXmRlGS/54+RnRKUHS6mMQtRfgY102CccFDb7dQoFG3QW8rJRzM3tzagdLbz9aWRvEoqFxsZBsa3pvaktsw6AyCu78DYQYAAAAAAAAAXAFxxgELvpqCMGMSOPT/be6KEP1zZBZdjO4Zy7ylhdpyyuiCTFSOFH3ZUSF2JseOLjKoYTgdAfkn+0Sn+bwuEd2OJueMVQwxbFeUQCMMYU6kCCuCysUgXaAxtSUp4yni/2ptqLdfqR2L2xV0BcHd3/qDtiu5DwAAAAAAAABZAnEmZRbcfM/NFDf5rylhbzjeUD5alo4Wtk48ae/V5ak9e5OII2IINKpYEBBVbLTJ+aFvT9m+G8aZ5rWZaWwhOFxRyD/JJ5o1Vhk2VOaeqUz+W9CcogQv1T1jEsN0gSYYn69CjyprM0WkMbWf/mpaX1R7xW0/JddMwTHzTQgzAAAAAAAAAOAaiDMpsuDr99TnmNHDbaKmWd0ylpwwaugTWcKbvGKCW6NAYwqxUcfrThp1+fqrPs20r1HYQnCIhRAOZ/JPGiM6yy+vylTNgaS3pzAoKmXvtfAmMsyjjlBeCofBGhIWKOM0oUZvuzghTDrGNlTWKUUZr1AuO7j73yDMAAAAAAAAAEAWQJxJiQX/VqcwUwvV3DJ6B94klticHh7rOxxaJIRSUUif3xR6oyb91UOeKKGoYAojIhov8Rxuk3TMzBgjOjtCmNG3nwxtWlqP5pCJ60qxCTSmMZFhTIpoQlq7GkPOqqA6s9T95CEIc/VIYWY9hBkAAAAAAAAAyAqIMymw4JspCzMRbpnI0CYRVOnAG3r0lk5+hYtGD2syuWiEQaghqnT01CoqkCZUqAOx20PmmOnxSZyjCDMmYarifWWIkmpGihRDbO4Zff4wFCq0zYQuGqXdRKBtn9F5pIkq9WAIZwpCGaajEBIW3L0OwgwAAAAAAAAAZAnEmYQs+FaKwozRvWFwmpjcH/r/ZPu/DoHGJGqYtoliiApxXTMmMUcY3hfCcIrCDJ0zVllNqjRvYE+6a3AYlXLPWBPGhOFgEQKNcb9YpNFyClmFGts5UG87hgJUmchVzNNTEGa+BmEGAAAAAAAAALIG4kwCFnzn7mtjJf+N6rubQpXIICaYBAXjqyYmGAWcmAINJwoumkssIo3u9ogKZzL9X62tTOJCyfFRDGvyp/lEs8dKYVnC1uBR228LdQr/MSUAJoNAozuM1GWVrV9x0mjrC7S2Mx2uWtox0I9J+N4bT/5bCGX6yrMgzAAAAAAAAABAA4A4UycLvnv3KiLqi/vpUjiSSWhRieGWKVuWTXxR3Rwmd0iUQKO7YxSRpuDsEAbxJ8rlYdq/auiCgirKCCX571SfaO5YMZRJdx3pzhhTOBYp4Umay6gUjVUtvMkjpQy4YQfV9qw4/qFVJ1x+hFijYmtjW/uGDhlN4GJhhu7+VwgzAAAAAAAAANAoIM7UwYLvF4SZdaku1BiKE+Hu0MWCqvlRxqsyjc+jzKg7PnTBR3kf8IiCQ0UXZKptB1GliGJqC+21sM5QTAiFmSljRHP96FCwCreKKJ83qv0CQ3hT1H6RlofGIHJZ21cV1irC06ooWsY21PLTCG2dniLM3PSsuJIZAAAAAAAAAAAHQJypkQU/TFmYEUrITNQ0oYsGFvcMVXHGeHrYDYsecQQAbTsC1fUhNLGGyLxPVdsj/Ggw/r/q8hBclWnyGNG5WvJfm7hl2f5yQUYTVdR9EIZ/qgo02oL09cYNBSsdyxrQl2dqRxnS1El095cgzAAAAAAAAABAo4E4UwMLNqQozJg64ia3jP4ZXXwgg1CgujBsoUsV4wwuGtM2RwgcJbHGtp4oTEICGUQFWZVpsk90rgxlMq9ACjtC35fSMg35Y8pCjESVJqsi0JQ5kMIwIksYGFle9baoB1P7CRovP95FdPcXIMwAAAAAAAAAQB6AOBOT+Ru3rxIkkgszuqODIv4Xls4/Ke4ZsuSeoToFGtVFY3J72FweJvdOLW1iey0JM5z8NxRmSturuI50cStKTDLti41ArwZuEGh0R4zafqqTpqqDJ2Eb6u2oilJqKNPnz4MwAwAAAAAAAAA5AeJMDOb/3+3xHTOmLq/QOvD6NFsoDpE9TwpZOva6oBNHoClze2giTTW3h03giCMsVGsroWyLDGWaEjpmDOuwtYtNCNPnL223oTQ2h4KVtaJaEtuWR6YijEldiEFUq0fgMol3pXFKnh5+DToDuudGCDMAAAAAAAAAkCcgzlRh/o+3v6XmUKZQDLEJMhTDLRPH6aG6Z2h8nNUVEydGpmy7lLwvJqHG5PaIg80NpAoyNC4olJL/PlPJMUPaZ7S2K4U2Gd1HlrZQw5t0gUbdVaGmgYlo10hhy9Bo+jqrtWvZ7EH5OKG0nyrM3ABhBgAAAAAAAADyBsSZCOb/tAbHTC1UuGUM7pgyAcQQvqNQlmPFhu6IUUNw/n975x9i2XnW8efcmdlNQTSgIv6hzaZ1Z2aT3Z2dapUutClBIyFiIGirtOmCRdKCNIb+UVBpK1hDqdr+UygIjRaKItQItmorZfsrRdTZTUJ2ZlLspFqKRQqJQv9odueVc+455z7neZ/3PefeuTOZH58PHO6955577jnP7P5xP3yf5+2TQsEIE5Hu9XoSISWkvNfeykJKKlRipmplKuLPTzMbp/O9odu+Za8vI2jcNqdC/KHKOdEWCa6BlqtXbql99fDfsCDy3J8iZgAAAAAAAA4jyJkEq/80ZzFTqBWWJJ346P0Rr1GyxhU0Wq7YtiUp/GNsa46b+nCEyh7q0qmHljOjevjva25lVmUyA36nqZ9tJ4oSR/2Cpk3RNHLFSppUTcW4mGnTR2Kut3BqWUqZ0bgl7LmPIGYAAAAAAAAOK8gZh9XPP/MOmcfwX0mkO6ZNfPQlPUJC0OSkTmr4b2qezLykghUK9nGkhv++alfktbWY8b7bJpCMZClXjyqKIv5soWrQHpyqVULQqHpFkqaZRyMqpZSrxcCOM/ez+nUx+buGJnlUtjJ9eBkxAwAAAAAAcIhBzhhW//mZ+S6XbWVCKi1TtIalm4JIzJqJkh7TCBotYrzhv15ixpMqMqWcybXjjKQ7/Pe2XZGfuRVfS66O9v6sVHJlyAAzogWNlVjqNB1JUyREjSeZpsGVW6GtQ3MNzapMzz2OmAEAAAAAADjsIGcUq1+cQsxYOZKTBZLYZz+Xe5QBSY8+QWPFgpU07T71HWKu2z4f8tM/lZYpxB/+e9uuhLO3pLDDf8XM3mnPaVqb1GNVC/2eS8+NFKIG+BZmv/m76fFBnqjx6ubNqnGvw2klK7rf1y6XvRDkuT9GzAAAAAAAABwFkDM1q1efedtMiRmdeHHf7wqDdu6Mm/gwrUvuT+sekeAJGisRxHx3lJIxoibVfjMk/ZFKeuh7VFLh1m23RM6WrUxFfB59vZ7MSiRl2vYme92drzA3p2vWEVJNSqWI62fllxE17a1HUq2nkPrSvfRRoQcoh/GMmT9CzAAAAAAAABwVkDOlmPnyM28SkU/N9aQ5aWOFgn1PEiLEtuJkZ8qMRVAVNkklM7zkR5TwcdIa7Vu5tElCOBT+FhaChNtCLWac4z2RNCR1JE4tU/v1jlRt25oZSePVz0keuVUZqlEK57ERMzKWMuX/6Of+EDEDAAAAAABwlDjxcmb1q09fFCmenNsJPZmQa3ny0jNWvHjCoXwxKvxz6U2naOx3WUHkSQX7XudeQ7w/JZtsO5NMxEIlZk4HCWdvduoWCuV4eusYunNzjIzppGeS0kvSgsamYqykEfPd7t8sUce+Gnr1K+oPNs8X6lamD64gZgAAAAAAAI4YJ1rOrH6tFDNyVURu3/PJPGFg3/damzKzY9x0jZegyWFSNJVDCCbtYU81VCoMqYmY89T3E0ZhImdKMbN8s7tcdlJghYkIGZI60q7Frt4kzvH+jvhaojamoD6WGPw74M/lfl/nuRIyup2pHP77fsQMAAAAAADAUeTEypnVr89BzKSEzJC0jNd6077X016TEjRe+45J0YwPCxNJ47XjpO51FkziIxQTKVOJmaVdCauZVZn65sx4qaOU2PLmz7g1Vamg0FNbK7hEzZCJPjNtEcPku3QNrJwpxcwfIGYAAAAAAACOKidSzqz+y9MXphIzjVjQyYvcLJn2tRnwq37MJ9Mz7fOemEWfoLFiI8TXHPTqQaJm0+wlNWPbiQojGEZWzNwc1yiSV4na9bUOpZ53bsesZuXdg07qFBJfo72u1LWIfj6wkFFaxrYzNTWsh//+PmIGAAAAAADgKHPi5Mzqvz59XkS+NJdWJk0q5SE9P+I9EWKlS8rT2GO1RMhJGpv2aJyMkTUNyYHC+lK8ZZ7FSIVmK2fMLIWxmLHLZZvrDdFy1M2cl6IrLtr77UkdtbetltjWm/1cocRKU19JHCd+XTv7+0iJGV3LUsqMxnNmwkK4Y+CZAQAAAAAA4JByouTMyr9fv1BIMV8xk0p5iJUG9T71uhIEoYh/kEfSZYCg0RJBt/YMSZz0zEUJqZWXUvUQ9Z027VGKmcUg4dzNYasyDZVcnWt32pu80qVanFL3FaQ7BFknjcR8lz1tXwm9y+jUUrWDFeP/uWExXLjx3tVvDbsBAAAAAAAAOKycGDmzsnF9fsN/JdG64ooEZxUhL5mRa88Z1efYdT7nzJSZCIREC1Zv+40h1cKV2lc4z+u0R5uY6dRJXUynhum2sFZa5VJHjXiZRtDounr3F+zftnmdGALch70u/TfU9WvrWEmZ8n/uhRuPrT47xTcBAAAAAADAIeVEyJmV63MUMzbhIVYoJFqKRLo/3uv3qtkzkkjGWNlgl87uu87gyIOcaMlIjM79pT6vX+u0R1G34OhWJi8pk6hjp7Updb0p8SXOsfbW6uWw2zYnfR/6vCmpJjIZAiwZOdOKpsQBObnV1LCUXEty+cajiBkAAAAAAIDjwrGXMyvPTCFmUmKicOSBfq9JUUgRSwLdZiO+RMgOqPVm0Nj9XtojamfaY8ojV5vOo1nquZEKi7vjViZPEqUkTe5R1OwZe33e/JmolnHtooHPqfavWQVXqthJuTXZQj38Vxbl8o33rD6V+gYAAAAAAAA4ehxrObPy7PXpVmUSI1RSKzOlUh72eeq1xD/i3fkz4h2b+fXvSQU3cRL8Uw0Y/Dv5LnVTkaBRs1GaVZnO3YwFUiYtY1ubovRMe5zT3tS8X17DbkLQeEKrXpFr/KdPtKKlWtBmxaudljLFZPivLIbLN37nHGIGAAAAAADgmHFs5czKjWvnZR+G/1bpiuaXtNdSI6kf8UYS2B/4qX0N3kyZ1NLXubRHtu1noGXwkh6ipEIrZsbDf3fvuimFHf4bCaz6wlJ16b125zobQRNUvYKSMl7NpCtpxpeUmRkUCa4B9bPXb+oXRA3/bbZT4b4b70bMAAAAAAAAHEeOpZxZ2SrFjHx57qsy9c6ZUb/UU+mKKMUyeZ6dP+Nils/2rneeaY+MUOg8b4b/1mKmuypT6CZ0UrIjxJ9p0zP2WnR7kyWXOrIpGns9TcBIz+4R6Q4QtnXsq2tCbLVCxgqu8fDfh26869zne84MAAAAAAAAR5RjJ2dWtq/Nf1WmXFtSXztTR8aY1YfEnKsUGrtG0Oj2G5t+ab9TSZpsO5P5Tsm8tjXwXketOKGVCuXw391zk1WZOmIlWUezbLhzD6630u1NHp6gsXUtzHdm5s2E9qbii8ktzR10kXvamao6LlZzZh668ci5zyRPCgAAAAAAAEee45ic2buY8X6U5xIWkkin2OM7Bxq0oGnmnoREisYKGJ3caa9D2Qzva1NtQPZ7cjVqhIyWCgsiuwu7EqrETOHXLZdUsUkcPXtGCqee+m8wQNA0IieoD1qxlrouWyNHaIVgdnqX0ytmmuG/4eEbv30XYgYAAAAAAOCYcxzlzDAxkxIeuR/q9fN27ownX4pEOkYzZK5MnbRIpmjEvLZCowhGKmRacVI18d5r4i9WotSzUXbLVZnubpbLTrQwpZ6nhIi+8KKIOsi6KaSEoHFrpWbR2O9OJY+859PgXXPTwtQd/vvwjXfe9akZvgEAAAAAAACOGMd+Ke0sOSEjTSqkiIVCKi2TS1oU5ryDVmaS7jLbroRJtDyJvc50K86gOkX30G3BKZd6LluZgmplio7VKRi7nHf7vll9KZJf5pYaMaTn2tgltoP5+0YpImeZcV1P9zNDi+fQScqY5FH5P3IhPHzjtxAzAAAAAAAAJ4WTLWdSpFpvUs+tkNHywUtaFDKZP9MnaPTyzvZgTy7tl1RIteCU17cwXpUpRMN/u1IkGuibEC9JwSTdyEwkaPQMmWBPbr432cbkiBpxjk+ljzyiNiYlZEStyFTVUR7ZvIKYAQAAAAAAOEkgZzTeD3FPzDStTTbh0ZE0jnyxM1VSgiaR5AmFkTS5JM1epIInc2xyZTRZ7rlKzGgx48mtaVJHYgYDx5XwBY17qErieC1h2TamRNooJdWiOqpie3Us9HLZZfJIHt58GDEDAAAAAABw0jjecibVrhQdZ9qX7GetXBjSziSJFqJg23BMyibVgmPEUPO80IN/cwkUfb6+mmWet1KmI2Ze9of/NrXVM3i81qaUGLGzYzr3M1DQFDIZAGwlTV/yyF5T+9U9RUyJG5s6GtUXvlAtO/7w5tvvRswAAAAAAACcQOYiZ5Z3Nh4UkbV6awbylqsmXS8ft8+sv3goS+vJBOmRCOr9ah6Ml54R57OaYNtwzGBaex1esqMJdeh0hyhZk7vnobVp0jpaKMgk7REW61WZgvjDf72aFUqkJFNHCbEUCbAeQRP9bR1J0yfZZmlfsq9tWqaQzvDfsBQe2fxNxAwAAAAAAMBJpYiW/h3I8s5GKWEerbcfyXzqJRF5sjzuICTNyva1UP2gLpMpu3UaZrfeSs9Q7quf6/2dY4OY50V3v3qvXe5aHy/O58V8Xh/X2dQvfO84cYSBtz/zZ20ETiicg6zccNtwQqX1du9+Oa6hV7/dbo2KTh0Lpwa6lk59o9oVnf1Fqmbec90i5tUuVcdOOsgrsvcYJiszVUOUReRUeGTzrXd/IvEtAAAAAAAAcAKYSc4s72ys1cLl1VN8rJQ0V7bPrD+5n2WdWc40AqZpN9LHBEfcaEEjjrwRe/xAQSPmeElIhZSosZ+ZBi/5EbXi1GLmvBIzuRo6gqvQEkv/jSSu7WBBI11J0/FOubpFdXRaqWauYehKmtGkjuUAZSmH/74FMQMAAAAAAHDSGU17/7WYuTqlmJE6XfO3yzsbVw5NzYuepEi7Lzj71A9tCbHEiI4L7mfbfaPuj/dKgtjP6GNGZvNEytDNO++C2uqVhGQplEkP2b3wcn8dU8+L2ju19x7iWtg6SYhrMDJ1KbrnKlMpQZ9rNOCxmQHTbBK6r/u25m82qremhov1Vl5T2cZ0qqrjY4gZAAAAAAAAkGnljBIzuTamPj65vLNxz56ueq+4CREjYLznKZkzeHMEzcgREp6k0Z/NSZrcfm9bMEKmsy+Mpcw45VHJmVtazET1CH79Uo/2vYT8igRNToQ1YsRImlA49bW1mqZuI1M3Wzu1LzTbYiu4Htv8tfN/tq//xgEAAAAAAODIMFjOzEnMNDxRz6w5cKI5K32JD53wkLQccNMz7hYmEiOXeLHCICVqvETJtEJBLec8SX2ENu0RiRlvVo1XU5uQUfuS6ZncNlTQWEkz6hE1e9k8uaWFTJWcqQXXkoicDo9tPoSYAQAAAAAAgAmDVmuas5iRuiWqbG/66IH8LZof4t4QV+sZ9LHeo/6METjl/J7CLv+sz9uZLTP9EtrdawjdY6MvS9UiIafcFEr9uBjk1t0vT96zdSx6rlfXTe0PzeUU0r25kLkFb4ltW1ub5mm/d7JKUzD34S7FnSljdFmRIDKtWAuV4PqTzV9FzAAAAAAAAECXXjmzD2Km4dF9kTOFfE+C/KgrZCbHdAVLKwzMstmRWKh3WtmgfsQnBY0rHOqL8CSNd+32e1Pn9YY8e6IhSgc56ZNFkVvN8N+c3LI1tcfa+kbvqZuxIsrSCBortTzJFX2/vvbJF7hzsVP/dtwamjST6ERN2cokf7n5wPn3Ju4IAAAAAAAATjDZtqZ9FDMlr67PP2++nTxfkZEUnqjwHsUIDPs6NSDYS6V0BtmamTLeo7flWpbsUN/cbJrO3JS6DeeUyK2LP4jKFbxatK/zw5PtvhDtV8tNe5/Xg3vtoGD73Ktb596dIb62tctu9v320dS2GQJc1vF0+OvNB86/Yx/+rQMAAAAAAMAxIJmc2Wcx03CHiFzf9zKaFE05d6ZKt3itS+I87widXKRDHZVrcbLXFkyyQ6dLbELFa6+ahqR8Ch15UQ6v3S1bmVIplNR9WNkStYDF6aS4s6mUHcWwVIw+wEseeX/fqPVphjrqz+r71cJt3Mr02c37L7x1D98AAAAAAAAAxxw3OXNAYqZkP5IzT7bPUqsHifOjWtSP6lwyRgYc4yVo7PLP9jOdJEtID/+ddXitTdDo5MhC6KRswpJZLjtKx4jfxqPrE9U189lUgqavxp1kjErSpJJHqdqkUknelkonLUo3SbMkEk6HL27ef+GB/n+yAAAAAAAAcJKJkjMHKGb2Dy9ZkkrGeO+LOYd93p1kO/6xvuskMIpxgmb8tOieOzsbxfuuKQf/dg7Tk28T0kQlZjZ/5Xyx/M2NMLiORaZ+XqKmPWn3+t0EjT7Gfpd3HcFcRC555F+Gjyf5mmu0wmhUCa6vbN134d4BZwYAAAAAAIATTic58wqImRf34ZxXO68SaZmgUzVRMiP3XiJBY2eemERGsKkOL9khzjm8+TDe0te5mSk26WETNYtjKRNOjcVMfR3/EdVRXWNn7oxXn1SypicNk03Q2Dq5y43rx6ASNWEyz2ZPS46bGTU6OVMvOR5OV2LmjQP/vQIAAAAAAMAJp03OvEKJmf2YN/NC+2xIWqZw0jHS/LB3Uhsam6DJHdukaJoFiXIzb7yUh5fuSM2cSSVBrPgY1dKoFgyb95/Xn/ymiLymrYW1MYW482OSs3rcmjirN3nBGT2LxzswmTjy/rZTDOpJpmUc0VSLm7AUvr5170XEDAAAAAAAAAxGJ2eeOGAx89L2mfWrA46biq2zl16QQr7ffsamYLwf2eL82Pb2eZv+Du/zNkUzmiRp2vTOkBkpQ67FO9aZkRIWQruVs1HKzYiZkqciOWFqFqdc7HFDZvPE82eaBE2IEkVNYsXUyd6vrV1fMsZLyiw4x9jHNnkUxsmj0+Ha1r0X3zDvf9MAAAAAAABwvKnkzPLOxhURuXjAd/rkgGNm5WvV57wESaHlQuiKA5lFLiSOTbUqGYlSXkMkaga1NPUsl73QlQmhTHXU22SA7bgNZ/OXIzEjnfawbIJkyuG/th4jc8zI1ichdEYJsZUSVkNal7wlx6Mhyl0pU7aNhbKWp8MzW2++uD7ff8YAAAAAAABwEmiSM4++Avf6gX089+cGp2Xs40zpD3WsTnakZEFCGlSiRsmaZpt2TkolYYoQCxm7lWLmly54YkaqVJNXB02qfrmaevv1LBfnM2GUkDR6/ktCfvX+3XJ/J32+tnaTWT6hWtmqFjNvunjQchMAAAAAAACOCYvLOxt3vAKpmQ9un1l/YcBxsxK3SxX+PJJSYlQrKXnvi6yBXg4AAAlSSURBVJkrM6pXZUrMlGlXEGqO7+wzz1PXJGYsSqjnwgwhlXDRzxsRUs6YudcXM4oviMgvds4Z1a98ambH6LNOM5enveGi+13NYlXqu4uolmZFKzvLZopRM7lUUCuJmjqekme2LiNmAAAAAAAAYHbKn5hrB1y/v9g+s76fqZly7sx1KeR/qhdNG4wVF32pj1SCxpsh46Y0QvczNjljv6uvhWma2Smp9py6DaeaMdMvZiRKIKXqpN/36qf/BkPSLDZFY2vTpGlGai5NVM+elayyq1t10zjV9yyMt8nqTKFMzGwjZgAAAAAAAGCvHLScKcXMlQP6rk9nW2zU/uCJg5ygyQkGLRLaz6h2p9RMmb72mr7NOzZqxamH/755kJiRai6QvX+nniElX6Ia9dQvqocjaZzHpu0pkjVDN/O9ISVkRnUtl2ox84a1len+SQIAAAAAAADEjA6wJgcpZqRefWpCofpaHDkQDQeeRdDkEjUiXVFjJUNKsAyVNnolIZ0AWVADgMsZM/cMFjNSt559IZmKUftaQZOtXY/ESUkubx5NbnbP0K2WL0kZ0yaQ6jou1qtbnQ7f2PoFxAwAAAAAAADMh/Kn54sHUMuDFjNNa9Nm9aIv7SJ7EDRRUiYjXVKiZpRJ1th2JXdlJuc8ekWm8vFUkM03XhwsZhSf7tZigICx79v62CRRn+QqlCRpRI23OlauLWyhr4aOEGqljKpjKWZev3Z2hjoCAAAAAAAAuJQ/Ra/vc2kOXMwoPtxNeCTSM7MKmpEjCHKtOrlUTHNOb4ukhBU7CfGgEzOXZxIzZXrmCSnku52dTnJGbHuTrZ2bBOqpX6KNKRIoRULYeGkcd7+p54KzutW4lWln62cRMwAAAAAAADBf9lvOvJJipkzPTMSCJ1js/tRxKckgSjB4aQ8rGnLJjmk3LWBs8mPUpGbq+ShLcsceS/l4VqqoGmUFjfsZZ/lxT6J4gsat5ywDgJ1lxheU3FoMEk6Fna3Xrd25xzoCAAAAAAAARIy2z6yXbU1/tw+leUXFjOJ9UXoml64oBwSPQrVFaY/kDJgQS4ZkO5MjeqbZbEuOPo8dAHxKJCyFOzZ/7uK39ljDbnrGuxf1erCgsTXsq58nyGwSaUgtbboo1Tq2OJ4zE06H/9y+dAkxAwAAAAAAAPvCqD7pR+d88sMiZpr0zKab/JB8qiNaxUkSgqDzfmbgb6qlacjmiaIFI2T0jJSlSsxc2Hrd2l7FjNQC733RfWaue6oWp877ifpl5diUsstL40QrW43rGE6Fb29fvPTqvdYQAAAAAAAAIEUlZ7bPrF8VkY/NqUqHRswo3t0+3aug6UtxuGma4B87bcrDSoROi07TglNJmXK7vHVp7dl5FbCePfM1V3IlkkGuoLGCJSWr9BwYr16pfUPFTWp1q2ZfKbdOh29vX7j0U/OqIQAAAAAAAIDHSO37gIg8vccqHUYxU6ZnrkohH0vKGG+fJ2j6WpSyKQ9vkG9mpaaOMDBDge0Q4MXJ81LMyCm5b2tt7al9KOXbpJDvt6+soBnp/XXtRlK3iDkyLJVi8USNHdqbElm9cit0N6+WVSvT7n9v34WYAQAAAAAAgP2nCGGygtHyzsbtIlKmaC7O8M2HUsxoVp6/9qwEuVuaW96VcbyjfN3sC84mk/eKUHSPE4mOSe6zz4dSqOMSUik0g20X5b6tu9c+v7dKpVne2bgiQT4Z16zw66ZeF81xXk1s3VKP8yAj1yqRtBRe2l5Zv32/aggAAAAAAACg6ciZhuWdjTJF8/6BlXqpTN1sn1mf99yaubPy/LVy1aJS0PxQVyQ4gkamkDTeo5j9qdd9aJEg6nmhZELRDrB9aOvcpc/sdx2XdzaekCDvmFrQyABJ4z2mjhuKrZ3E6Zx2APRS+L/ts+s/PIcyAQAAAAAAAAzClTMy/gFeioxHReRBEfEGoj5dreIj8kQ9MPZIsPL8tTUReUqCvKqToJGeFI34kkYkI2rs81mxKRnbStSImZX9FzMNyzsbn5Ug9w8WNJKQNJIQNZKop/c6hye11OvOqlxL4Xvbr13/semrAQAAAAAAADA7STmjqdud1ppd9QDhI4sraFIpGklIGklLmEjWOMdkcZIewRusO6rnsSzK5a3lS/sxYyZJ/W/iK22bmCdoZJikERkgajRD6lgk9hWhHlSstmoAcNjavnN9ddZ6AAAAAAAAAMzKIDlzHKkFzVfaFidxBI30S4VkusPZ10qbBKGwHzDP7fDcxbArhdy5tXxpz8tlz0ItaD4uQX7DFzCJJJKtT0rUiJI1mr5/sp26hclp7LyedkCwfGL7zvVH9q1QAAAAAAAAABlOrJyRyQyaz0mQVX/eTEbS5B7t8dNi58x4qxctyLe2li/dMd+KzEY7o8htA8sMUB4yU8apo3VY7aHW49j6iRFcoxBkQd6yfWb9b/apNAAAAAAAAAC9nGg5I2NBc3u1jHiQ91Q7UpJGnOSHTCcVOvtTIRpPKEiTlmnno/ze1tlLH+q/u4NjeWfjHhH5tAT5yWSKRgaKLVur3P5U+5J9HS3NLd+RQu46SvOSAAAAAAAA4Hhy4uVMw8rz1+6pW3RWXcmSmokyRCAMISkUghYL35BCXr919tKhFAp1m1OZonlPctaM9NRwr+mjZOqoreNNGcnbt8+s/9UMZwcAAAAAAACYO8gZw8rz166IyOMS5Ceqd4aImtRxQ4mEQrCtON+VQt65dfbS3+/Tbc+VeqWvD1WzaGSIhNnjAGVJLZethiiP5AdSyMe3z6z/7sFWAwAAAAAAACAPciZBLWkekSA/Xx0xtJ1JEkNsPfTwlDjxcUsK+bfyGrbOXro+7/s7CGpJUyZpfr1aGUumaGMaWkNJ1FHa1Mz3pJDHt8+sf+RwVQcAAAAAAABgDHKmh3po8IMi8i4ROdvbztT3XoM/K+WmFPKsiPx5Ob/lsLYvzcLyzsaDlaQpa2lFjX0+C902pu+KyJNlAmr7zPoLh7QkAAAAAAAAABXImSmohweXs2nKZbgfEJGfliA/Hp0hV9JusuM7IpWM+ZKI/MNRTchMy/LOxlpdx9eLyDkRuThVS1NXbP2vFPJfIvJVEflHEbnKkF8AAAAAAAA4SiBn5kCdrmmWtdbPNVfV8+vHKRUzD+phwmvqVPc4p32h3kpe3D6zfiJkFgAAAAAAABxjROT/AaGvnBGuLJbJAAAAAElFTkSuQmCC)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "5aISf1B-AGDQ" - }, - "source": [ - "# SuperGradients Semantic Segmentation How to Connect Custom Dataset\n", - "\n", - "In this tutorial we will explore how you can connect your custom Semantic Segmentation dataset to SG.\n", - "\n", - "Since SG trainer is fully compatible with PyTorch data loaders, we will demonstrate how to build one and use it.\n", - "\n", - "The notebook is divided into 5 sections:\n", - "1. Experiment setup\n", - "2. Dataset definition: create a proxy dataset and create a dataloader\n", - "3. Architecture definition: pre-trained PPLiteSeg on Cityscapes \n", - "4. Training setup\n", - "5. Training and Evaluation\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "-1nPOPmc1lGp" - }, - "source": [ - "#Install SG" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "VAssbjJw7Yt1" - }, - "source": [ - "The cell below will install **super_gradients** which will automatically get all its dependencies. Let's import all the installed libraries to make sure they installed succesfully." - ] - }, - { - "cell_type": "code", - "execution_count": 1, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "JKce1SM6voVH", - "outputId": "a6397510-a140-443f-f13c-eec1272cc1a8" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Requirement already satisfied: torch in /usr/local/lib/python3.10/dist-packages (2.1.0+cu118)\n", - "Requirement already satisfied: torchvision in /usr/local/lib/python3.10/dist-packages (0.16.0+cu118)\n", - "Requirement already satisfied: torchaudio in /usr/local/lib/python3.10/dist-packages (2.1.0+cu118)\n", - "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch) (3.13.1)\n", - "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.10/dist-packages (from torch) (4.5.0)\n", - "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch) (1.12)\n", - "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch) (3.2.1)\n", - "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch) (3.1.2)\n", - "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from torch) (2023.6.0)\n", - "Requirement already satisfied: triton==2.1.0 in /usr/local/lib/python3.10/dist-packages (from torch) (2.1.0)\n", - "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from torchvision) (1.23.5)\n", - "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from torchvision) (2.31.0)\n", - "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision) (9.4.0)\n", - "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch) (2.1.3)\n", - "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.3.2)\n", - "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.4)\n", - "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2.0.7)\n", - "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2023.7.22)\n", - "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch) (1.3.0)\n", - " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n", - " Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n", - " Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n", - " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Building wheel for pycocotools (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n", - " Building wheel for termcolor (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Building wheel for treelib (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Building wheel for coverage (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Building wheel for xhtml2pdf (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Building wheel for antlr4-python3-runtime (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Building wheel for stringcase (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - " Building wheel for svglib (setup.py) ... \u001b[?25l\u001b[?25hdone\n", - "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", - "lida 0.0.10 requires fastapi, which is not installed.\n", - "lida 0.0.10 requires kaleido, which is not installed.\n", - "lida 0.0.10 requires python-multipart, which is not installed.\n", - "lida 0.0.10 requires uvicorn, which is not installed.\n", - "tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.23.0 which is incompatible.\u001b[0m\u001b[31m\n", - "\u001b[0m" - ] - } - ], - "source": [ - "! pip install torch torchvision torchaudio\n", - "! pip install -qq super-gradients==3.4.1\n", - "! pip install -qq prettyformatter" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "njthhNJR1pJm" - }, - "source": [ - "# 1. Experiment setup" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "YPym4wvpOcOJ" - }, - "source": [ - "We will first initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", - "\n", - "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", - "\n", - "```\n", - "ckpt_root_dir\n", - "|─── experiment_name_1\n", - "│ ckpt_best.pth # Model checkpoint on best epoch\n", - "│ ckpt_latest.pth # Model checkpoint on last epoch\n", - "│ average_model.pth # Model checkpoint averaged over epochs\n", - "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", - "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", - "└─── experiment_name_2\n", - " ...\n", - "```\n", - "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", - " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`.\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "A2PlnTWpimnH" - }, - "source": [ - "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": { - "id": "_v1N3kXs3wo1" - }, - "outputs": [], - "source": [ - "from super_gradients.training import Trainer, MultiGPUMode\n", - "\n", - "\n", - "CHECKPOINT_DIR = '/home/notebook_ckpts/'\n", - "trainer = Trainer(experiment_name='transfer_learning_semantic_segementation_ppLite', ckpt_root_dir=CHECKPOINT_DIR)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "J9ZaMulSvwhr" - }, - "source": [ - "# 2. Dataset definition\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "_1TXuJKkKzFJ" - }, - "source": [ - "## 2.A Generate Proxy Dataset" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "Y7us7VHRig7M" - }, - "source": [ - "\n", - "A proxy dataset generation is available merely to demonstrate an end-to-end training pipeline in this notebook.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "metadata": { - "id": "wbdVYnIyjgv-" - }, - "outputs": [], - "source": [ - "from PIL import Image\n", - "import os\n", - "import numpy as np\n", - "\n", - "\n", - "# creation of proxy dataset to demonstrate usage\n", - "def generate_proxy_dataset(write_path: str, num_samples: int, num_classes: int, img_size: int = 256):\n", - " # Create training files and text\n", - " os.makedirs(os.path.join(write_path, 'images', 'train'), exist_ok=True)\n", - " os.makedirs(os.path.join(write_path, 'images', 'val'), exist_ok=True)\n", - " os.makedirs(os.path.join(write_path, 'labels', 'train'), exist_ok=True)\n", - " os.makedirs(os.path.join(write_path, 'labels', 'val'), exist_ok=True)\n", - "\n", - " train_fp = open(os.path.join(write_path, 'train.txt'), 'w')\n", - " val_fp = open(os.path.join(write_path, 'val.txt'), 'w')\n", - "\n", - " # Create random samples\n", - " for n in range(num_samples):\n", - " img = np.random.rand(img_size, img_size, 3) * 255\n", - " img = Image.fromarray(img.astype('uint8')).convert('RGB')\n", - "\n", - " lbl = np.random.randint(0, num_classes, size=(img_size, img_size))\n", - " lbl = Image.fromarray(lbl.astype('uint8')).convert('L')\n", - "\n", - " im_string = '%000d.jpg' % n\n", - " lbl_string = '%000d.png' % n\n", - "\n", - " img_train_fn = os.path.join(write_path, 'images', 'train', im_string)\n", - " img_val_fn = img_train_fn.replace(\"train\", \"val\")\n", - " img.save(img_train_fn)\n", - " img.save(img_val_fn)\n", - "\n", - " lbl_train_fn = os.path.join(write_path, 'labels', 'train', lbl_string)\n", - " lbl_val_fn = lbl_train_fn.replace(\"train\", \"val\")\n", - " lbl.save(lbl_train_fn)\n", - " lbl.save(lbl_val_fn)\n", - "\n", - " train_fp.write(f\"{img_train_fn} {lbl_train_fn}\\n\")\n", - " val_fp.write(f\"{img_val_fn} {lbl_val_fn}\\n\")\n", - "\n", - " train_fp.close()\n", - " val_fp.close()" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": { - "id": "DXu4yfuZoiv0", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "dccaf4ba-159f-4a47-d13d-ba4b60eaac80" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Train file `train.txt` content: \n", - "/content/example_data/images/train/0.jpg /content/example_data/labels/train/0.png\n", - "/content/example_data/images/train/1.jpg /content/example_data/labels/train/1.png\n", - "/content/example_data/images/train/2.jpg /content/example_data/labels/train/2.png\n", - "/content/example_data/images/train/3.jpg /content/example_data/labels/train/3.png\n", - "/content/example_data/images/train/4.jpg /content/example_data/labels/train/4.png\n", - "/content/example_data/images/train/5.jpg /content/example_data/labels/train/5.png\n", - "/content/example_data/images/train/6.jpg /content/example_data/labels/train/6.png\n", - "/content/example_data/images/train/7.jpg /content/example_data/labels/train/7.png\n", - "/content/example_data/images/train/8.jpg /content/example_data/labels/train/8.png\n", - "/content/example_data/images/train/9.jpg /content/example_data/labels/train/9.png\n" - ] - } - ], - "source": [ - "num_classes = 10\n", - "generate_proxy_dataset('/content/example_data', num_samples=10, num_classes=num_classes)\n", - "\n", - "print(\"Train file `train.txt` content: \")\n", - "! cat /content/example_data/train.txt" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "MDksFYrIqClt" - }, - "source": [ - "## 2.B Create Torch Dataset" - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "metadata": { - "id": "AGziBKSIqaUu" - }, - "outputs": [], - "source": [ - "import torch\n", - "from torch.utils.data import Dataset\n", - "from torchvision import transforms, utils\n", - "\n", - "\n", - "class CustomDataset(Dataset):\n", - " \"\"\"\n", - " A PyTorch Dataset class to be used in a PyTorch DataLoader to create batches.\n", - " \"\"\"\n", - "\n", - " def __init__(self, data_folder, split):\n", - " \"\"\"\n", - " :param data_folder: folder where data files are stored\n", - " :param split: split, one of 'TRAIN' or 'TEST'\n", - " \"\"\"\n", - " self.data_folder = data_folder\n", - " self.split = split.lower()\n", - " assert self.split in {'train', 'val'}\n", - "\n", - " # Read data files\n", - " with open(os.path.join(data_folder, self.split + '.txt'), 'r') as f:\n", - " data_lines = f.readlines()\n", - " self.samples_fn = [line.strip().split(\" \") for line in data_lines]\n", - "\n", - " self.transforms = transforms.Compose([transforms.ToTensor()])\n", - "\n", - " def __getitem__(self, i):\n", - " # Read image and label\n", - " image = Image.open(self.samples_fn[i][0]).convert('RGB')\n", - " label = Image.open(self.samples_fn[i][1])\n", - "\n", - " image_tensor = self.transforms(image)\n", - " label_tensor = torch.from_numpy(np.array(label)).long()\n", - "\n", - " return image_tensor, label_tensor\n", - "\n", - "\n", - " def __len__(self):\n", - " return len(self.samples_fn)\n" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "metadata": { - "id": "2B0hlas_1Rh-" - }, - "outputs": [], - "source": [ - "train_dataset = CustomDataset(\"/content/example_data\", split=\"train\")\n", - "val_dataset = CustomDataset(\"/content/example_data\", split=\"val\")\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "eIG5tsiuor9E" - }, - "source": [ - "Let's have a look at the first sample:" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "metadata": { - "id": "ZsHqcq1jpN0F", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "e9c1182a-5359-45b6-c0f3-ad430a4fc67d" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "torch.Size([3, 256, 256]) torch.Size([256, 256])\n", - "tensor([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])\n" - ] - } - ], - "source": [ - "img, lbl = train_dataset[0]\n", - "print(img.shape, lbl.shape)\n", - "print(torch.unique(lbl))" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "aWfFrYLzo9j8" - }, - "source": [ - "## 2.C Create Torch Dataloader" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "D3ThxDIopDDB" - }, - "source": [] - }, - { - "cell_type": "code", - "execution_count": 9, - "metadata": { - "id": "XrWjWfjXnw_r" - }, - "outputs": [], - "source": [ - "from torch.utils.data import Dataset, DataLoader\n", - "\n", - "train_dataloader = DataLoader(train_dataset, batch_size=4, shuffle=True, num_workers=2)\n", - "val_dataloader = DataLoader(val_dataset, batch_size=4, shuffle=False, num_workers=2)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "vB1sGPO8qwZJ" - }, - "source": [ - "Lets' have a look at the first batch:\n", - "\n" - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "metadata": { - "id": "O-KuZQ3XBduM", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "489fc05e-f972-464d-c150-360e4f2dbd7e" - }, - "outputs": [ - { - "output_type": "execute_result", - "data": { - "text/plain": [ - "torch.Size([4, 3, 256, 256])" - ] - }, - "metadata": {}, - "execution_count": 10 - } - ], - "source": [ - "next(iter(train_dataloader))[0].shape" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "fFfvyMHU32QF" - }, - "source": [ - "\n", - "# 3. Architecture definition" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "EpqgjQjl4awr" - }, - "source": [ - "SG includes implementations of many different architectures for object detection tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "GNM64JAa4sbF" - }, - "source": [ - "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", - "and extra Auxiliary heads aren't used for training.\n", - "\n", - "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "metadata": { - "id": "YDK4btf04Gbu", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "cc7f1ab1-3a01-49c1-c9a2-6ca434dcc192" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Downloading: \"https://sghub.deci.ai/models/pp_lite_t_seg75_cityscapes.pth\" to /root/.cache/torch/hub/checkpoints/pp_lite_t_seg75_cityscapes.pth\n", - "100%|██████████| 31.4M/31.4M [00:01<00:00, 32.4MB/s]\n", - "[2023-11-12 14:41:45] INFO - checkpoint_utils.py - Successfully loaded pretrained weights for architecture pp_lite_t_seg75\n" - ] - } - ], - "source": [ - "from super_gradients.training import models\n", - "from super_gradients.common.object_names import Models\n", - "\n", - "model = models.get(model_name=Models.PP_LITE_T_SEG75,\n", - " arch_params={\"use_aux_heads\": False},\n", - " num_classes=num_classes,\n", - " pretrained_weights=\"cityscapes\")" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "40UcYJ3u5JyF" - }, - "source": [ - "That being said, SG allows you to use one of SG implemented architectures or your custom architecture, as long as it inherits torch.nn.Module." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "LYPVR-XM4GsZ" - }, - "source": [ - "# 4. Training setup\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "6K_56lDV8azX" - }, - "source": [ - "\n", - "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n" - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": { - "id": "3eRe0hBz4G1n" - }, - "outputs": [], - "source": [ - "from super_gradients.training.metrics.segmentation_metrics import IoU\n", - "from super_gradients.training.utils.callbacks import BinarySegmentationVisualizationCallback, Phase\n", - "\n", - "\n", - "train_params = {\"max_epochs\": 10,\n", - " \"lr_mode\": \"cosine\",\n", - " \"initial_lr\": 0.005,\n", - " \"optimizer\": \"SGD\",\n", - " \"loss\": \"cross_entropy\",\n", - " \"average_best_models\": False,\n", - " \"metric_to_watch\": \"IoU\",\n", - " \"greater_metric_to_watch_is_better\": True,\n", - " \"train_metrics_list\": [IoU(num_classes=10)],\n", - " \"valid_metrics_list\": [IoU(num_classes=10)],\n", - " \"loss_logging_items_names\": [\"loss\"],\n", - " \"phase_callbacks\": [BinarySegmentationVisualizationCallback(phase=Phase.VALIDATION_BATCH_END,\n", - " freq=1,\n", - " last_img_idx_in_batch=4)],\n", - "\n", - " }" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "D3tVVUhy4OqP" - }, - "source": [ - "# 5. Training and evaluation\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "8tKUuxbe9NlQ" - }, - "source": [ - "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", - "\n", - "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders." - ] + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "sh6t_y7KzqBH" + }, + "source": [ + "![SG - Horizontal.png](data:image/png;base64,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)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "5aISf1B-AGDQ" + }, + "source": [ + "# SuperGradients Semantic Segmentation How to Connect Custom Dataset\n", + "\n", + "In this tutorial we will explore how you can connect your custom Semantic Segmentation dataset to SG.\n", + "\n", + "Since SG trainer is fully compatible with PyTorch data loaders, we will demonstrate how to build one and use it.\n", + "\n", + "The notebook is divided into 5 sections:\n", + "1. Experiment setup\n", + "2. Dataset definition: create a proxy dataset and create a dataloader\n", + "3. Architecture definition: pre-trained PPLiteSeg on Cityscapes \n", + "4. Training setup\n", + "5. Training and Evaluation\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-1nPOPmc1lGp" + }, + "source": [ + "#Install SG" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "VAssbjJw7Yt1" + }, + "source": [ + "The cell below will install **super_gradients** which will automatically get all its dependencies. Let's import all the installed libraries to make sure they installed succesfully." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "JKce1SM6voVH", + "outputId": "a6397510-a140-443f-f13c-eec1272cc1a8" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Requirement already satisfied: torch in /usr/local/lib/python3.10/dist-packages (2.1.0+cu118)\n", + "Requirement already satisfied: torchvision in /usr/local/lib/python3.10/dist-packages (0.16.0+cu118)\n", + "Requirement already satisfied: torchaudio in /usr/local/lib/python3.10/dist-packages (2.1.0+cu118)\n", + "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch) (3.13.1)\n", + "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.10/dist-packages (from torch) (4.5.0)\n", + "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch) (1.12)\n", + "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch) (3.2.1)\n", + "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch) (3.1.2)\n", + "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from torch) (2023.6.0)\n", + "Requirement already satisfied: triton==2.1.0 in /usr/local/lib/python3.10/dist-packages (from torch) (2.1.0)\n", + "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from torchvision) (1.23.5)\n", + "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from torchvision) (2.31.0)\n", + "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision) (9.4.0)\n", + "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch) (2.1.3)\n", + "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.3.2)\n", + "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.4)\n", + "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2.0.7)\n", + "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2023.7.22)\n", + "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch) (1.3.0)\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", + " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for pycocotools (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for termcolor (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for treelib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for coverage (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for xhtml2pdf (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for antlr4-python3-runtime (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for stringcase (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + " Building wheel for svglib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", + "lida 0.0.10 requires fastapi, which is not installed.\n", + "lida 0.0.10 requires kaleido, which is not installed.\n", + "lida 0.0.10 requires python-multipart, which is not installed.\n", + "lida 0.0.10 requires uvicorn, which is not installed.\n", + "tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.23.0 which is incompatible.\u001B[0m\u001B[31m\n", + "\u001B[0m" + ] + } + ], + "source": [ + "! pip install -qq super-gradients==3.4.1\n", + "! pip install -qq prettyformatter" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "njthhNJR1pJm" + }, + "source": [ + "# 1. Experiment setup" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "YPym4wvpOcOJ" + }, + "source": [ + "We will first initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", + "\n", + "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", + "\n", + "```\n", + "ckpt_root_dir\n", + "|─── experiment_name_1\n", + "│ ckpt_best.pth # Model checkpoint on best epoch\n", + "│ ckpt_latest.pth # Model checkpoint on last epoch\n", + "│ average_model.pth # Model checkpoint averaged over epochs\n", + "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", + "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", + "└─── experiment_name_2\n", + " ...\n", + "```\n", + "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", + " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`.\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "A2PlnTWpimnH" + }, + "source": [ + "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "id": "_v1N3kXs3wo1" + }, + "outputs": [], + "source": [ + "from super_gradients.training import Trainer, MultiGPUMode\n", + "\n", + "\n", + "CHECKPOINT_DIR = '/home/notebook_ckpts/'\n", + "trainer = Trainer(experiment_name='transfer_learning_semantic_segementation_ppLite', ckpt_root_dir=CHECKPOINT_DIR)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "J9ZaMulSvwhr" + }, + "source": [ + "# 2. Dataset definition\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "_1TXuJKkKzFJ" + }, + "source": [ + "## 2.A Generate Proxy Dataset" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Y7us7VHRig7M" + }, + "source": [ + "\n", + "A proxy dataset generation is available merely to demonstrate an end-to-end training pipeline in this notebook.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "wbdVYnIyjgv-" + }, + "outputs": [], + "source": [ + "from PIL import Image\n", + "import os\n", + "import numpy as np\n", + "\n", + "\n", + "# creation of proxy dataset to demonstrate usage\n", + "def generate_proxy_dataset(write_path: str, num_samples: int, num_classes: int, img_size: int = 256):\n", + " # Create training files and text\n", + " os.makedirs(os.path.join(write_path, 'images', 'train'), exist_ok=True)\n", + " os.makedirs(os.path.join(write_path, 'images', 'val'), exist_ok=True)\n", + " os.makedirs(os.path.join(write_path, 'labels', 'train'), exist_ok=True)\n", + " os.makedirs(os.path.join(write_path, 'labels', 'val'), exist_ok=True)\n", + "\n", + " train_fp = open(os.path.join(write_path, 'train.txt'), 'w')\n", + " val_fp = open(os.path.join(write_path, 'val.txt'), 'w')\n", + "\n", + " # Create random samples\n", + " for n in range(num_samples):\n", + " img = np.random.rand(img_size, img_size, 3) * 255\n", + " img = Image.fromarray(img.astype('uint8')).convert('RGB')\n", + "\n", + " lbl = np.random.randint(0, num_classes, size=(img_size, img_size))\n", + " lbl = Image.fromarray(lbl.astype('uint8')).convert('L')\n", + "\n", + " im_string = '%000d.jpg' % n\n", + " lbl_string = '%000d.png' % n\n", + "\n", + " img_train_fn = os.path.join(write_path, 'images', 'train', im_string)\n", + " img_val_fn = img_train_fn.replace(\"train\", \"val\")\n", + " img.save(img_train_fn)\n", + " img.save(img_val_fn)\n", + "\n", + " lbl_train_fn = os.path.join(write_path, 'labels', 'train', lbl_string)\n", + " lbl_val_fn = lbl_train_fn.replace(\"train\", \"val\")\n", + " lbl.save(lbl_train_fn)\n", + " lbl.save(lbl_val_fn)\n", + "\n", + " train_fp.write(f\"{img_train_fn} {lbl_train_fn}\\n\")\n", + " val_fp.write(f\"{img_val_fn} {lbl_val_fn}\\n\")\n", + "\n", + " train_fp.close()\n", + " val_fp.close()" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "id": "DXu4yfuZoiv0", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "dccaf4ba-159f-4a47-d13d-ba4b60eaac80" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Train file `train.txt` content: \n", + "/content/example_data/images/train/0.jpg /content/example_data/labels/train/0.png\n", + "/content/example_data/images/train/1.jpg /content/example_data/labels/train/1.png\n", + "/content/example_data/images/train/2.jpg /content/example_data/labels/train/2.png\n", + "/content/example_data/images/train/3.jpg /content/example_data/labels/train/3.png\n", + "/content/example_data/images/train/4.jpg /content/example_data/labels/train/4.png\n", + "/content/example_data/images/train/5.jpg /content/example_data/labels/train/5.png\n", + "/content/example_data/images/train/6.jpg /content/example_data/labels/train/6.png\n", + "/content/example_data/images/train/7.jpg /content/example_data/labels/train/7.png\n", + "/content/example_data/images/train/8.jpg /content/example_data/labels/train/8.png\n", + "/content/example_data/images/train/9.jpg /content/example_data/labels/train/9.png\n" + ] + } + ], + "source": [ + "num_classes = 10\n", + "generate_proxy_dataset('/content/example_data', num_samples=10, num_classes=num_classes)\n", + "\n", + "print(\"Train file `train.txt` content: \")\n", + "! cat /content/example_data/train.txt" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "MDksFYrIqClt" + }, + "source": [ + "## 2.B Create Torch Dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "id": "AGziBKSIqaUu" + }, + "outputs": [], + "source": [ + "import torch\n", + "from torch.utils.data import Dataset\n", + "from torchvision import transforms, utils\n", + "\n", + "\n", + "class CustomDataset(Dataset):\n", + " \"\"\"\n", + " A PyTorch Dataset class to be used in a PyTorch DataLoader to create batches.\n", + " \"\"\"\n", + "\n", + " def __init__(self, data_folder, split):\n", + " \"\"\"\n", + " :param data_folder: folder where data files are stored\n", + " :param split: split, one of 'TRAIN' or 'TEST'\n", + " \"\"\"\n", + " self.data_folder = data_folder\n", + " self.split = split.lower()\n", + " assert self.split in {'train', 'val'}\n", + "\n", + " # Read data files\n", + " with open(os.path.join(data_folder, self.split + '.txt'), 'r') as f:\n", + " data_lines = f.readlines()\n", + " self.samples_fn = [line.strip().split(\" \") for line in data_lines]\n", + "\n", + " self.transforms = transforms.Compose([transforms.ToTensor()])\n", + "\n", + " def __getitem__(self, i):\n", + " # Read image and label\n", + " image = Image.open(self.samples_fn[i][0]).convert('RGB')\n", + " label = Image.open(self.samples_fn[i][1])\n", + "\n", + " image_tensor = self.transforms(image)\n", + " label_tensor = torch.from_numpy(np.array(label)).long()\n", + "\n", + " return image_tensor, label_tensor\n", + "\n", + "\n", + " def __len__(self):\n", + " return len(self.samples_fn)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "id": "2B0hlas_1Rh-" + }, + "outputs": [], + "source": [ + "train_dataset = CustomDataset(\"/content/example_data\", split=\"train\")\n", + "val_dataset = CustomDataset(\"/content/example_data\", split=\"val\")\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "eIG5tsiuor9E" + }, + "source": [ + "Let's have a look at the first sample:" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "id": "ZsHqcq1jpN0F", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "e9c1182a-5359-45b6-c0f3-ad430a4fc67d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "torch.Size([3, 256, 256]) torch.Size([256, 256])\n", + "tensor([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])\n" + ] + } + ], + "source": [ + "img, lbl = train_dataset[0]\n", + "print(img.shape, lbl.shape)\n", + "print(torch.unique(lbl))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "aWfFrYLzo9j8" + }, + "source": [ + "## 2.C Create Torch Dataloader" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "D3ThxDIopDDB" + }, + "source": [] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": { + "id": "XrWjWfjXnw_r" + }, + "outputs": [], + "source": [ + "from torch.utils.data import Dataset, DataLoader\n", + "\n", + "train_dataloader = DataLoader(train_dataset, batch_size=4, shuffle=True, num_workers=2)\n", + "val_dataloader = DataLoader(val_dataset, batch_size=4, shuffle=False, num_workers=2)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vB1sGPO8qwZJ" + }, + "source": [ + "Lets' have a look at the first batch:\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "id": "O-KuZQ3XBduM", + "colab": { + "base_uri": "https://localhost:8080/" }, + "outputId": "489fc05e-f972-464d-c150-360e4f2dbd7e" + }, + "outputs": [ { - "cell_type": "code", - "execution_count": 13, - "metadata": { - "id": "-Ojnc1bk9L3s", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "b36b8b9b-b554-444e-d440-02bf623c3efa" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-12 14:41:51] WARNING - sg_trainer.py - Train dataset size % batch_size != 0 and drop_last=False, this might result in smaller last batch.\n", - "[2023-11-12 14:41:58] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231112_144158_892860`\n", - "[2023-11-12 14:41:58] INFO - sg_trainer.py - Checkpoints directory: /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860\n", - "/usr/local/lib/python3.10/dist-packages/super_gradients/common/registry/registry.py:72: DeprecationWarning: Object name `cross_entropy` is now deprecated. Please replace it with `CrossEntropyLoss`.\n", - " warnings.warn(f\"Object name `{name}` is now deprecated. Please replace it with `{deprecated_names[name]}`.\", DeprecationWarning)\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "The console stream is now moved to /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860/console_Nov12_14_41_58.txt\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-12 14:41:59] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", - " - Mode: Single GPU\n", - " - Number of GPUs: 1 (1 available on the machine)\n", - " - Full dataset size: 10 (len(train_set))\n", - " - Batch size per GPU: 4 (batch_size)\n", - " - Batch Accumulate: 1 (batch_accumulate)\n", - " - Total batch size: 4 (num_gpus * batch_size)\n", - " - Effective Batch size: 4 (num_gpus * batch_size * batch_accumulate)\n", - " - Iterations per epoch: 3 (len(train_loader))\n", - " - Gradient updates per epoch: 3 (len(train_loader) / batch_accumulate)\n", - "\n", - "[2023-11-12 14:41:59] INFO - sg_trainer.py - Started training for 10 epochs (0/9)\n", - "\n", - "Train epoch 0: 100%|██████████| 3/3 [00:08<00:00, 2.91s/it, CrossEntropyLoss=3.49, IoU=0.0319, gpu_mem=0.686]\n", - "Validating: 100%|██████████| 3/3 [00:00<00:00, 3.83it/s]\n", - "[2023-11-12 14:42:09] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860/ckpt_best.pth\n", - "[2023-11-12 14:42:09] INFO - sg_trainer.py - Best checkpoint overriden: validation IoU: 0.013753225095570087\n", - "Train epoch 1: 0%| | 0/3 [00:00 Date: Mon, 13 Nov 2023 11:13:28 +0200 Subject: [PATCH 3/5] updated readme links, removed infery refs --- README.md | 6 +++--- notebooks/quickstart_segmentation.ipynb | 3 --- 2 files changed, 3 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index 14a77653b8..7deab15cbe 100644 --- a/README.md +++ b/README.md @@ -211,9 +211,9 @@ model = models.get("model-name", pretrained_weights="pretrained-model-name") ### Semantic Segmentation -* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/quickstart_segmentation.ipynb) [Segmentation Quick Start](https://bit.ly/3qKx9m8) -* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/transfer_learning_semantic_segmentation.ipynb) [Segmentation Transfer Learning](https://bit.ly/3qKx9m8) -* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/segmentation_connect_custom_dataset.ipynb) [How to Connect Custom Dataset](https://bit.ly/3qKx9m8) +* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/quickstart_segmentation.ipynb) [Segmentation Quick Start](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/quickstart_segmentation.ipynb)) +* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/transfer_learning_semantic_segmentation.ipynb) [Segmentation Transfer Learning](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/transfer_learning_semantic_segmentation.ipynb) +* [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/segmentation_connect_custom_dataset.ipynb) [How to Connect Custom Dataset](https://colab.research.google.com/github/Deci-AI/super-gradients/blob/master/notebooks/segmentation_connect_custom_dataset.ipynb) ### Pose Estimation diff --git a/notebooks/quickstart_segmentation.ipynb b/notebooks/quickstart_segmentation.ipynb index 13798e9e3a..cea757196f 100644 --- a/notebooks/quickstart_segmentation.ipynb +++ b/notebooks/quickstart_segmentation.ipynb @@ -1502,8 +1502,6 @@ "id": "br7n55Szm4Nq" }, "source": [ - "SG is a production ready library. All the models implemented in SG can be compiled to ONNX and TensorRT. Deci also offers the [Infery](https://docs.deci.ai/docs/installing-infery) library that allows to do inference on models saved in various frameworks with the same API regardless of a framework.\n", - "\n", "Let's compile our model to ONNX." ] }, @@ -1536,7 +1534,6 @@ ], "source": [ "! pip install -qq onnx-simplifier\n", - "! pip install -qq infery\n", "! pip install -qq torch==1.12" ] }, From 4fb10d601d966831c2c7fe91eae7a8b1ee0a8a5d Mon Sep 17 00:00:00 2001 From: shayaharon Date: Mon, 13 Nov 2023 13:58:36 +0200 Subject: [PATCH 4/5] updated qs notebook --- notebooks/quickstart_segmentation.ipynb | 2762 ++++++++--------- ...nsfer_learning_semantic_segmentation.ipynb | 2 - 2 files changed, 1259 insertions(+), 1505 deletions(-) diff --git a/notebooks/quickstart_segmentation.ipynb b/notebooks/quickstart_segmentation.ipynb index cea757196f..003c6b8c7a 100644 --- a/notebooks/quickstart_segmentation.ipynb +++ b/notebooks/quickstart_segmentation.ipynb @@ -1,1586 +1,1342 @@ { - "cells": [ - { - "cell_type": "markdown", - "metadata": { - "id": "HY_HuQbxn7X0" - }, - "source": [ - "![SG - Horizontal.png](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAABGcAAAD5CAYAAACK9FhIAAAACXBIWXMAAAsSAAALEgHS3X78AAAgAElEQVR4nOydCZgcV3W2z62ZkUbryPIu20gywWhFEsZJzGLJkBCWEMsYsFk1hjiGgLEclgQIWCIsgQQ8Zov/sFgibGazxB8ghhiPCJD8EViSZUuyvEnYlm3ZljQaLaNZqv7ndp/quX373urqrrrV1d3f+zw13VNVXcut6uq6X33nHBEEAQHQZMwgoqW8yXN4MLGViA7x+H4c5DLUNlTf6+zhgbT2BAAAAAAAAACQEp1oSJBzpGiw4ukDY2+Y1O2dN3my6JGbe/x4EEyaJES1TR8bIxoZCfzubuGNjdHo0WP+U8PDwZZTTu74Lgs2e3K+/2kg23Dp0wfG/rqrU8ydPt07RS5zaKjYLnGWf/SYPzZlstch3x87FgwcH/J3T+gSP502zbuVRRsAAAAAAAAAAHUC5wzIIysHBvzVU6aKF1FAorOzughTL1JoGBkJftbT432ixUSGQhtOnCgu7OoSEzo63KxkdDQIhBD+oYGxO0+e2SHbcIObNQEAAAAAAABA6wJxBuSFOQMD/qenTBGXCSE8V2KCDd8vOEkOnBgOPnHSDO+rTRq+M+fAgbGPTJ7svdGlIGNDtmEQ0Njx4/7Ppk71/rpNXEkAAAAAAAAAkBiIM6DRzHnqqbFvnnJKx/PzcCRkGJQ0hAyd8L8wZbJ3XcM3KB5zjhzxvzR1qvfyvGyQ71Nw8NDYb0+e2fE6iDQAAAAAAAAAEA3EGdAwDhwY+9r06V6vy7ClepEizehYMDRxgnh9nkN1jh7zb5g4QVybxzYMGRjwv9fT4/0VkgkDAAAAAAAAgBmIM6ARLD182P/11KneZC9WOtrGIUN1jh7zfzttqvenORMXmqYNJSeGC0LXy1E1CwAAAAAAAAAqaYJuHWglDh7y3zM6GmyZPr05RAW5jdOmes87MRw8GlFuOlNkGxJR07ShZOIE0e37dId0S+VgcwAAAAAAAAAgV8A5AzLj0X2jPz7zjM5XNIugoDM2Rn5HB72kke6PJ54Y+59TT+34o2ZtQ+lEGhz0N/X0eCsR5gQAAAAAAAAAReCcAZlw3/0ju5pZmJF0dBS+L3ewcyVzdu4afqCZhRkKnUjTvOXHjgUPENGMHGwSAAAAAAAAADQcOGeAc7bddeLRxYsmzmpmUcHAxVk6aO7ceuKJpc+ZeForteHx48H+SZPE6TnYFAAAAAAAAABoKHDOAKf8y5cPbW9BYUaGON2eVQ6a7/1w8NFWE2YkEyeK0wYGfCQIBgAAAAAAALQ9EGeAM6665vH/eMdVMxa1mqhAHOJ07FjQ7zo050Nrn9p12cppLSduEYc4TZ3qLR8c9NfmYHMAAAAAAAAAoGEgrAk44cV/8fC3N37rrCumTW1t/e+pp8d+esrJHa9wseyLXv7w/97+o7Mv6OoSLhafN5YR0dZ22FEAAAAAAAAA0IFzBqTOc16455/efdVJLS/MSGae1PFyIlqR9nKf84I9v7zxH09rF2GG2IUEAAAAAAAAAG0JxBmQKs950Z4Pzpsz8b2vesXUtmhYGZpzYjj4cZrLfM6L9tx+wXMmvWjJcyamudhc090teo4e829omx0GAAAAAAAAAAWENYHUWLz8oevECfHZH37zLPqDZ3a1TcOOjUnnh//RadO865Mua/Hyh24TJ8RLf/WzZ9D06e2lnY6Nkd/RQScT0aEcbA4AAAAAAAAAZAacMyAVFl/80HViRHz2gqWT6Nxz20eYoWJyYJo82ftg0uUsfvFD3xEj4qWXv3o6TZvWfl9NmWT5wIGxz+ZgUwAAAAAAAAAgUyDOgMQsfnFRmKExog+8byZ57ZEmRafz4CH/PfV+ePFLHvq6GBaX05iga94+g0R7tiH19HSscl0BCwAAAAAAAADyBsQZkIjFf/LQdWJUfJZ8olmnd9HcOe3lmgkpumfER+v57OI/feifxbB4sxS3XnzRZJoypa2/lt7goH9dDrYDAAAAAAAAADID4gyom8UvfejqgjAzRiTFmbdf1VMQKdqVzg4xmYiW1rL7i1/60HU0It4j2498QW/rnd7WbSj3fcJE8d4cbAoAAAAAAAAAZAbEGVAXi1/24NU0SjcVhJmgKCy85OLJbd2YUlgYGPD74s6/+GUPXkej9FnB4pYsPb54YftUaLLR1Vm7yAUAAAAAAAAAzQzEGVAzi18hhRlxkxgTVHR8EM171gTqntiYRCmyWpIcqo3LgqlTvRfGWc3iVzwoHTOfLbVhUAxpGh1tXPW04eHydcv208dlgSxPPjDgJ06wDAAAAAAAAADNQieOFKiFxa988M00Im6iUSqJChQIesMV06irKztxRlaAP3LUp9vvOEb9vzxO+/aN0q7dw6Xp886bQLPO7KSX/slketELJ9HUKV4mSXaDIPCIhHR9bLXNs/jPH7xaCjOhsBU6j173mqnU2ZldG0rxxfcD+s1/D9F/3nGM9j02Spt/N1SaLtvvrFmd9IILJ9ErXjaZTju1syCcZEH3JPHKzBoCAAAAAAAAABqMCILGPakHzcXiVz34ZhoWX5fCjBgVVAhpKgyC/vPHZ9Npp2aTLGVkJKBP33CQbvn+YPSMyql9+Wum0fv/5qRMBKQDB8Zunjmz462maYv/4sGraVjcRMNEBdeM0oa/+/UzMhO4RkYD+pf/M0BfWT8Q7wMB0QXP7aYPvH8mnTu3y7lI4/sUeB6dS0R73K4JAAAAAAAAABoPwppALBZf8uCbaVR8PUz+W3R7FF9lrpSshJnHnxilFS9/hG754WEiL7APgs9u1jqkkLPizx4pfN41nV3i5aZVLF4pHTNUdB0F5a4ZKXxkhWyDV122j77yb4eqt2GoFQmizVuG6NVX7KMf/fiIFE+c4nmFNa/IrFEAAAAAAAAAoIFAnAFVWXzpg6+mUfp6mPxX+JwnhYohTTKESLpZXDN4xKfL3rSPBo+NFc/cyEFuaCg0FMWFwUGfLnv9Y87zqEyf5p2hj1t86QNvLggznGOm0IbBuLtnyeKJmbhmwjbc9+TIePt18FDRnkobKiLNRz76dCYCzdMHxv7a7RoAAAAAAAAAIB9AnAGRLL7sASnM/ICU5L8FgvFB5iRxLSz4AdHqv9tPg8f8cVeMOgiqcHqMjx93gUiB5h3X7C8szxUjI4VYwVK1ocWXPcCuI1Hefko7yrw4rpFiSqENj/vmttPHeYY2pHGBZv+Tbl1Ik7q985w3CgAAAAAAAADkAIgzwMri10phRvxADWUSgSgTZiTLlrov//z734/Q5m1DES4PTVRQRRpNXNh851Bhea7o6iqkHi6IM4tf98DVpXCwYDyMSW/DZ5ztPjf3trtPFNvQJG55BnFLaMKNJtB86CNPOxW5Jk8WPe6WDgAAAAAAAAD5AeIMMLL48gdeSiOhMCOUykzliXbl+9NPc5tvRjo+vvrtgUqHTIV4QHYXjSbQfHXdYadhOQcOjF206PIHri5UttLaUKjtGBQ36OST3bahrMz0ha8epKCDioNXPhidM2Rq03KRa/9+d+4ZmRRYdSABAAAAAAAAQKsCcQZUsPj1D7yURsVtZaWeyfRa7KXLkssukZWBbv/1UQq8wDhYBRtdoFFef7HpmNOKQ0dPBH8qRummsgTKJnFLlv1+1oSCeOKSY8d9+t/tx63Jf0vtqoc4GZ00QWncd79/xNlWB0FhbTPctgwAAAAAAAAANB6IM6CMxW+Qjhm6rVBRyGcBJiKkKQse+v1IMdeMJRSnKCpoIo3VQROUcs/s2esutImC4GwaDfP0iAqnTHEernY1zf3X8K4dJ6qLV4LGhZqoeWj89Ve/Oe5MWOoomolQsQkAAAAAAADQ8kCcASUWv+mB59Mo3VZIXBtYHB9aSNMF53fTseNulZqd90UIC4oIUybSUHX3zP33uxNnZkzvKG8v23t2HnU4rkT+698di65uZRCxAps4o7hndu0edrrtjz8xutDd0gEAAAAAAAAgH0CcAQUWveV+Kcz8uqwqU+iUoUpRRmXiBLeVmu6W4owpCbAlL0qFi4YMr0R03wPuxJlpU73K5L+k5Juh8VfXYWFSPNv10LA5mXJFVSb+kE2gIe09ET38iLu8M11dYrazhQMAAAAAAABAToA4A2jRqvtni1Hx64r8KGQIXzKMd+36KAgLtqpMtrwoFDGO+e3vhtxueFSuGeX9WbPcijOTJwl6VCbureI8soaBkVncCuc9dsxdZuXuiWKms4UDAAAAAAAAQE6AONPmLHrr/TPEmNhTqihkCmMK880YePZ5E5w24OARv7xKUJSoQAb3DBmEGWVZTkOybCFNDWDfkyPF5L9CSQSsO48sDhpreBNz7253DqQxn6Y0psUAAAAAAAAAIDsgzrQ7o3RwPPmv6vYwJP81CAuF8B3XKMlqSwIDRTg+SBFoqHycLtSMjbpRS2SS3FmndWluGW3l/P8557h1zhQw5pgxiDRGwUtzz2RLd0PWCgAAAAAAAAAZAnGmjVm06v5AjIry3ChkEWNi5J5xhsnVIYJKN02EUJO1sCBLV591OosuVdpMhh05R3cc6SJNVGiYQVcCAAAAAAAAAJAeGTyyB3lk0VvvD4SvJP8lQ46UPKGLLaVtDMonBtp8ooH7EtWOWW+Tp2yPUWjhCep0oU0GAAAAAAAAAOAEiDNtyKK33R8UQpnGIhLWksEt04AO+uRJXnXnS6CIMPprHlArNTUKoYlaRpEmKFpkbG3YAAcSAAAAAAAAALQDCGtqMxZddX8gRRlRlvxXC2myldBWySjORVaCKuS1MZZ6Dt8bctDkhbwIRFUSKEeVHC/+b8jfAwAAAAAAAAAgFSDOtBGLrr4vqCiXnQN3TDWec97E2oQFGv8/aGwy23EakatHJcwro+fvIV20UTYOThkAAAAAAAAAyASENbUJi95xX0CjxRwzQq/EZMuNEiUiiGLcy727h5034LzZE80iQVl+FCUkJy8hTXkSvArtFKj/jLedOjowjAcAAAAAAAAA4BQ4Z9qARe+UjpmwKpOlZDbZRRphC2ESRIODvnlaisw7d4K5MhMld3ZMm+bmK5BJifFa0MPA4oYzAQAAAAAAAABwDpwzLc6id903nvzXt4Qy6e9N/9tggcaVyCHpmeYVxI7BwaDc1aE6Pepwy2z+3RA954K96W9wB1HQ4160qglT5SVVkDElAAYAAAAAAAAAkAlwzrQwi969m3PMjIcxCRJ2IaaOjrkUOFwKM5LOTkEv/sMplflQkqC7cNIaJhIFM8fyFxpUsX+B2S2j5p1BzhkAAAAAAAAAyASIMy3Kout2F0OZ9OS/1fLMUD7dE5f/SU+5qKALB7rgEEU4r5fyMDGg4KSx7BsnBoGxzTSBBkIMAAAAAAAAADQEiDMtyKK/2R3QSCjMRAg0lI4Qs++xUeeNuPi8ieY8LvWICnrVojSGTqLgJD+/YUFCE2h0YQsAAAAAAAAAQMOAONNiLHrv7srkv2QJaQoT/ZrGxYFn3bNnJJNG/OvLZsZzekSNT1uU8TiU6WRDjpm8CTW2MCYAAAAAAAAAAA0F4kwLsej9u8uT/1KMEKYYBKYcJEqn/rd3nsikES+5eBqvu86dcSLMBBTM8PMvcoQ6HKoyAQAAAAAAAEDugDjTIiz8u3uLwoyS/LesTDZZwpmiKjbFZNv2EzQy4t4mMrnboyv+bHp9H3aR/FdWZWoGYYYikvyqOXwAAAAAAAAAADQEiDMtwMIP3huI0RqS/6alo3DHXlZs6upy37vv6CB631tOoWlTajhtpT7lBekPEzj5b96qMkVRkUA5f5sIAAAAAAAAAO0IxJkmZ+GHVGFGVOSZMb4mxdCx37M3m7wzEyYI+uvXziz+U01cCLcz7VCmCUQkc8x0ZLHHKaG0VS1phQAAAAAAAAAAuAfiTBOz8MP3BiIsl62LMlpIkyglHUmwvya3Bf9/28+PZdaQr395D82bOyF6Jj1xcNqhTM3glNGpN1dPFQ4PGpIhp0SHR087WzgAAAAAAAAA5AQRBM3YywQL19wb0DCRkCWzwwTAoYNmjHPPhKKNjL4JlGlB+KoJO4XpSqWnsM8t06r42vgwVIrXM++8CfTNdWdkEt4k2f/0KL3kqt+P709puzT3UJrIctnT/YKkKcbUdhOV26G2P88n9HHcpt/7xpn07POqiE0JWfSW+8vWKXxt+8u2S8ldpJwzpe3XzxV3l5CL79o8u99pwwAAAAAAAABADoBzpglZuHaXTyNSIFDcMKZYFb3THKcTrSaIjVuqWhDt2j1Mx49nJ/SdPKOTPvXu08wTvaLDJdVhAlEwc6z4HnpmOS6SLQu6DsIMAAAAAAAAoF2AONNkLPyHXT6NCjHuXLEk/nWZb0Z9r4TKfPu7g5k1pkwO/LKLptIVL9OqNxXEpZSHDlkue8wuVjUD1QS3epBt46U+DJMXXH/X/87ua+LWBgAAAAAAAICagDjTRCz8OAszapiRKxFGRa5KzVeidvKV97f+6AiNjWXXnp5H9P6rTqZ5505wk/jX41CmmU2W/LcaaQg0woE7qYOGyaO33PU/cz7qtgEAAAAAAAAAIF9AnGkSFn5yZ0mYqSyTrQs1Wu+7nvAmsnTiI0Kd9j02SrvvH860Qbs6BX3tY7No2lTPjTBzql/+LUFIUxE3oUwfv+s3c27Jw+4BAAAAAAAAQJZAnGkCFn5KEWZ8SzhTSBwHTRznhKkqk8Eto4c2/fNnD2bqnpFIYeaHN55dFGjSrMo0010VoqbFlUPJo0/d9Ss4ZgAAAAAAAADtCcSZnLPw0yzMKJVxROiMMYkycbG5YqJEGdP8VC5qbL5ziAaPZC9qnHFaJ33vhrNp2rQUHDSdSP5romDQ8pwMF9/1yzl/l789BgAAAAAAAIBsgDiTYxb+886xslCmUPOwhSmZXDPVxIW4VZnUvDNGEWd8RV+86RA1okL7mad20idkBaekwswpY62VYyYNCo4ZJwmAP7W9fw6qMgEAAAAAAADaGogzOWXhDTvHaEx4xVAmYcgz49DVEZbTNk5TXoVhHBHd8v1BGjicvXtGJgi+6HmT6fN/e0Z9QkFnUHTMhPsD10wRN/ll5PCp7b+YC8cMAAAAAAAAoO2BOJNDFvTtGKMR4dFYpRgjqiX/TYLuiLHlmbFOHxd13vP+JzPPPUOqQPP+M1GVKQ3cVGWSbX7F9tshzAAAAAAAAAAAQZzJHws+t2NMjIaOGYNLptbkv3GICm2i8mlloU0RrzL3zEN7RhrSvgWB5vzJ9LUPz4qXJNjjHDP4NpTjMPnv9p/PRVUmAAAAAAAAAGDQHc0RC74QCjOi3BkTtypTLSKN5nQpH294T9UFGX2Z77puf8MaVwo05y+YRLd8/OzoMtu2HDPtHtLkMpTpP+CYAQAAAAAAAAAViDM5YcGXFMeMXipbpZ6qTKbBNF/Z/0HlNGFJDGwUaYj2PTZK3/j2YEPCm4gFmnPO6KIffvIcs0DTSeSXqjKlGB7W7LhL/vvF7T+FMAMAAAAAAAAAOhBncsCCm9RQJlEZzhRVQjsNbHlmyCK8VHtV3DOf/uwBevpAg9QZ5oxTO+mHHz+Hpk4RBWGpMHQE48IMGMdlKNOPz30XWhoAAAAAAAAAKulEmzSWBf96z6gY8YrJf3W3jEmMsQkz9Qg2Qvuc+n/4XsSYT61upL8PiN517X767rfObGg7S4Hmjs/PpTd97BHa9cgJCk7ialLZF5XKL4ooN22KR/POnZjKpp59eue3P/ruU/+DiFbU8fFDRLQ15y0HAAAAAAAAAImAONNAFnz5nlEaER3jooylZLaJNJIA6wJLhTATFLfJkDBYuk8EiXJBpkzQGZ+w675h+twXD9E73z6DOhroVOmeIOjbHzmbrvjcXtr12PC4MNPu+WWo0gklhZkvf/RMOnY8mXolpFlJ0OuOHvNfV+tnOzwhhkeC/dOneY1V9gAAAAAAAADAMRBnGsSCr90zSqOiQ0SEMZVIuypTmQsmGHfomAQaivG/sLhmwjcB0VfWDdCFf9xNF5zf3dB27+oU9J13z6arvvIobb7vWEO3JTfo4ptPJA6LgpBWyNWTjLAYd10MjwSNPWEAAAAAAAAAIAOQc6YBLLj5npFyx0wVUUbHNN6WzzZM4qt/KCq3TKy8MpwYWP9cxHJWv+9JevyJ0Ya3vxRo1r39bLrij2c0fFtyQeE48jBGJB7vIBpBgmQAAAAAAAAAyAqIMxmzYP09wzQqOgvuhNA1Q4ZqQXq+mTiuGVsJbH0ekyhjel8KTzKIL2Sp3FSxHUFp2uCgT+/+mydpZDQfcUQfuuw0uuLCNhZoDMl/xZNeQaABAAAAAAAAAJAdEGcyZMG/3X2CRkVXRfLfODlmkmISbEoiSmAWVmyvttLc1nHjAs2u3cP09nfuz8XxkKW2P/Sa0+iK57ehQGMUZjogzAAAAAAAAABAA4A4kxELvnn3EI2ICYXOb1gW2ybM6AJNGrlmbP/HKpttcMco02OFNykCzeY7h+jjnzpAfg4qJUmB5oOvPY0+/ZYzGr8xGRJ4yiBdXI91EJ1om90HAAAAAAAAgFwBcSYDFnz77uM0IiaSKfkvaSFNaSX/pUoxpZB3plpuGZsTRv/flH+myjyqQ+eW7w/SJ/8pPwLNy543jb5w1azGb4xrFJEsTOTsPQXHDAAAAAAAAAA0EogzjlnwnbuP0ojoLokyFBHCZBJmahVpNBFkfJz2v/5eFVeMr0H5Z2IINnEEmtv7j1GQgxQ0niBavmQKfeGvWligCUOZOoLiIMuhH/CIRnKwbQAAAAAAoAzP616hDmgdAFobiDMOWfDduw/TqJhcFGbGXTNCd8rUI8CY/q9WYCfKLWNbvjAIKybhJSq8KWLce/72Sfrlfx3PhYNGhALN1S0q0KhtL6sy7ZehTNVOGgAAAAAAkBWe1z3D87r7PK/7EBHdoQ6e1x14Xvc6z+uegwMCQOsBccYR83+w/UkaFdMKgowfIwGwSaCJEm2qCTKx3DLavFHii+3zpvCmGhIES655z376r1/lSKBZOoW+8I4WE2j05L9PdxA1vqo5AAAAAABg2B2zh4iuJaIeS7usIqKHPK97NdoNgNYC4owD5v9w+8NiWJwyXpVJVIowSUKXojC5YBRhJCAtfEmdL+JzZQJOxbjyITCJPOryPLtAk4cQJynQXLRsCn34jac2fmPSQq/KhFAmAAAAAIDcwMLMHRGijM4Nnte9BkcQgNYB4kzKzN+w/WExKs4mSzUmoQo1OkmEiTjhSfp8RgdNYJ7H9FmTo0Z10OjzxXDQyBCnvOSgee3yHrpiRdzfx5zjFYXCYihTa+wSAAAAAEArIEOZiGhDHbtyPXLRANA6dOJYpsf8H23fI0bE2eOOGUsIE1kqNdVDQeDQBB/BK5DLFsq6+L10zwghysaVRBN1XmvIlLJs0j6rizX6uAqUGYKiQPOpj51Cf/anUwpVlBpJocz2m04tbNd3fjHQ2I1JShCQeKqTaLi5dwMkw/O6lxKRvImTr3P4tZoCuYlf+4loqxx8f2gPDgUAAACQGqtrcMzo9PHvOQCgyYE4kxLz/337g2JEzNaT/1aIL2m5ZgSVCy/qeE2MqRBPqr2SJr6oyyPDPOouqlqLrAZEomaB5m///im67/4Retc7ZuRCoPn7VafS1vuHaNee5rWcoCpT++J53SuJKBzqufFbrr3KZW5jsWad7w9tbfc2BgAAABKSJH/MEvnwBb/HADQ/EGdSYP5P7npQjHhzy5L/kiG/TFTeGRNRYURxRBnb+6hXUpZvcsao/+sijv5ZznFTj0DzlXUDNHjEpw+8b2bDBRrJtz9yNv35+39Pjz7RpFl0UZWprWB7dC/f7M12sO9LeLiWhZo+3x9a1+7tDkA7wdVi5rAbj5RXG7LjeIhf96AjCZoJz+vur3NzV1c71zksKWkc/Ur+bgEAmhiIMwmZ/9O7HqQRMbci+a8tnKkeTA4ZFatbRgttUsSdIODQJlNIUqCELlENoopldCyBRo73ywWaW74/WHj7ob+dmdbhqpuuLkHrP3gWXfqBh2nwaA7KSgFgwfO6pSizxpEoY0KKNDdzUsI1EGkAaE1YjFnBncB6OpPL1X88r5s4bFJ2ejdArAE5Z3mdmzcDBxYAEBeIMwmY/7O77qdRMVf4FkHGVKWJLP/Xih7WJAyOGKrBLWMQcIxqiklgqeqKiSHQBJy0NgiK77lNpUDz4IMjdNMXT6Ouzsa6P844tZM++fbT6V2feayh2wGACSWZYJwbyDAs6RC/2pij5KZZWkXwmc0ijXTr9KKjBUDzo7jwelmITZvlPMikpnv5GtaHvFYAAADaEYgzdTL/P+/aTSPimcbkv2RO/iuSJv8lg4tGdcjY5jN+RnHPmPLSkPZ5sog6ZEgQbCGWQFMqtx2UxK3Ndw7RG9/yOH3z62c0XKC56PzJdMWfTKdbfna4odsBgAon+u2v8iR7I3d85BPqQ/U0IHfUVipPz03rkx24LZ7Xvdb3h1DiE4AmhF0y8vu7KsOtlwLvtRwuuYmdePWGkgAAAIgBh9XN8P2heqqFgZRBKe06mH/7XbtoWDxLCjNCT/5rc8iYKjXVSKk8dUhUmevSOKVctaGcdUCBcfz4YCmJHXc+bQjC7bHNS9oy+f2u+4bpjasepxMnGltnW5bY/rtVp9CsU6FrgnxQRZiRZcbWEtFc3x9aKUOO6hVmJPKzvIxedtRcSUR7LbPLJ+H9LOgAAJoAKcp4XrcMTXwoY2FGR7pp7uBrCMoEg1YnDacp3KqgKnyNX+l53X18fZUdqzsSJqQGKQJxpkbm9991N42IZ1uT/5LmoEmKXg2JIspcmwQaivhfF0OoRuEl7nxpCTS7h+nFL3+EHm9wUt7OTkFf+/BZDd0GAKi6MCOdMrJ6wxoXIQKKUBMl0sgOFgQaAJoAzhvVaFFGJxRpNrCbB4CWgx+abEywXwNwPQAT8j5RhptL0d3zuvfwNf5WdinWm0cJOEXrj7cAACAASURBVATiTA3M37TtbhqmhaEoUwrRsSUANgk0NtHGJpbY5qn4jCKMqNP1aSaxRHfP2ISXqkKKe4FmcNCny17/WMMFmjNP7aS/XIn+JmgcSo4ZkzBzHTtlMsnboIg0aw2TDyVx6wAA3MI37/JacX2Om/oS6QzgnFYAtCJJwoD7cEYAGr+er1FcMVuI6AYW3bMqFAESAHEmJvP/a9udNCIWFhLVhmFMZblmtJCmNPLLqNgWV9X1YpjPNK/+uWrLiRyyEWge3TdKfoMKJ8ny3u+8fCbNOg3hTaBh2CoyXen7Qw25UeMcM8sUF80AJxIFAOQQFju2NMlNuxSib2AXDZ6OgJaCk+ibHnBUYxPyuwGFlSy0wxXTpKBnGYP5v962mUbEslIYUxyHDMWYFodQnODlSEGjLImvvnx9XPi/4A+EooeWlLeUHLjic+qy4iX+La0gal5RbMuakgQr/0iB5nVveoy+dfMZdM45XQWxJGtkeNM/vut0esvf78t+5Y6RIWRvu/qJWCt5W+90ev6Fk5p4b5sPtvdfa9jwKxtdylreYCrhVutQdQWAfMK5ZfIUwhSXSzhcElXhQEshRRYWHk2/7yY2cWccANAiQJypwvzfFISZ5xWrMo2HMYlAlAs0FFOsiUvBKaK4cVTRRWjiiSpehIKIqUw2kVmc4fdVqzcJsosuRnElQqAplc4urrfQpn7EvBaB5lWv2Uef/8xp9KIXTmqIQLNkXjddsGgSbd52PPuVO2TwiE+btwzFWsElr5rapHvZ1Jis/RsbLcyEcBjT0jxsCwCgHO789TsqjZ0VyD8DWhLfH5L5Qfo5VMnmaBvgamYIZwKgxUBYUwTz/2fbb0rCjB7GpGITY+oRaUz5ZkzhSFHTbblp9Lwx+vy15J/Rx3nm5UWGOGnjAy8oX16MECfJNe/ZT//1q+MNCXGSgtCHrzol+xW7RB6HOEPKkXugJkyhQsjFAACIpEWEGQlcM6Blkcl9OY/bMg51upGI1vP7S31/aAaEGQBaEzhnLMz/322/oFG6cLwqU8zkv/UKMuGrKSSJFGEisIQ2qeFLNmcLGabbwol0KuYzrMfjtqrVQaOGV6lhTp7Sxr7dQSPfSoHmO18/k+Y9e0LmDppnzOoqume2N7l7RtQo14453BZghUOG9CTA6xE+BACIooWEmStRmQa0AyxAQoQEoI2Ac8bAvN9u/QWN0MU0JirdMgXDgCX5bxxhxuYsMc1neq8vhyyvUW6aetwzxsGRg8aUKNgzLbvcQXPVO5+g/U9mX8Wpo4Pow29vAfeM2s7VBtBIVhjWjY4KAKAa61pEmMlF+CYAAACQNuhmacy7c+ttYkRcXFaVKTC4ZlxRJqoYRAqyvJZND8yfEdpyE4c3RQg0ccWcWgQao/iTjzLbBffM4iZNihtLiDOIfqBRmKqU9ONoAABsyNKqnEg3CwY4UWk4bEtpnTdCmAEAANDKIKxJYd7WrT8SI+KlxeS/ESFMlFI4k44exmRaphrapCf/1V/VbRXacqOqN4VGoDgVnMIPp1HFKSrESZ2/InyqPEmwFGj6bzuburqyUxKke+Ztl85ozsTAtYguDSpdDsqocM5wAl4AAKjA87pXcGlVV2xkgbg/Kg8Mh1Ut5WvYyhpdPDJ0E3m1AAAAZAo/3KiFdfw7V0hcX63UPacrCKuu9UOcYeZt37pejIhXFZP/GsKZQuKW0I6LMIgyqgiivyeLgKNPD8UXveS2aRlkH1cm0Him3C/Kh6uVzg5zx2Qg0Lxx1eP0ra+fUSh3nRUveN4kmjbFK6y/aRBBKSqvGoVqWgCAXMM/8mXuKt8fajlnVbvsJ42Xzg+rE22tRYxlQcRF2ONeWS1GLjvu9vB8/Tys4f3q5WTmeh4tFSnMmJKgNyXa8dzjOl+YIoqV0arfFxWtrUMOIZl0utjOsVqvV82M5Vxri+9ZFvBDBhPOr6F1PNzo59+25cTfD9vDBSUXXOk3EOKMFGbu3vplGqa3lBwzYd/ahTuGDIKMOj7CLaM7VwrChSj9Y3bHiKgOdbxswKkINOH8pSS/Ae+rYeYUBJpd9w3Tpz5zkD7wvpmZJgi+/OXT6SvfbYLfIT1ELgrXoXwAWDDd8DXyRiet7TEsp67OAgsUK3hYaiu76nnd4dtNnFxyay2d6kajPFUK99PYkef9HOD96+d9zKwTZhCMar5pVPZ1qeWYXlxjGGNfFeGjVmT7rk4jvIjbRoo0fSzQmG6At7kUZiJu+KsRq9PJ33X13NXdQmtZ5EoNz+tW17c8Ytuo1u9LRCe8Gs4FkRhtrc5LLDCq+97wpPquz8c0YBFiRZx2pvG23qYIs/1ZbKtNLIlB1XO1jnMttd+lGN/BevZZMiMv51/ca5gyPyltHA79OSqUca3ndW+w3C+u03+jhex4tzPz7tnyQRoWHxcjgmiUO/uj7JwZ4ypNimgjfGVaoFRy8hXRwjQuUD4fkGGaKJuvJBCZKkUpnxOBsjwiQziWtlwiw/8RlaiUoSSQqNtApnmFZXyV+fVt1+Yt7au+DxXzitK+ff4zp9Hyi7LLBfP4/lH6097flx9/3kahj+Pj+r1vnEnPPm+C0+1a/MoHqZDgup68SWPa9oftzsv72EdOob/48ylOtvvwoH9o+jTvJCcLb1LYXql3Yk5qxadTfKNwhzrO94caZuFKa3sMy9nk+0Oxbor4xmw1P5UxijE1IMNR1uWx8k2K+7mXb376XH9HPK+7X7uRXFvNzkzjnYjemPt6cVxB0HS+JmQjl7B20o4sTKlJi2WHboXL4+Z53fXeBEceB257eTxXVVlOrHOkGnwOreEOYxIxbi8LeutM7Z7gnIp9jauVGtq66jbyfhuFR1fnStbrqAdFjFidUlLx9dzWLrfZdK8UB+u5muK5VnAe1iNyO7iup0Hi809p26TXMJW97By1CSO1bJuKeg5s4++FylZer3o/IMWjOep11fO65edu0D67tq0TAs/bseU6GhEfF2qn1VQyO6TWS2a1XB5R0/VkuKZxYeJcy7Ti/0Hk58vmsSTnLa2LlPm8qHkD6zIi51crBNn21bYM0pbH4z645ik6fDi7MKNTZnbQrNNzbEhTjx+qMjUrpo6Kk5tekB/kzTE7DA7yDWdSYYY4Qeytnte9x/O6cxE2wvu5LsX9nM3LOSjbjzsZuUDZ14dSPKYqaToyrvP9oZUuhRL5NNn3h5Zyx825MOMCKZKwSHdHCh24qvD6wnNoVQqdmtncWdhTR56FTJFiXsptLTtSN/P1cGWM+VsevkbJ80A6EG5OsdqbPF53yOOXwK2RGfw925DiuTZbOdfa+v5N7r/2PU7T6Snb+Vo+12Rbr67nHkAKO+rA34eQQ/p0y+9WDz98KKAI6hW0bfdr3q4tr6Yx8Vnd3WAMYTLlnUmCLsqYBAYVW9Wmap+PK77EmScrgabKsmMJNMo0mf/luvc9SWNjCY9ZTGSOm4ueNzmbldVKxHG1HieQR0zqf8vkYwCV8NOVPXyT4YLwRnEruxcagrKfcW5+NxmGalzLnc6Gd7xq3Nd6lr8ijh08JrJ8dZ+L7TTBYUxOhSAX8DF9KMV2j4TXt9XROSQ7Etdzh6Zh1wQTilC9xVFbz2bRekOexNysUUSZ61PuMKss547zBu6s5rEdevl75qLa3Wze/7ZLds7f41DwyuKaWSY8N+i7fYnyIGyDNUQ7223KB/Pu3fJ8GhU/kGFMQnXKUIQQYxNkahFq4jhp9PmiXnXBIparxIFAk0Lp7NQFGmVfNt85RLf9/GgNByoZf/GSaZmtqyZkm3gxB5BbOFZ5r7Z9l7T705dWhG9e+vmGwnZzLM+FG4noUiKaK8Os9IGIlvH0Gw3njop8Krol6yfmMfdThtRcqezjCsMg93Uuz7fRspwe7nita8TNmXJDGrWvxBbojZyT5ErONXMxdxLikNYxvLIR5atzlCsgFuxe0e3prtYV9xxKg9l8TciLs24OP6BwJVSryM54QwXrRsCOpK2ORRmdsK1z5Vji7/XNGbTDDbyutoC/U3scCV7V6OFzO0uRRr0f6eNjHbrQBtgpWqLtxJl5926ZTWP067KcMbZcKSbq6bfGcrsE0dP1V6HNa1xmjcJLOJ9XZT5doNGHMrHGsUATOYy36cc/fYBGRrMRHeY9023+mLpQhTSEMrUCph/xXIVsgGTwzcvWiCdKMvRjme8PyRhmmaDVmtCSw0U28HxzWKxZzzcFJq7PSryosp8DLE6cxCE166p12uV0nm8lCzVrLfsp3Qb9WX5nlKoMthvSUGiTx3UG7/Ma3p8ou7S+nlhJFGOwthHCTLPBN9rOQ5go3jnkipsbLdAo14q0QmviMJuvE20h0PAx3pJxG4eEwnlmLr0osvxeM6vyIoK6hPexP0Phz0Yo0mTh2OpTBJge7bxaracraL+uWED3FxKZViTONSWjjRHbUc0Jo00PTCKM/t4oxkS7ReK5ZzISaEjv8CcUaCzzBqb59LbnQYY33fK9IxEHK0VEDgUaCDOtRp+hw7kk684mcAN3BPotOUjWs3ukt96KDyzW9PJNyXrLbM7Fiyr7uZaT562pN7yFhZo1Efu5JKsn40qn2tTp2csOlVBoS1rVJg2L/KY0EtW2Og0SZhrRcSZ2EDSk86hcKxrRoeupsUJaU6K4RBrNtY12kbBAlKUwE3JzKwuB7PDOwokUl00ZVnRcabhv32h6ANJW3bF59275DY2JzlL1H19xypAlz0zchMA2gUGfR59fnxZXoNGXETu8KYh0pNQk5FQTaNT/vaBymXEFmmpCkW2cuh9E9C9fPpSJe6arU9Czz21CcQY0DdxZNXWcQoEml7HboDoRnZABrojQm1bIhzyPWKS52BLutMRVpyRiP7exc6RuUUZH2c9LDTdHUhjKwiVk61RLEWppyg6VpKEBA8hjVR1+AtwuwkxI5h3WBgszIXnpTDqhAS6RaqxqoECzNKOwORu5cA6lDd+X5q0yZGbnGN+3qfft1t/ZHJeVSZd5u7e8iHy6sKxMtamcswuEJuyo/4fvS4KEKJ9GyvRAEx705enzBAaxIlDWo89H2nxSUFHKUpvm++6ac6jDE5WfjSKGPjI2FtD1Nz5Jux4YsewDvxXBeIlvfT5t/tA986bXu88Jc8GiSbTxtoycOnGoJr5kV9AKpIRM0Mk3rfoNVegGWJNlEk+QHO6AmZLEOa1cI8NllDLGerjEEnmTzOJGmvtpE2Zc7ucG3s8NWic3FKGcPLHkp7B6p3qA9zXVp3acsyFpR7Kv2XK+NIAZGXeiTOdQyxNxTQQpkUNhJkQKNJTmb09MGn2uLZcOE5dlxhvEupx9j/dmHbbL9+0r+D7LmvS+bcQZCujnapnsikTAZHDLpGGy0IUYmzBjE2AqxgfjQpJBjCmJFfrnSV+mbISY83kWgWbAowXnTkzUPDaOHQ9o2tQOIjFc3F/TttG4e6dMoCG9zcdFr298+zC9/nXTqKPDyWaXeM58N+1SN1FJfn2TogWahNXcodRv2ns4wZy8qZEiTd6eVgAz6wwhPjKRXK/ryjW8/JWWG3V5k7w1RbHPJMysz+ImXIoOfHOkuxCkCCVFiVSrZvC69KewLkWopInBB1r1yW3KrM6qo5GlQyeHmK6JICU4+Xuez620f3uahTUpXMtzQ8rVA9OiUeeUvJ6vjhLf2iKYYd7uLTeQTxMrymWHRFVrqgeTg0PVQfRcKWXzVUkMHCfUqFp4U63hQWEOGiUUShwT5D3iTuGQzpmyECfTPuj7HLVP/LrvsVF6+JERZ9sdcspMx+pPrdjaDjQ13LlbEZE3ZAkn2JMd0l7ko8kv3AHTXSvbOClsZiWFWSAxnU9r0giX486ALiZuSylPSiyU740e4nStg6pn+pM5p+6gFG7o+5qthHWDyKpcdtYOndzALrBGVHNpC/had30T7OsN7VYxi90zrRSenrf8ZQNZhjSpcJh1ZHu0R6YJn66xumLSTEFiCzkShnl0McQ0zjSNyJ4cWFlHLIHGJHpY1z+eq0YcFyQeysB0VYuYRJVtUL6s8QP9o393X1Z72tScfbWQY6ZlUfJprI3Yx9mchO2gLMEKoSZ3mDpgexv15IzPp23a6J6kNzN8s6l3BgYcixVGFIFGJ80btl7tqf9AlJU5KfydThr6gupM+aKvHUN62lmUygIlXCxN5PVtEw/670dS2vG6lKuy4vXCv/u5c83k+SFEy4c1zdu95b1E1KGGNFVUZQqpu0y2litGHa+HIOnvSREYbKFLRocDhyUFBnGitDkBiTB0yZavxjQhIsRJHPVIPJSRK0QXhwJDnhwl10xhf9V5TO6lgOgntx2ld/xVD3V1ubOOjI0VBZrBw/lI5hJoYoxAjpmWQyrxUnjhG9qoH8JLeJBVATZyeIe1DDPIBFOHOlPHjGn9XLZW7RjKp3krE4TJmW6wG7afMt+L53Wv1QSj2ZyvKY0nfXo4xkrH37OkT5e34TqQH7hT067hTKsRzuSUtES/TXxd7zddO9jx0su/J0mO55IUr8vNwkqLQLmnysO4ekOI9iYQwaJ+N9IQmTaymLgnDAfic2sGD0uVIc55llTsW6cUS7Ddv6zmbSO+l4pCnXdP6+ecCehjuiAj1OS/tQgyJreG6fMm8cWUGyYUEkz5ZnQRgjRxxzSPqjVEjasgpkBzyCPxcMbhOmX7qgg0tv2KFMGKgpYMbXLNseN+oZz25i1DGTRSDFRxxn3BKtAgOLnoCraD98X4kQqFGmkb3ss/fv18o4XQhsaxNsPyjkY4N8tqQ2nVvnqeuFpiztc3Oukhi5orNYFsNeefSfM7cGMG+5rUaeU0NxWfA67dYHuyTvLokLQ7ouE1PuzkFDpULAIt5WPT22inDrs60g5zDB9EyOvq1vC7zefkHO5AtkUIFe9zUtFvW7W8GTR+T7Kar6mr+Zyu9/xycV2ul23K9bKfzyH1e5TGd8gosBiq/pTBocP1iDN7HIlfSR4abOOcexX3Q9q40m8XX89W8rXM9OBrfdKHEPz5yGXUcg+nz9vS4sy83VveSIHMNaO4R0KiXDP1dl5FhDigvyeLIFOaR3HjmMQYdR6yiClSy/CC8ZLhtvlsE7Tkv96jHY3p2JeJMUFlgmBF3IoUvZRt3/v7UfqDZ3Y52+T8hTXxziP5b1vAzoYN3OlcHfOHejYnLi0kL/W87k2Kq6ahQkGbsTcvTwdlJ5dz4ajnz+w6K0no+zSQozh0+R25Q/m/h2/s0gqryGpfk4YqZiEeuc5xsakVQiBSds1EdqSVjsaGlDrRSUmj4ljIek6Kb+xIKW2yjts87wly0yDptaiu5O1cqWaDoVpeXHr4Wt3I343I8ymEfzcTu5NapGpTvblzBupxm/L8su37+Dvdq4UY5z5cstUzT1xjHJt2vhlbrhky5IdRXsuS2OqvtjwqcfLGaHlZAn2eWpYlOdIgYca2vaZpUfMYctHs2DmcxR7kB+SYaUukSOP7Q7JDNFc+uecnp3FZzh2pLZ7XfUhW8UGumkzIm23btD01PdW2xJznJpSOb343aaPTfHKf1dPeRGFNLVi6NU+E+Thu5JAEOVwcISSllW9CuvCW1nJsuTLOHAd5Q+KSxndPtvfFUkSIe52R87HocKkhWXhLkELVnCuTVNXjY7Eiwbm1ukH3IHJ7l8U9n9i9t7TGey4T7ZYIWWVrGg4X+bDL94fk9exKFhZz/7CxZbts83ZvmUEB/VFkjpkQNcypVkx5TaLeR71WCAsRwkvZZ4LydRqGmgQaT5lXVmV6OAeOGX0b9XE0/lohehFV/P/oPvehTQDkBf6BWs0/UMu4g1DLzVEPP01EUmG37M1bSAZ36vRz5ZIaj72ps5U3EUpv99kpVghphsSmLdkZbTDye3OdFMd9f2iGFMr5OryGB2OeDiaNsvJX1uvC40TzSyMqATqBv3NJk1rLdq9JkFJh56mpmlsrkOS8Wp/G75OSjL0egaanAYlywwp7NXXq+buddFvb+R4r1WpV8txNIixmSSs/T19hFGaclco2VFAii7ig/G90z+gCjWm9NoHFNE5dV5x5icNgZFWmPR1EjUoeaxFfKgSaaq8G7e23v3OfC2bW6a2f0gk0H/IGgzsIS9lRcyXH49dyI3oJCzV72FHTzk930iSv4RgmcaGWm0593k15SzzLnQ79O5DGjdz6JsnfhNDF9NjErg0pEPTVeq6z0yypQHFlSkLv6owdNEk7s3WFQuiE+dsSbkuuYEG93pCtbWl2bPmaWO/y0s5HFMXeJNUE+TxKInC21DlYI/IBSVtWD2xlcebdhb+m5L/V8suYjDShwKF/wBI2U/a+ViGhFuGFFLEixnyxBZqjIruqTNUwbStp/9um6+jt5JDBIyiJBPINO2rk0wR5MzuDXTXXsVgTh9BRI0Of+tkyDeonrzcipkSxsY41dzT1xNROE88mQN+uNM7nLPc1b+VK2w0pDFzKDpkkIWJJBYqNaTnwuFO6MkMXSdLvXGoV0bhjHVUVp9lIcl6lLogkEC6W8O9KFvSmIK6jJHz9rPK87q2cP7FtaF1xJqALErtkojrxcUQZ/b1pmu6esblEkgo03vi4qgLNEY/E3py4PiKFJLvoFdhCvTJk8CjEGdBcsKumj8UawTkR1sZ8cio7hndApKmbvXktY8w3p3pOlrjH2DRfXsUZvUO9JIXQPeRxaQ8G+Al7Gud2kuvnQEqOrxJKgs0sSCIwbkw7bxKHhSXNG5IX6u3gbnKYj6re8NYs7jFSqSbIIlS94mYrhDUldWVKF+GtSu7Dla0eUt+S4sy83Vukojq18I/mlhFJ8suEVBNcSBcRyJ5stx5nTOSQUKA5zOWyG6wrTJ7k2duMLO4ZnQYIMvVw7DhqW4N8wjkR1hhCoKIIRZp1yElTE3kVLEL0m9TZMY+vHvKWWxHKIqQkKgOKkvRtwUA9OSkiSHLOuUo+3efaPZOCqO8q3CVv+bHqpd72dSbM8W9BXKeuShbiTJrHvd5rQ9LwxjyQlrAXOrVv5dyHYVj96lYLrW9V58zSWHlmau0TVxUIqieiNX5OrahEBjFCd4ro80QJQfq8nubY0R0zjSqXrdGhRlRZBS+qTA6sf6YJePjh5k5OPHmSu8b2BB12tnBQE2oIFBGdxOFPUY4a+SO6FfloYpP3nB+m7YtzbPV58irMhB0FvQOa5PzNel91dxPIht6UK4DoYYC14CQ0kgUf1wJyku/aRleiryUfVVPBv8P1lHUeSMkNFkU9nXfX9xXbUj6f2tZByeePi+/PbL7PvIFD60OxpulDoFo1Y2m6imrY9wzCl4CEEOMihuBpguzj1Fd9mUJZrnT26CJDlc+VKBtXtmCrWFFYp5w4Ikg84uVCmAmZN3cCbb7neDFvkHYMCgiDwGaaz8DzntvtdNuHh/PlhjnrlC56dP+Ik2Wfc467y4gQ5GajQSL4Rl0+Tevjp51rLHb02fyjmVZyylYmt6IFY9q+OHH/+k103m9St2rnchL3VzMl2YWIWh+b0uy8JnSPpN2h1NmQIKFsHJJ811wLCK733TX15miRoSSunUP1bJtrR0nenazNhrxfvN7xNodijcxTM8DHsK8ZSmfrtL44E9VHjnrgL/jDqghDBgGmQnxRlJAKYSUYT1BsElmixJzwvYgQXaQrxq9doKFjROKB+kKZpAgxYYIb58SsU7rs7UGG8SaxxsIpJ7tNdjw6Fn9e12W95TGadUqnM3Fm8mR3BrwjR4NHp0xxtniQAhyTvYI7FX2Wm6abPa+bINDYcRjTnwryBkceQ404N9X1PK3NE+0iWvTIMDWEYdVMlpVjquG0QylFKMM1IE2SCFOur5/9TS7O1Hsdm51Bp7ouZFJgh2IkcoWlSx9fK7O6H+hRhBrpKF2T93ssldYUZwI6I/a8qgijulNsVZziuGX0+cuEBU1h0MSFkpMl0hkTIbpUiBhVBJphWZWp/tPgqafHaNaZbk6jhc+cWHwTilq6K0YVxPRcQhy2ZcwxFBAtWzrRyTaHyFCfzdvileve95hbcUaKZ/NmT6TNO46nt1DlPD/nbHeXkSAI9jtbOEgV/uFbyk/ZTDdz0mWztRmfYoBU6c15wmi9E9NMeZP6EyZUXeq4U1LrsnPZKVTY5OB6lqQKTRadj205zIMxkEEeq7w7G6vRivnf5jg8Ls1+vHOFFP053OiOBmxXmAdRijSrm+EetFWdM6cX/prKaNeK7sgwumWUdVQLZzKJO6RrKCzQRGIRXUL3TByBZliQ92Adp4DSBoODPtGZdbZtFRb+wcTqbWs7NlWY/YwGnvpafhwpzrh0IEleuGAy/dtP038gOu9ZE2hkJKCuLjfbfsbpnb92smDgDJlAWFZs4qe46lOSHs6HgPCJ9mZ2wpwawE7Si/wKlx18FnBjL9/zuvMuzrhwAmZVIrhe8uisyqKz1ewPFfC7WwMOxL62fyglr/8yxF06qRu0Ccs5zH4tV2HLLa1cSjsZel9TS7AblMo/Geaxlca2TTfMU1Fe2zifJfGvF2Pewx5593cS1RB+Y8JlMtvODkHz5k6sbDP1VX9fjUDQvPMmONvmkKcPxGhY3icpcHV0uBNmJH+4cFJ6C1MSbb/w+ZOcCTNjxSZs+x+0ZoQ7YaakbLI0caplXgEAJZJeL5s+kWLG5Momn5Ftvy3D3hDuBxKC82c8ufaVDU6wfb18gJjnaqIt55yZt3tLcru0LTzJ9j7uKylCQrVwJFuelbh5ZaIcNCOCvEfSqcp03wMj9CcvSb4cE55H9IY/66GP3LQ/2i0Tdz94vqve1uNMUAg5eDh+Ap9du4fLq1M5wPMEveT8KXT75mOpLrz3zdOdbbMQhSOGH7QmhZ+SrDWEJqxxVVEENAV7m8wy3jQCMX/nkixiieM8Dq1EnkvCuwTnBgCgbqRAI0Pc+T6wUSGS0kUjfy9X5FF4bdWwptoxhcSE+V0M+UyqqRQSfgAAIABJREFUCzJanhQdfXrxdSsFtIEE7ZFDQMEcQWIOBbSShGJJjJtXxuP/VTFnhMh7MIYwE1P0+O2dQzQ8PN1ZSM6fXzSV/ukbTxXDp/SEynWIS9OmerTiohRdJBa23B0v30zI00+P0ckOkxRL8Wf1a06h2zf/PtmCFNfMBed305Qp7sx3nlc4ynDONDEc4tSrhbHMlmU9kXumbVmXd0txk5M0J0gvC6ggmtyJFBkJawiPAQAkgu//whyFWSYKVlnC4fe5y4HXmmFNtXTareKJ9l4PpwlDmyJCkyo+S/r40ob2k6A5O65esGzH2xes3fH2Bet3/PWCTTveuWD9Pe+av/aea+YvK8QhC1pXc4iTGuY0wjlm0ohE4vVJ14fLXCldnYI+8Y7T6w9jCgmKQ9+nTy0s0zXb7o0pzvAxunf3sPNtesaZXfTKC6cmX1AgRa4O6vunU506fg4f9p9yt3SQIaaOHsInNHKeJLfQ8TOMjvPEqZH25XYkaWjL6jzbvXNEHit/5D1fjSuc77d8oNCwvUsHuJBB7uAHNfL7u5ZdtVmzPINS8TXTujlndEJjSShikKWTbxJibOOjXmPmliER9O74ywUX77hqQeRJec818/fec838K0nQSvLoUM0CjRRmHqhRmDHtvyY2SUfL4JE6anDXwEXnT6YrXjq98ngJw76rk0OXDR+Gy189nc5/rtMykAVkrpTNdw1Fi0haO/72zhPOt0sKKR97+2mFyk11obhmPnH9KQUXkkuODwU7nDcKyAJTJybXQkSDyHuH2NQBiuN+0ufBsXdL0pDBnpyVh243kog+WQgIeRQpskgw3uzCF5yqIJfIsCIp0vj+0BzOR7Mx4+283vLwqWG0ojhT342f3rkX2rSKccWXssTAJrHGlti3OPTueOvC9bVs5j3XzN9IglbUJNCEjpl6kv/qzh8D27e7FRZk7pkPvPVUWv2Gk2NvUwl2zLz/3TPpA++dSZ570wz5fkD7nmAVTDneQcR2/+o3xwtVj1wjXUPf/oez6IIFMUO7SoJMccOnTfFo/U1n0ote4D407PTTOhqV0R2kCNvs4Z6oTt5FC9P2xQmh0OdBWIRD2C6e9Ank6rzdrOaQPDoRnH63+JxwGX5Qt4CQgfMQ162ckVECbJAhMh+N7w9JZ/VJGQs1uXoggZwzUQlmbbll1M+q2OYzl9Lu3bGqNmEm5J5r5m9b+Pmda8ijPgpNK7YcNCdEMcdMwqpMFSht89OfHaPnX+i2sy4FmisvmUGvfOFU+uTXnqZf/M/RypmCyveXvHwqvfOtM+i0UzsLy8iCXQ/UEKLEeXR2ZRDWFCIFmq98eBb999bj1Petp2nXg9XXLV0yf3lFD73pde7yCxnAD2/rsJUTsIU0642uS3dL3sUZ/ZgNxMxvIY/9KuX/HuQcco6Mo782wUrCsvdwOdlxdf4mWa7rcFHXy08ieK10fM/Q7KG4/Ybk/HG5OPvNBe0MJ+ldFzpBPa97Jf8erXCURLg3TwJN+4ozulhCESKNabryv3TPCBLlyyx7ryT/LY7v3fHm+oSZkHuumX/jws/vlCFOK6wCzYhHoiDMpNiZNrTL5t/Vlvy2XqS4csapnfTZ955ecKfcvfsE3blziB56ZIT2PT5aSHw867TOQvntJfMn0rxnTShUKXJdCUnnlp8erj6TQRSUAs3ihXWGHNWIbMsXPHcSXbj0bDpy1Kftu0/QXbuHCsJSIUzNJ5p37gSae9YEet6Sbpr7jK5CuFZWbXnkqD84dYqHqhCtSyOSv6WBS1Ep75Vy9I563I6QLawN4ow7+hKKMxTG4iN5c7bITonndQ/UeY2UwudK3x/a4Gijex0tNyTJNcFZ54odQ42qKpMWSX5XZjg8pwCoCp9/hXOQc6KFYs3KlO4ne7hyUy4eCrefOKN3iqmKW0YYBBabQ0ZfntDEkuLne3e8KZkwo7COQ5zGqzKF6x0RLMyktCYytVtQqHe877FR2v/kGJ12ajY9dykQdHQIWrawuzDkjV/8xuDqKbWZ/fz77veP0ML5EzNz+BCLNNOnefSC8ycVhiiyFLnGRuk/slsbyAA9RGJTkza6a8fPSu5Y5wp+aqXfAMW6WZcOGc/r3qvlhejN4362ClLg87xuaQe/JOEuyVj8PdJqnnXTtEAC1iToTsNaWB33u1kLHDbkWqBIIs7IzlWvo3O16QVKvibo1+G4rHRxTgFQDwZXzQq+p1iVsEFX5MWx34o5Zyov7oZEtmXTSu8NOT9sCXa1aaXKTdGf6d3xxtSEGbnMDWXr8HiQVadTq8rESZS1cXpbfue7gymsrPnZed8JGjxqSZBcpR1/selYpsJMXvF9op4e7xPt3g4tRhYJG7MgaWe3GnlNxGp6Yl7LTYx+Y7+kzTvfWZCW+HUzl8PPDD432jmsNcm+L3d0vJwLFOwaTJIvqS/tSmN8Libt9OWFes+rVchBBfKKdLv4/lAvh9+1RH7DVuwKjifKNVTwKWGqRGR6X/YaVI5XE75Gl9bu3fGGFIUZGdr0rvkDJFiMCtcjHTMP1SHM2NrANJ/hMz+5LcIt0iZIUeGbP1ZCmqJEwbI2LZ5XsvLVb/77eLs3Ix077h9CyEPrYEnUmCTxY0NuEtk94prZWXeEq8HtrYtSm2oMvzIJBagI5BC2Z6flUJMCTSbHi8///iYOfUyDpMJUX5rXST729Tp5aiXJvvekUK1MJ3PXmEOSuF9aqR1Ag3B5f8O/eUnu03KTY60tn9OXOUFsbhqrQGPocNucMuPjendcvihVYSYkEMGh0nrCUKYTDhK2Wp1HRUFKhjZtv8d9Oeg8MzYW0MbbByMFmcAmgvH7r647XMjt0r5tSDQyQh/LwabkkiZ9emX6sYwrzpgEgEa1QVZiQt4s9KbtqelGnYUcXShYBfeMc9K8Eb7B87o3pO1MCJHL9bxueV7d3ObCTNjJSPIEWLZfKseKxfUbki6nBpIKU5fweZQYXk6z55opwXk76j2vCjmoHGwWyA9O761YmLnZ5e9Iq1Twah9xJo6gQlXcMfp0w2AIb+rd8Vo3wky4joLYVMoxk7IwU63dFPo+d6ithYVbfhKRCNjUjob3m+8cKibkbVOCIPBPmuF9tW0bIALuyG5N68YzC/gHuO6QGIs7I/OnG9xByerJ8eysXArVsFj699aZ18F0Y4+8Mw7h78+NKa5BOqj2pP30k5e3p4XCR9IgaY4PKSr0JxH02S2Yda6RJAJCyKqkv5P8+VY8H5O0y/VZODszcqmCSma7Ek2Y8B4g/B1J/T4n4fYnqRaXKi0nzuw6b1n5TX+1PDOm8XFeTZ3ryvCm3h2vcSjMhOscFST2eunkmFGXG6XzGMQrKSw8/MhIihvRPIyMBvSl7xys3o56qJ1B9PtM38G2FbkOH/bX5+kCmRf4B2cdPxFNfOOZIasNT8FrDYnRHReNuHHL+onhmpy4pEznWV1tYQmzWZ4XIaqFWZMwj4dODz/9LIg09d4Ms1NmtVwO3DJG0rjGL2FBv6bvGB8bKZzemvVx4WSfaQhC8ndya63uPDm//FwLC4VJBfGbXTlopCjD14NbkeOmYTi5v2JRT8092MNuzLTF/iTnd27SKbS2c0YXF9QEvsKSwDfGZ0sd7KhwJinMvNqxMEOFUKYVnkz+O5ySY8aUp0dNCmxsq/EwsX/4xAHyDXmVWx3pmilLBGxKLq21aUV78uvGfz9CTz6VptLWHIyOBv7MmR1/03Y7Ho8+zV6de4GGb4qvN0yqdbv1H8wlWd648VO8LFwzqnDR0+jqGNw50y392xJWQzF1EtcgvMkd3Nl1ccM9m0WVg2xTX23JL1VCTuf55Ll9kMNl0k4WntdS9DXBYmYaopraCVodde2U07jjvSeFUuxJSMtRJ69fW/j8jPwOsDAgr21bWimUSYcfjCTtm0gHTX8a120WAntZELtVuR4ghKp+koT2rKkmuPN1YkONoq9t3tmK2N9X7znFvy39CUVVOGecImhb2eLjJLgtE1YCk9hSPr++3PL5enesdC/MLPjSjku8WqoyRbVBXIGKLO2huGfuu2845ga1BsPDAX3ploOVbaO3L4VCl6U9FQHsQx95upBguF0IAqLjx4OPwTVjpc9g9V7FN0gubah1wdtkEhfq6dybbjQyuXHjjoy6vS4rAehPj5Y0SoDjJ1mmzlmiJ1yyrDYRrdVGp5YfA0S2+3UOm+cSFlru8LzugId+ZSiMk9N5PldVz/a2WKLpNK9zs7ntH+KOkDwu63joZ8fCQyyoN9TFxOdrWsmsic+3W7Xzsk89P1kYaJewutUp/JYtZ+Grv9YwJEWQWcdC4M0GQWxVNbEXWElyHz3bFg6pCJgP8Xcq1rU2Zhn+2XzPIc+pPXxdWs2iS4VgwwKRnLaGhb07UniIlpt8NZ052AYX7C8tU3Z4w95w2AkOtPeleSP+r+h0B+MjRNm/vTsvWezeMSNPzoc7V9ddlUnwBuv7rM6n77veVmX7Pd7O1773Sfr3H86izs6U89/kEBl+9JEvPkmDx3xzG5HWjqb3pLVlUBS5frdliC44v7vl21AyeMQ/OH2aZ3JZAL5Z5RugO7T2WM629d68JELjTna/5Yl4zZ17mcTQ87oHtA6DvHFbU2N4VE0oApO63j6LGygxcl88r3uttny5n8RlIjMhTNpnWNda7jQlwveH1vC5rN6shTeEK9jpkSnsElrq+0Mt2xnw/aG+jMsCZ5WjSWVlI84fV0ghm50sabuLZvPQiGMUlzWG37s0CPc5z/vuFPkd4fMqjUTPyzk8lVhQ28riwB4e5DUnFN5XcNLZuOfzmjxV0GkW+H4xydYuYRF3Lx/DGRZxReaoWRrjvqBWkXk2/06VfqsS7k8c9qZxf5MWrRrW9Js4bpmK0Kay14jQJbLOs27nn2cjzCx67+5L6JhI/6IV4ZYJTMmS9Veu3PSl/zPQFnlT7to1RD/+1ZHKCTaHFZE56XRpvvFpq9/3ZMGV0+qMjgY0fZr34pbf0YSw+HKp4YnXbH5i3ddo9wE/Idlj+SFP0rk32dydhf0oApO6HwOuE9hK4cJS1WhdFseWbcomYWYTb1tarDCcx2EC08xCnJQKQde2Q/4bFvkyuUdpAFfm6eY6RXJVWj8r+PduY3vsbfZIsTZldxKxUHMtP2C4WXHKXc/D8hqFxuVwz9RNGsc2FHCjXC+Rv5kZF1NIQq7SBLSqODP+BDlKXLFNjxO6VDnPup2vXHylk73RWPS+3Uto1MGJFCUo6ONswgL//5V1A3Tn1qHUNzFPyKpK7/z0Y5VbFCOELtDPI8O5JZf/jmv2t7TIJUO3Bo8E781TIq48w6UwTR1b4pui1CupxEFJIHmHxRK/PmHn3hTW5STsh+28ujAj6c3oqbx0lWzTxq1KWnklCj5+GyxPUrelnbOE2zFKoHGe9JlvGvXEnze0Qf6b1Ybzq9m5MmEupNzS5iJFr+NQ0nZnZcrJwl2A3DP1kVXOupVVHhw1g7g8AHEmGyo7etXypZjmMTlrSO9IB0Vh5uUZCTPX37uExgodh3SeohqFJkPyX9VtRJVCgmle6fx4/InWTGwrqzO99aP7xpMA19iO1pLtVH7eyfCmf/znAy2Zf0bu02OPj/7upBneZ3KwOU0DPx1eaulgqZVUVrt2W3CnvloCyfVJw3K4M29yraSad4dFga0GYeZGFsacw/tq6pSElVdSvVlVyhib8oDIc8xJqBGfxyaBpofzQ2xwIUZxrPo6FhL1p7h576gkRh5L3x9a2kIOmpYVZhR62+Hc1FGuhcBd+67MuQAG90x9ZCXO9Nge3vDvdzPkcepzGSJfDy0pzuw6b9khErS78I/aSQ4RFrGBTJ1sS3JgNZTpzzISZj6wu4eOiuTCjC3kK4bjo+K1QqAZb6/BQZ8ue/1jNHSi9UJz3nvjE7Rrzwn7+aW/N7R5tHtm/Ly85fuD9F+/Ot5SAo3cl533Du85a1bn83KwOU2H/CHhDtaNlm0Pkz8e5JCYak83YqMk8wsrrkQlkFybVr4US8gPsWV2DyeGq7esb5jp31Q6VopLmYa7RAhwPVwpI5H4phzDqDLGzoSZEEWgMQmNl3Dc+7o0RBo+xmEyQ9MN4ybOO9MWLj7+XtquH82A7FBe2gbCTLN0op3Aongzn6e5JkIkzxNwz9QIiw1ph63ZsN3jNcNx25ZyyHYqtHIp7Z9aO8n6q81NYxVsFGHmpc/JRpj5yL09NJrAMWMTZPRpMcLAIgUtg0Dzlrc+3lICzce++iTd/tuj1cO+bNMNAkxUeJN8veY9+wsCTdACzSiFmTu3Dj29cP6EuTnYnKaGRYOLq4QprGLR4aBSpaI3ztMo7sSv4Pn7OCv+Qe7QR1VckU95L3bwo2cK+aFQtGCRZh1vb2SHXoavsMARlek/seunXvjmaoUlpEEV3zbUsL96hQxb/L/c76VZhHEpnQNb6MYqFmn6q5UCVuFzdyWft3v4GNue4kkRsSHJiBsJXz+ubMJOfygcNrTcfJbw9yQPOZEyP1f4PM2D06slxbEmEGiWZxHq2oJkJTost/wu5z1EeCDtkO20EEEr9PYMzNu9ZSkFtKVQAUcOvqDS+8L/VPa/0Ker8xHbHNTPSGHm4myEmYUf29UjBr1+8mlpaZt83qYxZTt9ZTv98vHCND4wzK9PM7RVaXnWttI+HxBNm+bRD759Jp1xevMWCJOhTH/18X20+d4hPmditqPPKove9lScJmzt5yttyONXv+sk6n3zdPKaVFaV+XM2/25o8I//sHt6DjanpeDwlCSVPbZxlYVaqinoDLBF1NlNgSVZbxTbtNKStsoDKnI/VpueynPZ1RK+P2STvaP2YYVeiSRqOZyodk3MErf6/tZyPAc4t05DOr18A74uxn4OKFVBVKfLHGWIs8+beH9TsTSz+0oV+dbm8amcjlIyvhkSN0oXxZq0hDT9+1wDFzeiQl5ENbUsCEuV31rHujYlrYbGwnKjwiQG2CFQz77HPlcaeT7W8duaJbKaTk0OSg4Brqu6Yj2/61W2peI3P4ttYYdz1IO0tLhRdxgrFS/z+LsywAJ/Lp2yLeuc2XXesq0k6FjhnzihJ1Xyp2jhTdkJMx/d1UPHCqFM6SqQZieQ3fGh/W/MPVPh/qgMcdr8u6GmS3Art1fmznn93z9Cm+87Xj4xTjtGurKCUqV3a3iTMr7vCwfpk/90oCAUNRuyHb+y7tBBCDNukEIC37xcWWfCzyV1VFMIkTftsgz0HNedUSVnxtqYHwn3a3mMygOkhLfkJlyCK2vMifn0WN/fOMdzQDl+DXMj8Lrn8LZEPcXt4X27RKkEcj133OLs8ybuzKzIW6x5I+AwSdl5uC7HT8+38TFb3W4OJxW+Lpmq9rkmTAzesLZvYLWxbewsaenzro7f1qzYlFeHQxPQm1EC+Irjw+dTHn9XtvG9Tm5DmFs5rEnyZWu4iCYoBKr4Ypp/fNy6nSsyEmY+vquHTqQszFQLb9LnqdZ+pmll48oFmre9/Qn64k2HmkZckCE4t/38KF36t7+nXftOjE+oCHULyqeR1gYU1U6Wsu2eWaCROWhe9ep9dP8DI00R5iRFmcOHfXrzVY89cfXbZszMwSa1NCzSyGvGMn7K7CqR5ADfKMu8DwVRJstOE4tAc1O8Wc91h51vdHp5n29M6WanTFTLQ6eX93MNizTXpXj+hufrMj7Gmbse8o4iAuapc7aXk/4uxTErolTty6rq1qa8PGXma2CWoXgb8/yE3QV8/V2WYc4SG+F3v63aP00iCgykzWxbpUP+XclLEvq1WYVsJ6Flw5rIFNpkCm9SwkmsoU3j86zb+cIl2Qgzn97VQ4OiFMokxkRFqFJNYU1kCbtRQ24M4UtqaFJ5OA6Hg6lhN/rnyRziJJl1Zid98h9OoSWLJ+YyREd+LZ7YP0ofWvM0/b8Dg0TdleeQiBMiZmpHPVSMtPAmY/tp7cjT/rK3h97xVz3U1ZWqAzM1pAh3y/eO0E3rDh781W3PgDDTIDhsYSX/QC6t07a8jUNI5NCfp5slts+uVPYxrvtnE9tuN8QVZPRcPfV0GGsNa4pYjrrPcY/pJraub2iWG16+6VvJHdKlMcO7iPd1K++r8449b6eaF25Pszpz+JrRy2Escds7TbZxiKRTB1uzhTXpcOiGq2M0wCFkpUp5CcIzEoc1qWQQilcR2ppg35sirMkE7/OajENTUvnuI6ypbP1z+F7HRcjaJv6uVL2f4O1Y04DwxPV8LWuK3+OWFmeoKNDsIJ/mF/6p1klWxYbK/B/rdr4gI2HmH3f10HEWZkIRpB5xhhRxplpuGXXZpEwLxRtT7hm9zSiGQEPl0y54bjd9+IMz6Zyzu6ijI4vWjUa6PI4d9+lzXzpEt/zwMPlzRzVhplzgq8g7ox4LvT2qtmV9Ao3M5/OOq2bQ5a+dSp4QuWhHyc9/cYw+87mD9OhTw0e2/2rutBxsElDgH0k1hju8cdbzeGxtxjACRUSZobgPwzwlhxopTKQlzliWS0reFfVYNq1QoKOdu6Eg0pL7mgc4z8nKDHIXDHAHoi+r72ezizM0Lk6vTlGkKeQO4+NQdu3PizijbU+a4sFeFn3S3PemFWdCWHhezdcBV0Jgqt99iDPG7UhTzN3Ix6ueB1QzWPzvdZjjaED5LjfV/UA7iDO9FNDNcd0zpCcHJhZmLsxImPnMzh6SyX8DFmaqiTOmJL9+5bzCJhz4ZE/8a3N9RAk0piHEIi4UjtN5E+idV8+g5z+/mzo80RA3zZ69I9T3hUP0i18eI+oIyJ89QjQloo0Djlgqa1N7EmVrO5YJgTUINFQ+XYo0l182rZAweOpUL/M2DIWtjf/3KH3jO4dp3xMjFEyg49t/OWdytlsCQL5xJc4A4BLFobYigQNPp6FOLk7iXA+xnhRnieYgrFVICzvH/VGOBe6k99mmR7BVTxiaJizYruZzs9bzcq9yDlpzbiXY99jnSjOcj+zYXFFnW6uE3/1+R66f3ogyz5GkLSQmOHdcbEsSYaRmt3GM7ZnD51KtzmcT25TvctOGwba8OENFgeYI+TSl8I+tk6yE7GjhTet2/lFGwsxnd/bQUdFPY0KrypRQnAk094y+z6ZqQqQLDJbwJpuDxibSlL2vFGmkwPDi5ZPpda+ZWhBsJC5CdkZGgsJypSDzo38/Sj+57Sjte2KUyAsKuV78OSNEkw1trLpnKCK0SRdddHeN5dyLJdCUta8m0vB72XZvuGIaPf+PJ9Fpp3YUhBMXjhq53KNHffrVb47Tz24/Rr/YdKzYhh1EQWcwtH3T3EnprxWA5gbiDGgF+CZ/qVYlK6qiSnizvJXdTcgj4RC+zoTHZIaWvzA8FntYOGmZY6Gcl2GnVg03VB12/XDZJUc7z0hrb1Icq8Rt3lDnKhiHhZGlyrVBFYL28HCIrxGZiB2G3xUy/Lao3+OtfE61TE6ydhFn1lBA1xvdM2R2jLDQsG7nBRkJM307e+iIKDpmxjSXSxJxRnXP2PLOqKJBlHuG9M/Q+LhaBBrSP1/pAgmRIsPzzu+mF17YTeec00XnnF0sxX3seECdHUQTJtj7MwUnxzG/IPhInn56jH7/yCht2XqCfv3fxwuVo0pJeD1OuttJRcfMZIOrSA9RCvfZ19qpom0t551pvtJ4i0Cjt6syv9524f9y/2X42AsunETLlk6k00/voGlTvUL7nBgOaPKk6D6hbOuJE4rhUsPDAe17bJR27Bym//5/Q7T5zqHC/6XExV4xkXHQFZzYfsfc7sgFA9CmQJwBAAAAAAA67SLOSBXukZJ7RutcVwgRxXnW7Tp/aTbCzOd3liX/TSzOWMYLk6tDD22qbIfy5ZJBdKDy/40CDRk+Q9o0ihZqVC44v7s0XYoPz2aXjUSKBfv2jZYqHO3aPVyoFFXAViXJY2HmnBGiqZa2LGszRZypEGKEWZjRBRzSx5sFl6oiTcV7i1CjtblMynzWrM7S/89+9oSCaBPy21C8omKlLdmOlaXDtTYMhZnbIcwAYAPiDAAAAAAA0GkLcYaKAs1qCuiG8s6y0PJ9lDrJ/bueu/TiLLZr4RfYMaMk/61bnAkM40lbjkk4iOOeCbR5TIKBTaAhy7zqq3WcqJyuoo8zdW9Kgkww/r86SEGhI6DgnFGiaZY21tulWmiT3rZkmUdvG4srRlSMi2g3m+Bla8M4VGtDuTsybKozGNn+n3Mn1LkWANoCiDMAAAAAAEAnh0WM3bDrvGV9JOjR8g5loDkpCsNWEoWkRM5Z+CXOMVMeB2wm6rZdVBElot6bpkUOgXm69tnACwpD2TR2VpTeq6/q58vmCzjkKBh/rw4d2mCaFr4Pl9uhDJ0FpwcFMpTJmLs8sLdtaFCp2H9DG1W0r/Hcs7RpUNRyPK2NTO1mfLW0XdxBb8Ow/ZT3sg1pgg9hBgAAAAAAAADqoLPNGu1NpaeVgl0EYQe3aAiQwsyKXUuXDrjekIU37eiho148YSZE74wXhANhcI9o4/V9NY0LlydVAHU+/XP2kZXLLrwURYiSk0Zo6w7K5y97tYlLtWIRPgrikQxlOnuExHTFaWTYD/N2K+1Vfh6VT7PCC4tqzlKbBcXDQ4LbU9kOU5vp2xPjsFUQJe6VxCIWbLqCke3/cS6EGQAAAAAAAACog7ZxzlDRPdNPgn5iyTtSFGaWuBdmChyL6ZipBZNLQ3sfkBaWok6vEH/MrpiydvM0p4vFzREIg5PG5vYwDVUdPYbPd2iDp4QwSadHF+eYUXPK29omql1C90yEk6hqO5qmeQaHDLuICs6kUCDRB9v6bG1brc21Ng2KlZiKThop704MRiHMAAAAAAAAAED9tJtzRvJGEvQIEScHLhBsJRIrdi3KRphZeMPOm1MVZoQhl4jJGWRzrJTNY3CDUISDpqo7pHx7AnaACDVprcntUXXdlnVY3geqmCSKIoN0zNB0ze1TvLj7AAAgAElEQVSiO46iXD6628jUlupG6fPoO6g7i9Rjqo4rtVdQWn24DULfBn01dbRhoK5PF2w6g7HtPz63K8aSAQAAAAAAAABYaCvnDBXdM4c4vKlI6JhZmJEwc2NBmOlNbYF6R7+Kc8boqND/tyV9NY3T59VdF7pzg1/DnDRluWniuGHiOj3CvDcdxaHMRdMVFIWZGYb2MbZXUDnO4KIxu2eCyv23taPNWaS7ikzHg/PHlNqUXTXqUK39Co6YcP6OoNR+4/l7qJSnpzBMKAgz7SjwAgAAAAAAAECqtGXHatd5yzbM273lRiJaTgGt2DV/WTbCzOdTEmZUd4U6jlRnhcHREYoIQUBCCIPTQ12Jxa5idX5Q5QR9G3WXjOJECUKRx7B8ERi2g5TP6G1gei2JEAH5s0aIThKVThh9m03vje0+3talzbc5aDzNFaM6YgKlAaJyBEVVZ1JFIb29bM4ZmyvJJpYVwpkCf/vGZ0KYAQAAAAAAAIAUaJtS2ibm7d4yddd5y45ksa6FX9h5MwXUW1b62lgem0s0+1RRJlv4hvlJmV8v3xxZNltJ0hsoyyqbV1SO0xPQVkxXevr6dH2ciWrjo4QE/X/VrSKKzo+iMEPl5cZ534WvtCWRVkK7SplxfzwcSpjmIYpdhnz8f6X9TW1qa8+or3RUO1ZzV3nsnpHCzK3P7IhYCwAgApTSBgAAAAAAOm395DsrYWbBl3bcLEjU7pgxOWT06ZEuD0sOFTI4VIRhnOyM+zFzypSWE4wLC7pLhqqICpHLjTHeGH41Xv7ZP4tDmSKrMiltVuZW4fFq+5HW1ux4Mabh0R00FdM0Fw0pbVmyPFmcR1SDOGMjynUklDLkXYG//QcQZgBIiAyv3YRGBAAAAAAAIW3tnMmCBTftuJl86hW+4ryI65zRxgvd1aE7ZHTHh809ozhABBmmk8kdUsVFQ5bPk+amIcNnTO/jYBUUtDwvqmPGNzhhwrZQXTI2J5LBgVTRToGyvDgOGr29TG1JyvKrtVct7Whz0OgJlGUemq7Av/t7EGYAAAAAAAAAIG2QM8IhC/51R/rJf6PcMlF5USqcHsV/S/lnVIwGD4slRHV7mLYhdIwE+ocq3yZy0QhLUuFOIv+MUaKZehsoOxnRPuXvNVeNPl1rJ1szlhw0ulOmbB5DW5a2W98uUfnZWtsxqqR3oXS279/93T+AMAMAAAAAAAAADoA444gFX0lRmDGFruihOGryX/UzpAgBVEPnvZpAUxGCo4k0+jaqr2ryX+OORqAmurXmRlGS/54+RnRKUHS6mMQtRfgY102CccFDb7dQoFG3QW8rJRzM3tzagdLbz9aWRvEoqFxsZBsa3pvaktsw6AyCu78DYQYAAAAAAAAAXAFxxgELvpqCMGMSOPT/be6KEP1zZBZdjO4Zy7ylhdpyyuiCTFSOFH3ZUSF2JseOLjKoYTgdAfkn+0Sn+bwuEd2OJueMVQwxbFeUQCMMYU6kCCuCysUgXaAxtSUp4yni/2ptqLdfqR2L2xV0BcHd3/qDtiu5DwAAAAAAAABZAnEmZRbcfM/NFDf5rylhbzjeUD5alo4Wtk48ae/V5ak9e5OII2IINKpYEBBVbLTJ+aFvT9m+G8aZ5rWZaWwhOFxRyD/JJ5o1Vhk2VOaeqUz+W9CcogQv1T1jEsN0gSYYn69CjyprM0WkMbWf/mpaX1R7xW0/JddMwTHzTQgzAAAAAAAAAOAaiDMpsuDr99TnmNHDbaKmWd0ylpwwaugTWcKbvGKCW6NAYwqxUcfrThp1+fqrPs20r1HYQnCIhRAOZ/JPGiM6yy+vylTNgaS3pzAoKmXvtfAmMsyjjlBeCofBGhIWKOM0oUZvuzghTDrGNlTWKUUZr1AuO7j73yDMAAAAAAAAAEAWQJxJiQX/VqcwUwvV3DJ6B94klticHh7rOxxaJIRSUUif3xR6oyb91UOeKKGoYAojIhov8Rxuk3TMzBgjOjtCmNG3nwxtWlqP5pCJ60qxCTSmMZFhTIpoQlq7GkPOqqA6s9T95CEIc/VIYWY9hBkAAAAAAAAAyAqIMymw4JspCzMRbpnI0CYRVOnAG3r0lk5+hYtGD2syuWiEQaghqnT01CoqkCZUqAOx20PmmOnxSZyjCDMmYarifWWIkmpGihRDbO4Zff4wFCq0zYQuGqXdRKBtn9F5pIkq9WAIZwpCGaajEBIW3L0OwgwAAAAAAAAAZAnEmYQs+FaKwozRvWFwmpjcH/r/ZPu/DoHGJGqYtoliiApxXTMmMUcY3hfCcIrCDJ0zVllNqjRvYE+6a3AYlXLPWBPGhOFgEQKNcb9YpNFyClmFGts5UG87hgJUmchVzNNTEGa+BmEGAAAAAAAAALIG4kwCFnzn7mtjJf+N6rubQpXIICaYBAXjqyYmGAWcmAINJwoumkssIo3u9ogKZzL9X62tTOJCyfFRDGvyp/lEs8dKYVnC1uBR228LdQr/MSUAJoNAozuM1GWVrV9x0mjrC7S2Mx2uWtox0I9J+N4bT/5bCGX6yrMgzAAAAAAAAABAA4A4UycLvnv3KiLqi/vpUjiSSWhRieGWKVuWTXxR3Rwmd0iUQKO7YxSRpuDsEAbxJ8rlYdq/auiCgirKCCX571SfaO5YMZRJdx3pzhhTOBYp4Umay6gUjVUtvMkjpQy4YQfV9qw4/qFVJ1x+hFijYmtjW/uGDhlN4GJhhu7+VwgzAAAAAAAAANAoIM7UwYLvF4SZdaku1BiKE+Hu0MWCqvlRxqsyjc+jzKg7PnTBR3kf8IiCQ0UXZKptB1GliGJqC+21sM5QTAiFmSljRHP96FCwCreKKJ83qv0CQ3hT1H6RlofGIHJZ21cV1irC06ooWsY21PLTCG2dniLM3PSsuJIZAAAAAAAAAAAHQJypkQU/TFmYEUrITNQ0oYsGFvcMVXHGeHrYDYsecQQAbTsC1fUhNLGGyLxPVdsj/Ggw/r/q8hBclWnyGNG5WvJfm7hl2f5yQUYTVdR9EIZ/qgo02oL09cYNBSsdyxrQl2dqRxnS1El095cgzAAAAAAAAABAo4E4UwMLNqQozJg64ia3jP4ZXXwgg1CgujBsoUsV4wwuGtM2RwgcJbHGtp4oTEICGUQFWZVpsk90rgxlMq9ACjtC35fSMg35Y8pCjESVJqsi0JQ5kMIwIksYGFle9baoB1P7CRovP95FdPcXIMwAAAAAAAAAQB6AOBOT+Ru3rxIkkgszuqODIv4Xls4/Ke4ZsuSeoToFGtVFY3J72FweJvdOLW1iey0JM5z8NxRmSturuI50cStKTDLti41ArwZuEGh0R4zafqqTpqqDJ2Eb6u2oilJqKNPnz4MwAwAAAAAAAAA5AeJMDOb/3+3xHTOmLq/QOvD6NFsoDpE9TwpZOva6oBNHoClze2giTTW3h03giCMsVGsroWyLDGWaEjpmDOuwtYtNCNPnL223oTQ2h4KVtaJaEtuWR6YijEldiEFUq0fgMol3pXFKnh5+DToDuudGCDMAAAAAAAAAkCcgzlRh/o+3v6XmUKZQDLEJMhTDLRPH6aG6Z2h8nNUVEydGpmy7lLwvJqHG5PaIg80NpAoyNC4olJL/PlPJMUPaZ7S2K4U2Gd1HlrZQw5t0gUbdVaGmgYlo10hhy9Bo+jqrtWvZ7EH5OKG0nyrM3ABhBgAAAAAAAADyBsSZCOb/tAbHTC1UuGUM7pgyAcQQvqNQlmPFhu6IUUNw/n975x9i2XnW8efcmdlNQTSgIv6hzaZ1Z2aT3Z2dapUutClBIyFiIGirtOmCRdKCNIb+UVBpK1hDqdr+UygIjRaKItQItmorZfsrRdTZTUJ2ZlLspFqKRQqJQv9odueVc+455z7neZ/3PefeuTOZH58PHO6955577jnP7P5xP3yf5+2TQsEIE5Hu9XoSISWkvNfeykJKKlRipmplKuLPTzMbp/O9odu+Za8vI2jcNqdC/KHKOdEWCa6BlqtXbql99fDfsCDy3J8iZgAAAAAAAA4jyJkEq/80ZzFTqBWWJJ346P0Rr1GyxhU0Wq7YtiUp/GNsa46b+nCEyh7q0qmHljOjevjva25lVmUyA36nqZ9tJ4oSR/2Cpk3RNHLFSppUTcW4mGnTR2Kut3BqWUqZ0bgl7LmPIGYAAAAAAAAOK8gZh9XPP/MOmcfwX0mkO6ZNfPQlPUJC0OSkTmr4b2qezLykghUK9nGkhv++alfktbWY8b7bJpCMZClXjyqKIv5soWrQHpyqVULQqHpFkqaZRyMqpZSrxcCOM/ez+nUx+buGJnlUtjJ9eBkxAwAAAAAAcIhBzhhW//mZ+S6XbWVCKi1TtIalm4JIzJqJkh7TCBotYrzhv15ixpMqMqWcybXjjKQ7/Pe2XZGfuRVfS66O9v6sVHJlyAAzogWNlVjqNB1JUyREjSeZpsGVW6GtQ3MNzapMzz2OmAEAAAAAADjsIGcUq1+cQsxYOZKTBZLYZz+Xe5QBSY8+QWPFgpU07T71HWKu2z4f8tM/lZYpxB/+e9uuhLO3pLDDf8XM3mnPaVqb1GNVC/2eS8+NFKIG+BZmv/m76fFBnqjx6ubNqnGvw2klK7rf1y6XvRDkuT9GzAAAAAAAABwFkDM1q1efedtMiRmdeHHf7wqDdu6Mm/gwrUvuT+sekeAJGisRxHx3lJIxoibVfjMk/ZFKeuh7VFLh1m23RM6WrUxFfB59vZ7MSiRl2vYme92drzA3p2vWEVJNSqWI62fllxE17a1HUq2nkPrSvfRRoQcoh/GMmT9CzAAAAAAAABwVkDOlmPnyM28SkU/N9aQ5aWOFgn1PEiLEtuJkZ8qMRVAVNkklM7zkR5TwcdIa7Vu5tElCOBT+FhaChNtCLWac4z2RNCR1JE4tU/v1jlRt25oZSePVz0keuVUZqlEK57ERMzKWMuX/6Of+EDEDAAAAAABwlDjxcmb1q09fFCmenNsJPZmQa3ny0jNWvHjCoXwxKvxz6U2naOx3WUHkSQX7XudeQ7w/JZtsO5NMxEIlZk4HCWdvduoWCuV4eusYunNzjIzppGeS0kvSgsamYqykEfPd7t8sUce+Gnr1K+oPNs8X6lamD64gZgAAAAAAAI4YJ1rOrH6tFDNyVURu3/PJPGFg3/damzKzY9x0jZegyWFSNJVDCCbtYU81VCoMqYmY89T3E0ZhImdKMbN8s7tcdlJghYkIGZI60q7Frt4kzvH+jvhaojamoD6WGPw74M/lfl/nuRIyup2pHP77fsQMAAAAAADAUeTEypnVr89BzKSEzJC0jNd6077X016TEjRe+45J0YwPCxNJ47XjpO51FkziIxQTKVOJmaVdCauZVZn65sx4qaOU2PLmz7g1Vamg0FNbK7hEzZCJPjNtEcPku3QNrJwpxcwfIGYAAAAAAACOKidSzqz+y9MXphIzjVjQyYvcLJn2tRnwq37MJ9Mz7fOemEWfoLFiI8TXHPTqQaJm0+wlNWPbiQojGEZWzNwc1yiSV4na9bUOpZ53bsesZuXdg07qFBJfo72u1LWIfj6wkFFaxrYzNTWsh//+PmIGAAAAAADgKHPi5Mzqvz59XkS+NJdWJk0q5SE9P+I9EWKlS8rT2GO1RMhJGpv2aJyMkTUNyYHC+lK8ZZ7FSIVmK2fMLIWxmLHLZZvrDdFy1M2cl6IrLtr77UkdtbetltjWm/1cocRKU19JHCd+XTv7+0iJGV3LUsqMxnNmwkK4Y+CZAQAAAAAA4JByouTMyr9fv1BIMV8xk0p5iJUG9T71uhIEoYh/kEfSZYCg0RJBt/YMSZz0zEUJqZWXUvUQ9Z027VGKmcUg4dzNYasyDZVcnWt32pu80qVanFL3FaQ7BFknjcR8lz1tXwm9y+jUUrWDFeP/uWExXLjx3tVvDbsBAAAAAAAAOKycGDmzsnF9fsN/JdG64ooEZxUhL5mRa88Z1efYdT7nzJSZCIREC1Zv+40h1cKV2lc4z+u0R5uY6dRJXUynhum2sFZa5VJHjXiZRtDounr3F+zftnmdGALch70u/TfU9WvrWEmZ8n/uhRuPrT47xTcBAAAAAADAIeVEyJmV63MUMzbhIVYoJFqKRLo/3uv3qtkzkkjGWNlgl87uu87gyIOcaMlIjM79pT6vX+u0R1G34OhWJi8pk6hjp7Updb0p8SXOsfbW6uWw2zYnfR/6vCmpJjIZAiwZOdOKpsQBObnV1LCUXEty+cajiBkAAAAAAIDjwrGXMyvPTCFmUmKicOSBfq9JUUgRSwLdZiO+RMgOqPVm0Nj9XtojamfaY8ojV5vOo1nquZEKi7vjViZPEqUkTe5R1OwZe33e/JmolnHtooHPqfavWQVXqthJuTXZQj38Vxbl8o33rD6V+gYAAAAAAAA4ehxrObPy7PXpVmUSI1RSKzOlUh72eeq1xD/i3fkz4h2b+fXvSQU3cRL8Uw0Y/Dv5LnVTkaBRs1GaVZnO3YwFUiYtY1ubovRMe5zT3tS8X17DbkLQeEKrXpFr/KdPtKKlWtBmxaudljLFZPivLIbLN37nHGIGAAAAAADgmHFs5czKjWvnZR+G/1bpiuaXtNdSI6kf8UYS2B/4qX0N3kyZ1NLXubRHtu1noGXwkh6ipEIrZsbDf3fvuimFHf4bCaz6wlJ16b125zobQRNUvYKSMl7NpCtpxpeUmRkUCa4B9bPXb+oXRA3/bbZT4b4b70bMAAAAAAAAHEeOpZxZ2SrFjHx57qsy9c6ZUb/UU+mKKMUyeZ6dP+Nils/2rneeaY+MUOg8b4b/1mKmuypT6CZ0UrIjxJ9p0zP2WnR7kyWXOrIpGns9TcBIz+4R6Q4QtnXsq2tCbLVCxgqu8fDfh26869zne84MAAAAAAAAR5RjJ2dWtq/Nf1WmXFtSXztTR8aY1YfEnKsUGrtG0Oj2G5t+ab9TSZpsO5P5Tsm8tjXwXketOKGVCuXw391zk1WZOmIlWUezbLhzD6630u1NHp6gsXUtzHdm5s2E9qbii8ktzR10kXvamao6LlZzZh668ci5zyRPCgAAAAAAAEee45ic2buY8X6U5xIWkkin2OM7Bxq0oGnmnoREisYKGJ3caa9D2Qzva1NtQPZ7cjVqhIyWCgsiuwu7EqrETOHXLZdUsUkcPXtGCqee+m8wQNA0IieoD1qxlrouWyNHaIVgdnqX0ytmmuG/4eEbv30XYgYAAAAAAOCYcxzlzDAxkxIeuR/q9fN27ownX4pEOkYzZK5MnbRIpmjEvLZCowhGKmRacVI18d5r4i9WotSzUXbLVZnubpbLTrQwpZ6nhIi+8KKIOsi6KaSEoHFrpWbR2O9OJY+859PgXXPTwtQd/vvwjXfe9akZvgEAAAAAAACOGMd+Ke0sOSEjTSqkiIVCKi2TS1oU5ryDVmaS7jLbroRJtDyJvc50K86gOkX30G3BKZd6LluZgmplio7VKRi7nHf7vll9KZJf5pYaMaTn2tgltoP5+0YpImeZcV1P9zNDi+fQScqY5FH5P3IhPHzjtxAzAAAAAAAAJ4WTLWdSpFpvUs+tkNHywUtaFDKZP9MnaPTyzvZgTy7tl1RIteCU17cwXpUpRMN/u1IkGuibEC9JwSTdyEwkaPQMmWBPbr432cbkiBpxjk+ljzyiNiYlZEStyFTVUR7ZvIKYAQAAAAAAOEkgZzTeD3FPzDStTTbh0ZE0jnyxM1VSgiaR5AmFkTS5JM1epIInc2xyZTRZ7rlKzGgx48mtaVJHYgYDx5XwBY17qErieC1h2TamRNooJdWiOqpie3Us9HLZZfJIHt58GDEDAAAAAABw0jjecibVrhQdZ9qX7GetXBjSziSJFqJg23BMyibVgmPEUPO80IN/cwkUfb6+mmWet1KmI2Ze9of/NrXVM3i81qaUGLGzYzr3M1DQFDIZAGwlTV/yyF5T+9U9RUyJG5s6GtUXvlAtO/7w5tvvRswAAAAAAACcQOYiZ5Z3Nh4UkbV6awbylqsmXS8ft8+sv3goS+vJBOmRCOr9ah6Ml54R57OaYNtwzGBaex1esqMJdeh0hyhZk7vnobVp0jpaKMgk7REW61WZgvjDf72aFUqkJFNHCbEUCbAeQRP9bR1J0yfZZmlfsq9tWqaQzvDfsBQe2fxNxAwAAAAAAMBJpYiW/h3I8s5GKWEerbcfyXzqJRF5sjzuICTNyva1UP2gLpMpu3UaZrfeSs9Q7quf6/2dY4OY50V3v3qvXe5aHy/O58V8Xh/X2dQvfO84cYSBtz/zZ20ETiicg6zccNtwQqX1du9+Oa6hV7/dbo2KTh0Lpwa6lk59o9oVnf1Fqmbec90i5tUuVcdOOsgrsvcYJiszVUOUReRUeGTzrXd/IvEtAAAAAAAAcAKYSc4s72ys1cLl1VN8rJQ0V7bPrD+5n2WdWc40AqZpN9LHBEfcaEEjjrwRe/xAQSPmeElIhZSosZ+ZBi/5EbXi1GLmvBIzuRo6gqvQEkv/jSSu7WBBI11J0/FOubpFdXRaqWauYehKmtGkjuUAZSmH/74FMQMAAAAAAHDSGU17/7WYuTqlmJE6XfO3yzsbVw5NzYuepEi7Lzj71A9tCbHEiI4L7mfbfaPuj/dKgtjP6GNGZvNEytDNO++C2uqVhGQplEkP2b3wcn8dU8+L2ju19x7iWtg6SYhrMDJ1KbrnKlMpQZ9rNOCxmQHTbBK6r/u25m82qremhov1Vl5T2cZ0qqrjY4gZAAAAAAAAkGnljBIzuTamPj65vLNxz56ueq+4CREjYLznKZkzeHMEzcgREp6k0Z/NSZrcfm9bMEKmsy+Mpcw45VHJmVtazET1CH79Uo/2vYT8igRNToQ1YsRImlA49bW1mqZuI1M3Wzu1LzTbYiu4Htv8tfN/tq//xgEAAAAAAODIMFjOzEnMNDxRz6w5cKI5K32JD53wkLQccNMz7hYmEiOXeLHCICVqvETJtEJBLec8SX2ENu0RiRlvVo1XU5uQUfuS6ZncNlTQWEkz6hE1e9k8uaWFTJWcqQXXkoicDo9tPoSYAQAAAAAAgAmDVmuas5iRuiWqbG/66IH8LZof4t4QV+sZ9LHeo/6METjl/J7CLv+sz9uZLTP9EtrdawjdY6MvS9UiIafcFEr9uBjk1t0vT96zdSx6rlfXTe0PzeUU0r25kLkFb4ltW1ub5mm/d7JKUzD34S7FnSljdFmRIDKtWAuV4PqTzV9FzAAAAAAAAECXXjmzD2Km4dF9kTOFfE+C/KgrZCbHdAVLKwzMstmRWKh3WtmgfsQnBY0rHOqL8CSNd+32e1Pn9YY8e6IhSgc56ZNFkVvN8N+c3LI1tcfa+kbvqZuxIsrSCBortTzJFX2/vvbJF7hzsVP/dtwamjST6ERN2cokf7n5wPn3Ju4IAAAAAAAATjDZtqZ9FDMlr67PP2++nTxfkZEUnqjwHsUIDPs6NSDYS6V0BtmamTLeo7flWpbsUN/cbJrO3JS6DeeUyK2LP4jKFbxatK/zw5PtvhDtV8tNe5/Xg3vtoGD73Ktb596dIb62tctu9v320dS2GQJc1vF0+OvNB86/Yx/+rQMAAAAAAMAxIJmc2Wcx03CHiFzf9zKaFE05d6ZKt3itS+I87widXKRDHZVrcbLXFkyyQ6dLbELFa6+ahqR8Ch15UQ6v3S1bmVIplNR9WNkStYDF6aS4s6mUHcWwVIw+wEseeX/fqPVphjrqz+r71cJt3Mr02c37L7x1D98AAAAAAAAAxxw3OXNAYqZkP5IzT7bPUqsHifOjWtSP6lwyRgYc4yVo7PLP9jOdJEtID/+ddXitTdDo5MhC6KRswpJZLjtKx4jfxqPrE9U189lUgqavxp1kjErSpJJHqdqkUknelkonLUo3SbMkEk6HL27ef+GB/n+yAAAAAAAAcJKJkjMHKGb2Dy9ZkkrGeO+LOYd93p1kO/6xvuskMIpxgmb8tOieOzsbxfuuKQf/dg7Tk28T0kQlZjZ/5Xyx/M2NMLiORaZ+XqKmPWn3+t0EjT7Gfpd3HcFcRC555F+Gjyf5mmu0wmhUCa6vbN134d4BZwYAAAAAAIATTic58wqImRf34ZxXO68SaZmgUzVRMiP3XiJBY2eemERGsKkOL9khzjm8+TDe0te5mSk26WETNYtjKRNOjcVMfR3/EdVRXWNn7oxXn1SypicNk03Q2Dq5y43rx6ASNWEyz2ZPS46bGTU6OVMvOR5OV2LmjQP/vQIAAAAAAMAJp03OvEKJmf2YN/NC+2xIWqZw0jHS/LB3Uhsam6DJHdukaJoFiXIzb7yUh5fuSM2cSSVBrPgY1dKoFgyb95/Xn/ymiLymrYW1MYW482OSs3rcmjirN3nBGT2LxzswmTjy/rZTDOpJpmUc0VSLm7AUvr5170XEDAAAAAAAAAxGJ2eeOGAx89L2mfWrA46biq2zl16QQr7ffsamYLwf2eL82Pb2eZv+Du/zNkUzmiRp2vTOkBkpQ67FO9aZkRIWQruVs1HKzYiZkqciOWFqFqdc7HFDZvPE82eaBE2IEkVNYsXUyd6vrV1fMsZLyiw4x9jHNnkUxsmj0+Ha1r0X3zDvf9MAAAAAAABwvKnkzPLOxhURuXjAd/rkgGNm5WvV57wESaHlQuiKA5lFLiSOTbUqGYlSXkMkaga1NPUsl73QlQmhTHXU22SA7bgNZ/OXIzEjnfawbIJkyuG/th4jc8zI1ichdEYJsZUSVkNal7wlx6Mhyl0pU7aNhbKWp8MzW2++uD7ff8YAAAAAAABwEmiSM4++Avf6gX089+cGp2Xs40zpD3WsTnakZEFCGlSiRsmaZpt2TkolYYoQCxm7lWLmly54YkaqVJNXB02qfrmaevv1LBfnM2GUkDR6/ktCfvX+3XJ/J32+tnaTWT6hWtmqFjNvunjQchMAAAAAAACOCYvLOxt3vAKpmQ9un1l/YcBxsxK3SxX+PJJSYlQrKXnvi6yBXg4AAAlSSURBVJkrM6pXZUrMlGlXEGqO7+wzz1PXJGYsSqjnwgwhlXDRzxsRUs6YudcXM4oviMgvds4Z1a98ambH6LNOM5enveGi+13NYlXqu4uolmZFKzvLZopRM7lUUCuJmjqekme2LiNmAAAAAAAAYHbKn5hrB1y/v9g+s76fqZly7sx1KeR/qhdNG4wVF32pj1SCxpsh46Y0QvczNjljv6uvhWma2Smp9py6DaeaMdMvZiRKIKXqpN/36qf/BkPSLDZFY2vTpGlGai5NVM+elayyq1t10zjV9yyMt8nqTKFMzGwjZgAAAAAAAGCvHLScKcXMlQP6rk9nW2zU/uCJg5ygyQkGLRLaz6h2p9RMmb72mr7NOzZqxamH/755kJiRai6QvX+nniElX6Ia9dQvqocjaZzHpu0pkjVDN/O9ISVkRnUtl2ox84a1len+SQIAAAAAAADEjA6wJgcpZqRefWpCofpaHDkQDQeeRdDkEjUiXVFjJUNKsAyVNnolIZ0AWVADgMsZM/cMFjNSt559IZmKUftaQZOtXY/ESUkubx5NbnbP0K2WL0kZ0yaQ6jou1qtbnQ7f2PoFxAwAAAAAAADMh/Kn54sHUMuDFjNNa9Nm9aIv7SJ7EDRRUiYjXVKiZpRJ1th2JXdlJuc8ekWm8vFUkM03XhwsZhSf7tZigICx79v62CRRn+QqlCRpRI23OlauLWyhr4aOEGqljKpjKWZev3Z2hjoCAAAAAAAAuJQ/Ra/vc2kOXMwoPtxNeCTSM7MKmpEjCHKtOrlUTHNOb4ukhBU7CfGgEzOXZxIzZXrmCSnku52dTnJGbHuTrZ2bBOqpX6KNKRIoRULYeGkcd7+p54KzutW4lWln62cRMwAAAAAAADBf9lvOvJJipkzPTMSCJ1js/tRxKckgSjB4aQ8rGnLJjmk3LWBs8mPUpGbq+ShLcsceS/l4VqqoGmUFjfsZZ/lxT6J4gsat5ywDgJ1lxheU3FoMEk6Fna3Xrd25xzoCAAAAAAAARIy2z6yXbU1/tw+leUXFjOJ9UXoml64oBwSPQrVFaY/kDJgQS4ZkO5MjeqbZbEuOPo8dAHxKJCyFOzZ/7uK39ljDbnrGuxf1erCgsTXsq58nyGwSaUgtbboo1Tq2OJ4zE06H/9y+dAkxAwAAAAAAAPvCqD7pR+d88sMiZpr0zKab/JB8qiNaxUkSgqDzfmbgb6qlacjmiaIFI2T0jJSlSsxc2Hrd2l7FjNQC733RfWaue6oWp877ifpl5diUsstL40QrW43rGE6Fb29fvPTqvdYQAAAAAAAAIEUlZ7bPrF8VkY/NqUqHRswo3t0+3aug6UtxuGma4B87bcrDSoROi07TglNJmXK7vHVp7dl5FbCePfM1V3IlkkGuoLGCJSWr9BwYr16pfUPFTWp1q2ZfKbdOh29vX7j0U/OqIQAAAAAAAIDHSO37gIg8vccqHUYxU6ZnrkohH0vKGG+fJ2j6WpSyKQ9vkG9mpaaOMDBDge0Q4MXJ81LMyCm5b2tt7al9KOXbpJDvt6+soBnp/XXtRlK3iDkyLJVi8USNHdqbElm9cit0N6+WVSvT7n9v34WYAQAAAAAAgP2nCGGygtHyzsbtIlKmaC7O8M2HUsxoVp6/9qwEuVuaW96VcbyjfN3sC84mk/eKUHSPE4mOSe6zz4dSqOMSUik0g20X5b6tu9c+v7dKpVne2bgiQT4Z16zw66ZeF81xXk1s3VKP8yAj1yqRtBRe2l5Zv32/aggAAAAAAACg6ciZhuWdjTJF8/6BlXqpTN1sn1mf99yaubPy/LVy1aJS0PxQVyQ4gkamkDTeo5j9qdd9aJEg6nmhZELRDrB9aOvcpc/sdx2XdzaekCDvmFrQyABJ4z2mjhuKrZ3E6Zx2APRS+L/ts+s/PIcyAQAAAAAAAAzClTMy/gFeioxHReRBEfEGoj5dreIj8kQ9MPZIsPL8tTUReUqCvKqToJGeFI34kkYkI2rs81mxKRnbStSImZX9FzMNyzsbn5Ug9w8WNJKQNJIQNZKop/c6hye11OvOqlxL4Xvbr13/semrAQAAAAAAADA7STmjqdud1ppd9QDhI4sraFIpGklIGklLmEjWOMdkcZIewRusO6rnsSzK5a3lS/sxYyZJ/W/iK22bmCdoZJikERkgajRD6lgk9hWhHlSstmoAcNjavnN9ddZ6AAAAAAAAAMzKIDlzHKkFzVfaFidxBI30S4VkusPZ10qbBKGwHzDP7fDcxbArhdy5tXxpz8tlz0ItaD4uQX7DFzCJJJKtT0rUiJI1mr5/sp26hclp7LyedkCwfGL7zvVH9q1QAAAAAAAAABlOrJyRyQyaz0mQVX/eTEbS5B7t8dNi58x4qxctyLe2li/dMd+KzEY7o8htA8sMUB4yU8apo3VY7aHW49j6iRFcoxBkQd6yfWb9b/apNAAAAAAAAAC9nGg5I2NBc3u1jHiQ91Q7UpJGnOSHTCcVOvtTIRpPKEiTlmnno/ze1tlLH+q/u4NjeWfjHhH5tAT5yWSKRgaKLVur3P5U+5J9HS3NLd+RQu46SvOSAAAAAAAA4Hhy4uVMw8rz1+6pW3RWXcmSmokyRCAMISkUghYL35BCXr919tKhFAp1m1OZonlPctaM9NRwr+mjZOqoreNNGcnbt8+s/9UMZwcAAAAAAACYO8gZw8rz166IyOMS5Ceqd4aImtRxQ4mEQrCtON+VQt65dfbS3+/Tbc+VeqWvD1WzaGSIhNnjAGVJLZethiiP5AdSyMe3z6z/7sFWAwAAAAAAACAPciZBLWkekSA/Xx0xtJ1JEkNsPfTwlDjxcUsK+bfyGrbOXro+7/s7CGpJUyZpfr1aGUumaGMaWkNJ1FHa1Mz3pJDHt8+sf+RwVQcAAAAAAABgDHKmh3po8IMi8i4ROdvbztT3XoM/K+WmFPKsiPx5Ob/lsLYvzcLyzsaDlaQpa2lFjX0+C902pu+KyJNlAmr7zPoLh7QkAAAAAAAAABXImSmohweXs2nKZbgfEJGfliA/Hp0hV9JusuM7IpWM+ZKI/MNRTchMy/LOxlpdx9eLyDkRuThVS1NXbP2vFPJfIvJVEflHEbnKkF8AAAAAAAA4SiBn5kCdrmmWtdbPNVfV8+vHKRUzD+phwmvqVPc4p32h3kpe3D6zfiJkFgAAAAAAABxjROT/AaGvnBGuLJbJAAAAAElFTkSuQmCC)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "oA_p5zIsoAJQ" - }, - "source": [ - "# SuperGradients quick start Semantic Segmentation\n", - "\n", - "In this tutorial we will train PPLiteSeg model on Supervisely semantic segmentation dataset\n", - "\n", - "The notebook is divided into 7 sections:\n", - "1. Experiment setup\n", - "2. Dataset definition\n", - "3. Architecture definition\n", - "4. Training setup\n", - "5. Training and Evaluation\n", - "6. Predict\n", - "7. Convert to ONNX\\TensorRT" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "GqH4VGMroWec" - }, - "source": [ - "#Install SG" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "Q8uA6AWEhHN6" - }, - "source": [ - "The cell below will install **super_gradients** which will automatically get all its dependencies." - ] - }, - { - "cell_type": "code", - "execution_count": 1, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "-mm-E4xRoNEm", - "outputId": "ce0b8873-49f3-44a4-f8f1-e53087c4f96b" - }, - "outputs": [ + "cells": [ { - "output_type": "stream", - "name": "stdout", - "text": [ - " Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", - " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m135.8/135.8 kB\u001B[0m \u001B[31m4.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m684.5/684.5 kB\u001B[0m \u001B[31m14.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.9/2.9 MB\u001B[0m \u001B[31m26.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m42.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m408.6/408.6 kB\u001B[0m \u001B[31m36.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m154.5/154.5 kB\u001B[0m \u001B[31m24.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.5/79.5 kB\u001B[0m \u001B[31m13.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m4.5/4.5 MB\u001B[0m \u001B[31m62.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m13.5/13.5 MB\u001B[0m \u001B[31m62.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m68.0/68.0 kB\u001B[0m \u001B[31m9.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", - " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m17.0/17.0 MB\u001B[0m \u001B[31m16.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m3.3/3.3 MB\u001B[0m \u001B[31m89.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.2/2.2 MB\u001B[0m \u001B[31m84.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m455.0/455.0 kB\u001B[0m \u001B[31m44.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.0/46.0 kB\u001B[0m \u001B[31m7.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m11.3/11.3 MB\u001B[0m \u001B[31m115.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.8/79.8 kB\u001B[0m \u001B[31m12.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m108.3/108.3 kB\u001B[0m \u001B[31m17.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m176.0/176.0 kB\u001B[0m \u001B[31m26.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m117.0/117.0 kB\u001B[0m \u001B[31m19.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m575.5/575.5 kB\u001B[0m \u001B[31m57.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m121.1/121.1 kB\u001B[0m \u001B[31m19.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m86.8/86.8 kB\u001B[0m \u001B[31m14.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m18.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m17.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.6/120.6 kB\u001B[0m \u001B[31m20.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m14.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m14.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m84.7/84.7 kB\u001B[0m \u001B[31m15.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m16.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m12.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.8/99.8 kB\u001B[0m \u001B[31m15.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m14.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m14.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m90.6/90.6 kB\u001B[0m \u001B[31m14.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m15.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m14.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m16.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m94.0/94.0 kB\u001B[0m \u001B[31m16.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m407.7/407.7 kB\u001B[0m \u001B[31m54.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m107.7/107.7 kB\u001B[0m \u001B[31m16.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m277.4/277.4 kB\u001B[0m \u001B[31m40.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m113.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m913.9/913.9 kB\u001B[0m \u001B[31m80.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m105.0/105.0 kB\u001B[0m \u001B[31m16.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.2/46.2 kB\u001B[0m \u001B[31m7.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m106.8/106.8 kB\u001B[0m \u001B[31m17.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m194.6/194.6 kB\u001B[0m \u001B[31m28.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m58.1/58.1 kB\u001B[0m \u001B[31m8.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Building wheel for super-gradients (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for pycocotools (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for termcolor (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for treelib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for coverage (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for antlr4-python3-runtime (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for stringcase (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for xhtml2pdf (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for svglib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", - "lida 0.0.10 requires fastapi, which is not installed.\n", - "lida 0.0.10 requires kaleido, which is not installed.\n", - "lida 0.0.10 requires python-multipart, which is not installed.\n", - "lida 0.0.10 requires uvicorn, which is not installed.\n", - "tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.23.0 which is incompatible.\u001B[0m\u001B[31m\n", - "\u001B[0m" - ] - } - ], - "source": [ - "! pip install -qq super-gradients==3.4.1\n", - "! pip install -qq prettyformatter\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "892xArqDsGsQ" - }, - "source": [ - "# 1. Experiment setup\n", - "We will initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", - "\n", - "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", - "\n", - "```\n", - "ckpt_root_dir\n", - "|─── experiment_name_1\n", - "│ ckpt_best.pth # Model checkpoint on best epoch\n", - "│ ckpt_latest.pth # Model checkpoint on last epoch\n", - "│ average_model.pth # Model checkpoint averaged over epochs\n", - "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", - "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", - "└─── experiment_name_2\n", - " ...\n", - "```\n", - "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", - " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "pl0WPz1HisFz" - }, - "source": [ - "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "cell_type": "markdown", + "metadata": { + "id": "HY_HuQbxn7X0" + }, + "source": [ + "![SG - Horizontal.png](data:image/png;base64,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)" + ] }, - "id": "HAff--HysJmP", - "outputId": "aa92f470-b3ce-448d-c1d9-39aa8b9cca72" - }, - "outputs": [ { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-08 10:54:04] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "oA_p5zIsoAJQ" + }, + "source": [ + "# SuperGradients quick start Semantic Segmentation\n", + "\n", + "In this tutorial we will train PPLiteSeg model on Supervisely semantic segmentation dataset\n", + "\n", + "The notebook is divided into 7 sections:\n", + "1. Experiment setup\n", + "2. Dataset definition\n", + "3. Architecture definition\n", + "4. Training setup\n", + "5. Training and Evaluation\n", + "6. Predict\n", + "7. Convert to ONNX\\TensorRT" + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "The console stream is logged into /root/sg_logs/console.log\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "GqH4VGMroWec" + }, + "source": [ + "#Install SG" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-08 10:54:04] WARNING - __init__.py - Failed to import pytorch_quantization\n", - "[2023-11-08 10:54:04] INFO - utils.py - NumExpr defaulting to 2 threads.\n", - "[2023-11-08 10:54:17] WARNING - calibrator.py - Failed to import pytorch_quantization\n", - "[2023-11-08 10:54:17] WARNING - export.py - Failed to import pytorch_quantization\n", - "[2023-11-08 10:54:17] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n", - "[2023-11-08 10:54:17] INFO - env_sanity_check.py - Library check is not supported when super_gradients installed through \"git+https://github.com/...\" command\n" - ] - } - ], - "source": [ - "from super_gradients import Trainer\n", - "\n", - "CHECKPOINT_DIR = '/home/notebook_ckpts/'\n", - "trainer = Trainer(experiment_name=\"segmentation_quick_start\", ckpt_root_dir=CHECKPOINT_DIR)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "dwVMY4gMjQSL" - }, - "source": [ - "# 2. Dataset definition\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "fpIWhnR9j2rm" - }, - "source": [ - "\n", - "For the sake of this presentation, we'll use **Supervisely** semantic segmentation dataset." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "ZACgRb-qjzDJ" - }, - "source": [ - "SG trainer is fully compatible with PyTorch data loaders, so you can definitely use your own data for the experiment below if you prefer." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "6ulV6Hpao3IN" - }, - "source": [ - "## 2.A. Download data\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "mVwslNv-j-2C" - }, - "source": [ - "Feel free to change the download path by editing SUPERVISELY_DATASET_DOWNLOAD_PATH" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": { - "id": "dfR18Rmbo00y", - "colab": { - "base_uri": "https://localhost:8080/" + "cell_type": "markdown", + "metadata": { + "id": "Q8uA6AWEhHN6" + }, + "source": [ + "The cell below will install **super_gradients** which will automatically get all its dependencies." + ] }, - "outputId": "8ed988c8-190a-4637-c0a5-ebcb173e7329" - }, - "outputs": [ { - "output_type": "stream", - "name": "stdout", - "text": [ - "Downloading and extracting supervisely dataset to: /home/data\n", - "/home/data\n", - "--2023-11-08 10:54:17-- https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", - "Resolving deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)... 52.216.49.177, 52.217.138.241, 52.217.171.137, ...\n", - "Connecting to deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)|52.216.49.177|:443... connected.\n", - "HTTP request sent, awaiting response... 200 OK\n", - "Length: 3564001012 (3.3G) [application/zip]\n", - "Saving to: ‘supervisely-persons.zip’\n", - "\n", - "supervisely-persons 100%[===================>] 3.32G 41.2MB/s in 72s \n", - "\n", - "2023-11-08 10:55:30 (47.0 MB/s) - ‘supervisely-persons.zip’ saved [3564001012/3564001012]\n", - "\n" - ] - } - ], - "source": [ - "import os\n", - "\n", - "SUPERVISELY_DATASET_DOWNLOAD_PATH=\"/home/data\"\n", - "\n", - "supervisely_dataset_dir_path = SUPERVISELY_DATASET_DOWNLOAD_PATH + os.path.sep + 'supervisely-persons'\n", - "\n", - "if os.path.isdir(supervisely_dataset_dir_path):\n", - " print('supervisely dataset already downloaded...')\n", - "else:\n", - " print('Downloading and extracting supervisely dataset to: ' + SUPERVISELY_DATASET_DOWNLOAD_PATH)\n", - " ! mkdir $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", - " %cd $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", - " ! wget https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", - " ! unzip --qq supervisely-persons.zip" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "collapsed": false, - "id": "V9ZcklupX8Qx" - }, - "source": [ - "## 2.B. Create data loaders\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "3Mk_YixjlEhj" - }, - "source": [ - "The dataloaders are initiated with the default parameters defined in the [yaml](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/recipes/dataset_params/supervisely_persons_dataset_params.yaml)\n", - "file. Parameters as batch_size, transforms, root_dir and others can be overridden by passing as `dataset_params` and\n", - "`dataloader_params`, as implemented bellow." - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "metadata": { - "id": "S3BzMRhSX8Qx" - }, - "outputs": [], - "source": [ - "from super_gradients.training import dataloaders\n", - "root_dir = supervisely_dataset_dir_path\n", - "batch_size = 8\n", - "\n", - "train_loader = dataloaders.supervisely_persons_train(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})\n", - "valid_loader = dataloaders.supervisely_persons_val(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "6dHIwvs46-dk" - }, - "source": [ - "As you can see, we didn't have to pass many parameters into the dataloaders construction. That's because defaults are pre-defined for your convenience, and you might be curious to know what they are. Let's print them and see which resolution and transformations are defined." - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": { - "id": "76tzhKxi6aS-" - }, - "outputs": [], - "source": [ - "from prettyformatter import pprint\n", - "\n", - "print('Dataloader parameters:')\n", - "pprint(train_loader.dataloader_params)\n", - "print('Dataset parameters')\n", - "pprint(train_loader.dataset.dataset_params)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "I4QEOkKyy93R" - }, - "source": [ - "We can take a look at some images from the dataset." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "l5GcDAg_pUGJ" - }, - "source": [ - "# 3. Architecture definition\n", - "\n" - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "metadata": { - "id": "xXPMJQCJzmb4" - }, - "outputs": [], - "source": [ - "from super_gradients.training import models\n", - "from super_gradients.common.object_names import Models\n", - "\n", - "model = models.get(model_name=Models.PP_LITE_T_SEG,\n", - " arch_params={\"use_aux_heads\": False},\n", - " num_classes=1)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "fU8orO7wlwIK" - }, - "source": [ - "SG includes implementations of many different architectures for semantic segmentation tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "-oGSU3V8lqcm" - }, - "source": [ - "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", - "and extra Auxiliary heads aren't used for training.\n", - "\n", - "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "X-_dBewgr1dG" - }, - "source": [ - "# 4. Training setup\n", - "\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "H1Rll8Orl-Dy" - }, - "source": [ - "\n", - "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n", - "\n", - "We will be using an average of BCE and Dice loss for segmentation, with different learning rates for the replaced segmentation head layer, and the rest of the network- this is controlled by the `multiply_head_lr` parameter which is the multiplication factor of the learning rate for the newly replaced layer.\n", - "\n", - "As our `metric_to_watch`, we will be monitoring the `target_IOU` which is one of the components of `BinaryIOU` torchmetrics object (the other components are `mean_IOU` which is the mean of the background and target IOUs, and `background_IOU`)." - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "metadata": { - "id": "NShu3zLgr5qD" - }, - "outputs": [], - "source": [ - "from super_gradients.training.metrics.segmentation_metrics import BinaryIOU\n", - "from super_gradients.training.utils.callbacks import BinarySegmentationVisualizationCallback, Phase\n", - "\n", - "train_params = {\"max_epochs\": 15,\n", - " \"lr_mode\": \"cosine\",\n", - " \"initial_lr\": 0.01,\n", - " \"lr_warmup_epochs\": 5,\n", - " \"multiply_head_lr\": 10,\n", - " \"optimizer\": \"SGD\",\n", - " \"loss\": \"bce_dice_loss\",\n", - " \"ema\": True,\n", - " \"zero_weight_decay_on_bias_and_bn\": True,\n", - " \"average_best_models\": True,\n", - " \"metric_to_watch\": \"target_IOU\",\n", - " \"greater_metric_to_watch_is_better\": True,\n", - " \"train_metrics_list\": [BinaryIOU()],\n", - " \"valid_metrics_list\": [BinaryIOU()],\n", - " \"loss_logging_items_names\": [\"loss\"]\n", - " }" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "qTECVyhcs506" - }, - "source": [ - "# 5. Training and evaluation\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "S1K5MU2kmmDb" - }, - "source": [ - "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", - "\n", - "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "metadata": { - "id": "u6roEj9ktFTi", - "colab": { - "base_uri": "https://localhost:8080/" + "cell_type": "code", + "execution_count": 1, + "metadata": { + "id": "-mm-E4xRoNEm" + }, + "outputs": [], + "source": [ + "! pip install -qq super-gradients==3.4.1" + ] }, - "outputId": "9b530548-7596-4db2-f1cf-2c461e9cb5bd" - }, - "outputs": [ { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "Dataloader parameters:\n", - "{\"batch_size\": 8, \"shuffle\": True, \"drop_last\": True}\n", - "Dataset parameters\n", - "{'root_dir': '/home/data/supervisely-persons', 'list_file': 'train.csv', 'cache_labels': False, 'cache_images': False, 'transforms': [{'SegRandomRescale': {'scales': [0.25, 1.0]}}, {'SegColorJitter': {'brightness': 0.5, 'contrast': 0.5, 'saturation': 0.5}}, {'SegRandomFlip': {'prob': 0.5}}, {'SegPadShortToCropSize': {'crop_size': [320, 480], 'fill_mask': 0}}, {'SegCropImageAndMask': {'crop_size': [320, 480], 'mode': 'random'}}]}\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "892xArqDsGsQ" + }, + "source": [ + "# 1. Experiment setup\n", + "We will initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", + "\n", + "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", + "\n", + "```\n", + "ckpt_root_dir\n", + "|─── experiment_name_1\n", + "│ ckpt_best.pth # Model checkpoint on best epoch\n", + "│ ckpt_latest.pth # Model checkpoint on last epoch\n", + "│ average_model.pth # Model checkpoint averaged over epochs\n", + "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", + "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", + "└─── experiment_name_2\n", + " ...\n", + "```\n", + "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", + " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`." + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "[2023-11-08 10:56:13] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231108_105613_531244`\n", - "[2023-11-08 10:56:13] INFO - sg_trainer.py - Checkpoints directory: /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244\n", - "/usr/local/lib/python3.10/dist-packages/super_gradients/common/registry/registry.py:72: DeprecationWarning: Object name `bce_dice_loss` is now deprecated. Please replace it with `BCEDiceLoss`.\n", - " warnings.warn(f\"Object name `{name}` is now deprecated. Please replace it with `{deprecated_names[name]}`.\", DeprecationWarning)\n", - "[2023-11-08 10:56:13] INFO - sg_trainer.py - Using EMA with params {}\n", - "[2023-11-08 10:56:13] WARNING - ema.py - Parameter `decay` is not specified for EMA params. Please specify `decay` parameter explicitly in your config:\n", - "ema: True\n", - "ema_params: \n", - " decay: 0.9999\n", - " decay_type: exp\n", - " beta: 15\n", - "Will default to decay: 0.9999\n", - "In the next major release of SG this warning will become an error.\n", - "[2023-11-08 10:56:13] WARNING - ema.py - Parameter decay_type is not specified for EMA model. Please specify decay_type parameter explicitly in your config:\n", - "ema: True\n", - "ema_params: \n", - " decay: 0.9999\n", - " decay_type: constant|exp|threshold\n", - "Will default to `exp` decay with beta = 15\n", - "In the next major release of SG this warning will become an error.\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "pl0WPz1HisFz" + }, + "source": [ + "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "The console stream is now moved to /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/console_Nov08_10_56_13.txt\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "[2023-11-08 10:56:14] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", - " - Mode: Single GPU\n", - " - Number of GPUs: 1 (1 available on the machine)\n", - " - Full dataset size: 2477 (len(train_set))\n", - " - Batch size per GPU: 8 (batch_size)\n", - " - Batch Accumulate: 1 (batch_accumulate)\n", - " - Total batch size: 8 (num_gpus * batch_size)\n", - " - Effective Batch size: 8 (num_gpus * batch_size * batch_accumulate)\n", - " - Iterations per epoch: 309 (len(train_loader))\n", - " - Gradient updates per epoch: 309 (len(train_loader) / batch_accumulate)\n", - "\n", - "[2023-11-08 10:56:14] INFO - sg_trainer.py - Started training for 15 epochs (0/14)\n", - "\n", - "Train epoch 0: 100%|██████████| 309/309 [02:08<00:00, 2.41it/s, BCEDiceLoss=0.405, background_IOU=0.539, gpu_mem=1.14, mean_IOU=0.604, target_IOU=0.669]\n", - "Validating: 100%|██████████| 65/65 [00:16<00:00, 3.86it/s]\n", - "[2023-11-08 10:58:39] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 10:58:39] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.6919947266578674\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 0\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.4047\n", - "│ ├── Target_iou = 0.6685\n", - "│ ├── Background_iou = 0.5393\n", - "│ └── Mean_iou = 0.6039\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3857\n", - " ├── Target_iou = 0.692\n", - " ├── Background_iou = 0.4666\n", - " └── Mean_iou = 0.5793\n", - "\n", - "===========================================================\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 1: 100%|██████████| 309/309 [01:57<00:00, 2.63it/s, BCEDiceLoss=0.338, background_IOU=0.606, gpu_mem=1.14, mean_IOU=0.663, target_IOU=0.719]\n", - "Validating epoch 1: 100%|██████████| 65/65 [00:16<00:00, 4.02it/s]\n", - "[2023-11-08 11:00:56] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:00:56] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7162820100784302\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 1\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.3375\n", - "│ │ ├── Epoch N-1 = 0.4047 (\u001B[32m↘ -0.0672\u001B[0m)\n", - "│ │ └── Best until now = 0.4047 (\u001B[32m↘ -0.0672\u001B[0m)\n", - "│ ├── Target_iou = 0.7191\n", - "│ │ ├── Epoch N-1 = 0.6685 (\u001B[32m↗ 0.0506\u001B[0m)\n", - "│ │ └── Best until now = 0.6685 (\u001B[32m↗ 0.0506\u001B[0m)\n", - "│ ├── Background_iou = 0.6064\n", - "│ │ ├── Epoch N-1 = 0.5393 (\u001B[32m↗ 0.0671\u001B[0m)\n", - "│ │ └── Best until now = 0.5393 (\u001B[32m↗ 0.0671\u001B[0m)\n", - "│ └── Mean_iou = 0.6628\n", - "│ ├── Epoch N-1 = 0.6039 (\u001B[32m↗ 0.0588\u001B[0m)\n", - "│ └── Best until now = 0.6039 (\u001B[32m↗ 0.0588\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.388\n", - " │ ├── Epoch N-1 = 0.3857 (\u001B[31m↗ 0.0023\u001B[0m)\n", - " │ └── Best until now = 0.3857 (\u001B[31m↗ 0.0023\u001B[0m)\n", - " ├── Target_iou = 0.7163\n", - " │ ├── Epoch N-1 = 0.692 (\u001B[32m↗ 0.0243\u001B[0m)\n", - " │ └── Best until now = 0.692 (\u001B[32m↗ 0.0243\u001B[0m)\n", - " ├── Background_iou = 0.3871\n", - " │ ├── Epoch N-1 = 0.4666 (\u001B[31m↘ -0.0795\u001B[0m)\n", - " │ └── Best until now = 0.4666 (\u001B[31m↘ -0.0795\u001B[0m)\n", - " └── Mean_iou = 0.5517\n", - " ├── Epoch N-1 = 0.5793 (\u001B[31m↘ -0.0276\u001B[0m)\n", - " └── Best until now = 0.5793 (\u001B[31m↘ -0.0276\u001B[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 2: 100%|██████████| 309/309 [01:57<00:00, 2.64it/s, BCEDiceLoss=0.326, background_IOU=0.629, gpu_mem=1.14, mean_IOU=0.679, target_IOU=0.729]\n", - "Validating epoch 2: 100%|██████████| 65/65 [00:16<00:00, 3.94it/s]\n", - "[2023-11-08 11:03:11] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:03:11] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7318407297134399\n" - ] + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "HAff--HysJmP", + "outputId": "63e96426-a29b-4cdc-9a72-60da27d6aaa7" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is logged into /root/sg_logs/console.log\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 11:11:11] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n", + "[2023-11-13 11:11:11] WARNING - __init__.py - Failed to import pytorch_quantization\n", + "[2023-11-13 11:11:11] INFO - utils.py - NumExpr defaulting to 2 threads.\n", + "[2023-11-13 11:11:23] WARNING - calibrator.py - Failed to import pytorch_quantization\n", + "[2023-11-13 11:11:23] WARNING - export.py - Failed to import pytorch_quantization\n", + "[2023-11-13 11:11:23] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n" + ] + } + ], + "source": [ + "from super_gradients import Trainer\n", + "\n", + "CHECKPOINT_DIR = './notebook_ckpts/'\n", + "trainer = Trainer(experiment_name=\"segmentation_quick_start\", ckpt_root_dir=CHECKPOINT_DIR)" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 2\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.3256\n", - "│ │ ├── Epoch N-1 = 0.3375 (\u001B[32m↘ -0.0119\u001B[0m)\n", - "│ │ └── Best until now = 0.3375 (\u001B[32m↘ -0.0119\u001B[0m)\n", - "│ ├── Target_iou = 0.7294\n", - "│ │ ├── Epoch N-1 = 0.7191 (\u001B[32m↗ 0.0102\u001B[0m)\n", - "│ │ └── Best until now = 0.7191 (\u001B[32m↗ 0.0102\u001B[0m)\n", - "│ ├── Background_iou = 0.6291\n", - "│ │ ├── Epoch N-1 = 0.6064 (\u001B[32m↗ 0.0227\u001B[0m)\n", - "│ │ └── Best until now = 0.6064 (\u001B[32m↗ 0.0227\u001B[0m)\n", - "│ └── Mean_iou = 0.6792\n", - "│ ├── Epoch N-1 = 0.6628 (\u001B[32m↗ 0.0164\u001B[0m)\n", - "│ └── Best until now = 0.6628 (\u001B[32m↗ 0.0164\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3604\n", - " │ ├── Epoch N-1 = 0.388 (\u001B[32m↘ -0.0275\u001B[0m)\n", - " │ └── Best until now = 0.3857 (\u001B[32m↘ -0.0252\u001B[0m)\n", - " ├── Target_iou = 0.7318\n", - " │ ├── Epoch N-1 = 0.7163 (\u001B[32m↗ 0.0156\u001B[0m)\n", - " │ └── Best until now = 0.7163 (\u001B[32m↗ 0.0156\u001B[0m)\n", - " ├── Background_iou = 0.4626\n", - " │ ├── Epoch N-1 = 0.3871 (\u001B[32m↗ 0.0755\u001B[0m)\n", - " │ └── Best until now = 0.4666 (\u001B[31m↘ -0.004\u001B[0m)\n", - " └── Mean_iou = 0.5972\n", - " ├── Epoch N-1 = 0.5517 (\u001B[32m↗ 0.0455\u001B[0m)\n", - " └── Best until now = 0.5793 (\u001B[32m↗ 0.0179\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "dwVMY4gMjQSL" + }, + "source": [ + "# 2. Dataset definition\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 3: 100%|██████████| 309/309 [01:57<00:00, 2.63it/s, BCEDiceLoss=0.306, background_IOU=0.642, gpu_mem=1.14, mean_IOU=0.695, target_IOU=0.748]\n", - "Validating epoch 3: 100%|██████████| 65/65 [00:17<00:00, 3.80it/s]\n", - "[2023-11-08 11:05:27] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:05:27] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7394765019416809\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "fpIWhnR9j2rm" + }, + "source": [ + "\n", + "For the sake of this presentation, we'll use **Supervisely** semantic segmentation dataset." + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 3\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.306\n", - "│ │ ├── Epoch N-1 = 0.3256 (\u001B[32m↘ -0.0196\u001B[0m)\n", - "│ │ └── Best until now = 0.3256 (\u001B[32m↘ -0.0196\u001B[0m)\n", - "│ ├── Target_iou = 0.7483\n", - "│ │ ├── Epoch N-1 = 0.7294 (\u001B[32m↗ 0.0189\u001B[0m)\n", - "│ │ └── Best until now = 0.7294 (\u001B[32m↗ 0.0189\u001B[0m)\n", - "│ ├── Background_iou = 0.6421\n", - "│ │ ├── Epoch N-1 = 0.6291 (\u001B[32m↗ 0.0131\u001B[0m)\n", - "│ │ └── Best until now = 0.6291 (\u001B[32m↗ 0.0131\u001B[0m)\n", - "│ └── Mean_iou = 0.6952\n", - "│ ├── Epoch N-1 = 0.6792 (\u001B[32m↗ 0.016\u001B[0m)\n", - "│ └── Best until now = 0.6792 (\u001B[32m↗ 0.016\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3468\n", - " │ ├── Epoch N-1 = 0.3604 (\u001B[32m↘ -0.0136\u001B[0m)\n", - " │ └── Best until now = 0.3604 (\u001B[32m↘ -0.0136\u001B[0m)\n", - " ├── Target_iou = 0.7395\n", - " │ ├── Epoch N-1 = 0.7318 (\u001B[32m↗ 0.0076\u001B[0m)\n", - " │ └── Best until now = 0.7318 (\u001B[32m↗ 0.0076\u001B[0m)\n", - " ├── Background_iou = 0.4784\n", - " │ ├── Epoch N-1 = 0.4626 (\u001B[32m↗ 0.0158\u001B[0m)\n", - " │ └── Best until now = 0.4666 (\u001B[32m↗ 0.0118\u001B[0m)\n", - " └── Mean_iou = 0.6089\n", - " ├── Epoch N-1 = 0.5972 (\u001B[32m↗ 0.0117\u001B[0m)\n", - " └── Best until now = 0.5972 (\u001B[32m↗ 0.0117\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "ZACgRb-qjzDJ" + }, + "source": [ + "SG trainer is fully compatible with PyTorch data loaders, so you can definitely use your own data for the experiment below if you prefer." + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 4: 100%|██████████| 309/309 [01:58<00:00, 2.61it/s, BCEDiceLoss=0.29, background_IOU=0.668, gpu_mem=1.14, mean_IOU=0.714, target_IOU=0.76]\n", - "Validating epoch 4: 100%|██████████| 65/65 [00:16<00:00, 3.86it/s]\n", - "[2023-11-08 11:07:45] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:07:45] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7402159571647644\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "6ulV6Hpao3IN" + }, + "source": [ + "## 2.A. Download data\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 4\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2896\n", - "│ │ ├── Epoch N-1 = 0.306 (\u001B[32m↘ -0.0164\u001B[0m)\n", - "│ │ └── Best until now = 0.306 (\u001B[32m↘ -0.0164\u001B[0m)\n", - "│ ├── Target_iou = 0.76\n", - "│ │ ├── Epoch N-1 = 0.7483 (\u001B[32m↗ 0.0117\u001B[0m)\n", - "│ │ └── Best until now = 0.7483 (\u001B[32m↗ 0.0117\u001B[0m)\n", - "│ ├── Background_iou = 0.668\n", - "│ │ ├── Epoch N-1 = 0.6421 (\u001B[32m↗ 0.0259\u001B[0m)\n", - "│ │ └── Best until now = 0.6421 (\u001B[32m↗ 0.0259\u001B[0m)\n", - "│ └── Mean_iou = 0.714\n", - "│ ├── Epoch N-1 = 0.6952 (\u001B[32m↗ 0.0188\u001B[0m)\n", - "│ └── Best until now = 0.6952 (\u001B[32m↗ 0.0188\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3405\n", - " │ ├── Epoch N-1 = 0.3468 (\u001B[32m↘ -0.0063\u001B[0m)\n", - " │ └── Best until now = 0.3468 (\u001B[32m↘ -0.0063\u001B[0m)\n", - " ├── Target_iou = 0.7402\n", - " │ ├── Epoch N-1 = 0.7395 (\u001B[32m↗ 0.0007\u001B[0m)\n", - " │ └── Best until now = 0.7395 (\u001B[32m↗ 0.0007\u001B[0m)\n", - " ├── Background_iou = 0.4861\n", - " │ ├── Epoch N-1 = 0.4784 (\u001B[32m↗ 0.0077\u001B[0m)\n", - " │ └── Best until now = 0.4784 (\u001B[32m↗ 0.0077\u001B[0m)\n", - " └── Mean_iou = 0.6131\n", - " ├── Epoch N-1 = 0.6089 (\u001B[32m↗ 0.0042\u001B[0m)\n", - " └── Best until now = 0.6089 (\u001B[32m↗ 0.0042\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "mVwslNv-j-2C" + }, + "source": [ + "Feel free to change the download path by editing SUPERVISELY_DATASET_DOWNLOAD_PATH" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 5: 100%|██████████| 309/309 [01:59<00:00, 2.58it/s, BCEDiceLoss=0.289, background_IOU=0.663, gpu_mem=1.14, mean_IOU=0.712, target_IOU=0.761]\n", - "Validating epoch 5: 100%|██████████| 65/65 [00:17<00:00, 3.71it/s]\n", - "[2023-11-08 11:10:03] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:10:03] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7486929297447205\n" - ] + "cell_type": "code", + "execution_count": 3, + "metadata": { + "id": "dfR18Rmbo00y" + }, + "outputs": [], + "source": [ + "import os\n", + "\n", + "SUPERVISELY_DATASET_DOWNLOAD_PATH=os.path.join(os.getcwd(),\"data\")\n", + "\n", + "supervisely_dataset_dir_path = os.path.join(SUPERVISELY_DATASET_DOWNLOAD_PATH, 'supervisely-persons')\n", + "\n", + "if os.path.isdir(supervisely_dataset_dir_path):\n", + " print('supervisely dataset already downloaded...')\n", + "else:\n", + " print('Downloading and extracting supervisely dataset to: ' + SUPERVISELY_DATASET_DOWNLOAD_PATH)\n", + " ! mkdir $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " %cd $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " ! wget https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", + " ! unzip --qq supervisely-persons.zip" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 5\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2888\n", - "│ │ ├── Epoch N-1 = 0.2896 (\u001B[32m↘ -0.0008\u001B[0m)\n", - "│ │ └── Best until now = 0.2896 (\u001B[32m↘ -0.0008\u001B[0m)\n", - "│ ├── Target_iou = 0.7608\n", - "│ │ ├── Epoch N-1 = 0.76 (\u001B[32m↗ 0.0008\u001B[0m)\n", - "│ │ └── Best until now = 0.76 (\u001B[32m↗ 0.0008\u001B[0m)\n", - "│ ├── Background_iou = 0.6632\n", - "│ │ ├── Epoch N-1 = 0.668 (\u001B[31m↘ -0.0048\u001B[0m)\n", - "│ │ └── Best until now = 0.668 (\u001B[31m↘ -0.0048\u001B[0m)\n", - "│ └── Mean_iou = 0.712\n", - "│ ├── Epoch N-1 = 0.714 (\u001B[31m↘ -0.002\u001B[0m)\n", - "│ └── Best until now = 0.714 (\u001B[31m↘ -0.002\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3203\n", - " │ ├── Epoch N-1 = 0.3405 (\u001B[32m↘ -0.0202\u001B[0m)\n", - " │ └── Best until now = 0.3405 (\u001B[32m↘ -0.0202\u001B[0m)\n", - " ├── Target_iou = 0.7487\n", - " │ ├── Epoch N-1 = 0.7402 (\u001B[32m↗ 0.0085\u001B[0m)\n", - " │ └── Best until now = 0.7402 (\u001B[32m↗ 0.0085\u001B[0m)\n", - " ├── Background_iou = 0.5274\n", - " │ ├── Epoch N-1 = 0.4861 (\u001B[32m↗ 0.0413\u001B[0m)\n", - " │ └── Best until now = 0.4861 (\u001B[32m↗ 0.0413\u001B[0m)\n", - " └── Mean_iou = 0.638\n", - " ├── Epoch N-1 = 0.6131 (\u001B[32m↗ 0.0249\u001B[0m)\n", - " └── Best until now = 0.6131 (\u001B[32m↗ 0.0249\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "collapsed": false, + "id": "V9ZcklupX8Qx" + }, + "source": [ + "## 2.B. Create data loaders\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 6: 100%|██████████| 309/309 [01:58<00:00, 2.61it/s, BCEDiceLoss=0.273, background_IOU=0.686, gpu_mem=1.14, mean_IOU=0.728, target_IOU=0.77]\n", - "Validating epoch 6: 100%|██████████| 65/65 [00:16<00:00, 3.93it/s]\n", - "[2023-11-08 11:12:20] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:12:20] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.751246988773346\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "3Mk_YixjlEhj" + }, + "source": [ + "The dataloaders are initiated with the default parameters defined in the [yaml](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/recipes/dataset_params/supervisely_persons_dataset_params.yaml)\n", + "file. Parameters as batch_size, transforms, root_dir and others can be overridden by passing as `dataset_params` and\n", + "`dataloader_params`, as implemented bellow." + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 6\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2731\n", - "│ │ ├── Epoch N-1 = 0.2888 (\u001B[32m↘ -0.0158\u001B[0m)\n", - "│ │ └── Best until now = 0.2888 (\u001B[32m↘ -0.0158\u001B[0m)\n", - "│ ├── Target_iou = 0.7704\n", - "│ │ ├── Epoch N-1 = 0.7608 (\u001B[32m↗ 0.0096\u001B[0m)\n", - "│ │ └── Best until now = 0.7608 (\u001B[32m↗ 0.0096\u001B[0m)\n", - "│ ├── Background_iou = 0.686\n", - "│ │ ├── Epoch N-1 = 0.6632 (\u001B[32m↗ 0.0228\u001B[0m)\n", - "│ │ └── Best until now = 0.668 (\u001B[32m↗ 0.018\u001B[0m)\n", - "│ └── Mean_iou = 0.7282\n", - "│ ├── Epoch N-1 = 0.712 (\u001B[32m↗ 0.0162\u001B[0m)\n", - "│ └── Best until now = 0.714 (\u001B[32m↗ 0.0142\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3158\n", - " │ ├── Epoch N-1 = 0.3203 (\u001B[32m↘ -0.0045\u001B[0m)\n", - " │ └── Best until now = 0.3203 (\u001B[32m↘ -0.0045\u001B[0m)\n", - " ├── Target_iou = 0.7512\n", - " │ ├── Epoch N-1 = 0.7487 (\u001B[32m↗ 0.0026\u001B[0m)\n", - " │ └── Best until now = 0.7487 (\u001B[32m↗ 0.0026\u001B[0m)\n", - " ├── Background_iou = 0.5394\n", - " │ ├── Epoch N-1 = 0.5274 (\u001B[32m↗ 0.012\u001B[0m)\n", - " │ └── Best until now = 0.5274 (\u001B[32m↗ 0.012\u001B[0m)\n", - " └── Mean_iou = 0.6453\n", - " ├── Epoch N-1 = 0.638 (\u001B[32m↗ 0.0073\u001B[0m)\n", - " └── Best until now = 0.638 (\u001B[32m↗ 0.0073\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "S3BzMRhSX8Qx", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "87b5092d-fe93-4c0a-8b2e-febe215b52bd" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "supervisely dataset already downloaded...\n" + ] + } + ], + "source": [ + "from super_gradients.training import dataloaders\n", + "root_dir = supervisely_dataset_dir_path\n", + "batch_size = 8\n", + "\n", + "train_loader = dataloaders.supervisely_persons_train(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})\n", + "valid_loader = dataloaders.supervisely_persons_val(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 7: 100%|██████████| 309/309 [01:58<00:00, 2.62it/s, BCEDiceLoss=0.26, background_IOU=0.699, gpu_mem=1.14, mean_IOU=0.739, target_IOU=0.78]\n", - "Validating epoch 7: 100%|██████████| 65/65 [00:17<00:00, 3.79it/s]\n", - "[2023-11-08 11:14:37] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:14:37] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7533503174781799\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "6dHIwvs46-dk" + }, + "source": [ + "As you can see, we didn't have to pass many parameters into the dataloaders construction. That's because defaults are pre-defined for your convenience, and you might be curious to know what they are. Let's print them and see which resolution and transformations are defined." + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 7\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.26\n", - "│ │ ├── Epoch N-1 = 0.2731 (\u001B[32m↘ -0.013\u001B[0m)\n", - "│ │ └── Best until now = 0.2731 (\u001B[32m↘ -0.013\u001B[0m)\n", - "│ ├── Target_iou = 0.7799\n", - "│ │ ├── Epoch N-1 = 0.7704 (\u001B[32m↗ 0.0095\u001B[0m)\n", - "│ │ └── Best until now = 0.7704 (\u001B[32m↗ 0.0095\u001B[0m)\n", - "│ ├── Background_iou = 0.6987\n", - "│ │ ├── Epoch N-1 = 0.686 (\u001B[32m↗ 0.0127\u001B[0m)\n", - "│ │ └── Best until now = 0.686 (\u001B[32m↗ 0.0127\u001B[0m)\n", - "│ └── Mean_iou = 0.7393\n", - "│ ├── Epoch N-1 = 0.7282 (\u001B[32m↗ 0.0111\u001B[0m)\n", - "│ └── Best until now = 0.7282 (\u001B[32m↗ 0.0111\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3143\n", - " │ ├── Epoch N-1 = 0.3158 (\u001B[32m↘ -0.0015\u001B[0m)\n", - " │ └── Best until now = 0.3158 (\u001B[32m↘ -0.0015\u001B[0m)\n", - " ├── Target_iou = 0.7534\n", - " │ ├── Epoch N-1 = 0.7512 (\u001B[32m↗ 0.0021\u001B[0m)\n", - " │ └── Best until now = 0.7512 (\u001B[32m↗ 0.0021\u001B[0m)\n", - " ├── Background_iou = 0.5443\n", - " │ ├── Epoch N-1 = 0.5394 (\u001B[32m↗ 0.0049\u001B[0m)\n", - " │ └── Best until now = 0.5394 (\u001B[32m↗ 0.0049\u001B[0m)\n", - " └── Mean_iou = 0.6488\n", - " ├── Epoch N-1 = 0.6453 (\u001B[32m↗ 0.0035\u001B[0m)\n", - " └── Best until now = 0.6453 (\u001B[32m↗ 0.0035\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "code", + "execution_count": 7, + "metadata": { + "id": "76tzhKxi6aS-", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "3b5c8f34-673c-4f4c-d243-80e82c347f3d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Dataloader parameters:\n", + "{'batch_size': 8, 'shuffle': True, 'drop_last': True}\n", + "Dataset parameters\n", + "{'root_dir': '/content/data/supervisely-persons', 'list_file': 'train.csv', 'cache_labels': False, 'cache_images': False, 'transforms': [{'SegRandomRescale': {'scales': [0.25, 1.0]}}, {'SegColorJitter': {'brightness': 0.5, 'contrast': 0.5, 'saturation': 0.5}}, {'SegRandomFlip': {'prob': 0.5}}, {'SegPadShortToCropSize': {'crop_size': [320, 480], 'fill_mask': 0}}, {'SegCropImageAndMask': {'crop_size': [320, 480], 'mode': 'random'}}]}\n" + ] + } + ], + "source": [ + "print('Dataloader parameters:')\n", + "print(train_loader.dataloader_params)\n", + "print('Dataset parameters')\n", + "print(train_loader.dataset.dataset_params)" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 8: 100%|██████████| 309/309 [01:58<00:00, 2.60it/s, BCEDiceLoss=0.252, background_IOU=0.711, gpu_mem=1.14, mean_IOU=0.749, target_IOU=0.786]\n", - "Validating epoch 8: 100%|██████████| 65/65 [00:16<00:00, 3.92it/s]\n", - "[2023-11-08 11:16:56] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:16:56] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7535856366157532\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "l5GcDAg_pUGJ" + }, + "source": [ + "# 3. Architecture definition\n", + "\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 8\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2517\n", - "│ │ ├── Epoch N-1 = 0.26 (\u001B[32m↘ -0.0084\u001B[0m)\n", - "│ │ └── Best until now = 0.26 (\u001B[32m↘ -0.0084\u001B[0m)\n", - "│ ├── Target_iou = 0.7861\n", - "│ │ ├── Epoch N-1 = 0.7799 (\u001B[32m↗ 0.0062\u001B[0m)\n", - "│ │ └── Best until now = 0.7799 (\u001B[32m↗ 0.0062\u001B[0m)\n", - "│ ├── Background_iou = 0.7112\n", - "│ │ ├── Epoch N-1 = 0.6987 (\u001B[32m↗ 0.0125\u001B[0m)\n", - "│ │ └── Best until now = 0.6987 (\u001B[32m↗ 0.0125\u001B[0m)\n", - "│ └── Mean_iou = 0.7487\n", - "│ ├── Epoch N-1 = 0.7393 (\u001B[32m↗ 0.0093\u001B[0m)\n", - "│ └── Best until now = 0.7393 (\u001B[32m↗ 0.0093\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.314\n", - " │ ├── Epoch N-1 = 0.3143 (\u001B[32m↘ -0.0003\u001B[0m)\n", - " │ └── Best until now = 0.3143 (\u001B[32m↘ -0.0003\u001B[0m)\n", - " ├── Target_iou = 0.7536\n", - " │ ├── Epoch N-1 = 0.7534 (\u001B[32m↗ 0.0002\u001B[0m)\n", - " │ └── Best until now = 0.7534 (\u001B[32m↗ 0.0002\u001B[0m)\n", - " ├── Background_iou = 0.5448\n", - " │ ├── Epoch N-1 = 0.5443 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " │ └── Best until now = 0.5443 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " └── Mean_iou = 0.6492\n", - " ├── Epoch N-1 = 0.6488 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " └── Best until now = 0.6488 (\u001B[32m↗ 0.0004\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "code", + "execution_count": 8, + "metadata": { + "id": "xXPMJQCJzmb4" + }, + "outputs": [], + "source": [ + "from super_gradients.training import models\n", + "from super_gradients.common.object_names import Models\n", + "\n", + "model = models.get(model_name=Models.PP_LITE_T_SEG,\n", + " arch_params={\"use_aux_heads\": False},\n", + " num_classes=1)" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 9: 100%|██████████| 309/309 [01:58<00:00, 2.61it/s, BCEDiceLoss=0.248, background_IOU=0.711, gpu_mem=1.14, mean_IOU=0.75, target_IOU=0.789]\n", - "Validating epoch 9: 100%|██████████| 65/65 [00:17<00:00, 3.76it/s]\n", - "[2023-11-08 11:19:14] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:19:14] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7540615200996399\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "fU8orO7wlwIK" + }, + "source": [ + "SG includes implementations of many different architectures for semantic segmentation tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 9\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2479\n", - "│ │ ├── Epoch N-1 = 0.2517 (\u001B[32m↘ -0.0037\u001B[0m)\n", - "│ │ └── Best until now = 0.2517 (\u001B[32m↘ -0.0037\u001B[0m)\n", - "│ ├── Target_iou = 0.7895\n", - "│ │ ├── Epoch N-1 = 0.7861 (\u001B[32m↗ 0.0034\u001B[0m)\n", - "│ │ └── Best until now = 0.7861 (\u001B[32m↗ 0.0034\u001B[0m)\n", - "│ ├── Background_iou = 0.7109\n", - "│ │ ├── Epoch N-1 = 0.7112 (\u001B[31m↘ -0.0003\u001B[0m)\n", - "│ │ └── Best until now = 0.7112 (\u001B[31m↘ -0.0003\u001B[0m)\n", - "│ └── Mean_iou = 0.7502\n", - "│ ├── Epoch N-1 = 0.7487 (\u001B[32m↗ 0.0015\u001B[0m)\n", - "│ └── Best until now = 0.7487 (\u001B[32m↗ 0.0015\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3133\n", - " │ ├── Epoch N-1 = 0.314 (\u001B[32m↘ -0.0007\u001B[0m)\n", - " │ └── Best until now = 0.314 (\u001B[32m↘ -0.0007\u001B[0m)\n", - " ├── Target_iou = 0.7541\n", - " │ ├── Epoch N-1 = 0.7536 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " │ └── Best until now = 0.7536 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " ├── Background_iou = 0.5458\n", - " │ ├── Epoch N-1 = 0.5448 (\u001B[32m↗ 0.001\u001B[0m)\n", - " │ └── Best until now = 0.5448 (\u001B[32m↗ 0.001\u001B[0m)\n", - " └── Mean_iou = 0.6499\n", - " ├── Epoch N-1 = 0.6492 (\u001B[32m↗ 0.0007\u001B[0m)\n", - " └── Best until now = 0.6492 (\u001B[32m↗ 0.0007\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "-oGSU3V8lqcm" + }, + "source": [ + "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", + "and extra Auxiliary heads aren't used for training.\n", + "\n", + "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 10: 100%|██████████| 309/309 [01:58<00:00, 2.61it/s, BCEDiceLoss=0.241, background_IOU=0.724, gpu_mem=1.14, mean_IOU=0.76, target_IOU=0.796]\n", - "Validating epoch 10: 100%|██████████| 65/65 [00:16<00:00, 3.98it/s]\n", - "[2023-11-08 11:21:31] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:21:31] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7544161081314087\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "X-_dBewgr1dG" + }, + "source": [ + "# 4. Training setup\n", + "\n", + "\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 10\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2409\n", - "│ │ ├── Epoch N-1 = 0.2479 (\u001B[32m↘ -0.007\u001B[0m)\n", - "│ │ └── Best until now = 0.2479 (\u001B[32m↘ -0.007\u001B[0m)\n", - "│ ├── Target_iou = 0.7962\n", - "│ │ ├── Epoch N-1 = 0.7895 (\u001B[32m↗ 0.0068\u001B[0m)\n", - "│ │ └── Best until now = 0.7895 (\u001B[32m↗ 0.0068\u001B[0m)\n", - "│ ├── Background_iou = 0.7243\n", - "│ │ ├── Epoch N-1 = 0.7109 (\u001B[32m↗ 0.0134\u001B[0m)\n", - "│ │ └── Best until now = 0.7112 (\u001B[32m↗ 0.0132\u001B[0m)\n", - "│ └── Mean_iou = 0.7603\n", - "│ ├── Epoch N-1 = 0.7502 (\u001B[32m↗ 0.0101\u001B[0m)\n", - "│ └── Best until now = 0.7502 (\u001B[32m↗ 0.0101\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3126\n", - " │ ├── Epoch N-1 = 0.3133 (\u001B[32m↘ -0.0007\u001B[0m)\n", - " │ └── Best until now = 0.3133 (\u001B[32m↘ -0.0007\u001B[0m)\n", - " ├── Target_iou = 0.7544\n", - " │ ├── Epoch N-1 = 0.7541 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " │ └── Best until now = 0.7541 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " ├── Background_iou = 0.5464\n", - " │ ├── Epoch N-1 = 0.5458 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " │ └── Best until now = 0.5458 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " └── Mean_iou = 0.6504\n", - " ├── Epoch N-1 = 0.6499 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " └── Best until now = 0.6499 (\u001B[32m↗ 0.0005\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "H1Rll8Orl-Dy" + }, + "source": [ + "\n", + "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n", + "\n", + "We will be using an average of BCE and Dice loss for segmentation, with different learning rates for the replaced segmentation head layer, and the rest of the network- this is controlled by the `multiply_head_lr` parameter which is the multiplication factor of the learning rate for the newly replaced layer.\n", + "\n", + "As our `metric_to_watch`, we will be monitoring the `target_IOU` which is one of the components of `BinaryIOU` torchmetrics object (the other components are `mean_IOU` which is the mean of the background and target IOUs, and `background_IOU`)." + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 11: 100%|██████████| 309/309 [01:56<00:00, 2.64it/s, BCEDiceLoss=0.235, background_IOU=0.73, gpu_mem=1.14, mean_IOU=0.764, target_IOU=0.799]\n", - "Validating epoch 11: 100%|██████████| 65/65 [00:16<00:00, 4.00it/s]\n", - "[2023-11-08 11:23:47] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:23:47] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7546034455299377\n" - ] + "cell_type": "code", + "execution_count": 10, + "metadata": { + "id": "NShu3zLgr5qD" + }, + "outputs": [], + "source": [ + "from super_gradients.training.metrics.segmentation_metrics import BinaryIOU\n", + "\n", + "train_params = {\"max_epochs\": 15,\n", + " \"lr_mode\": \"cosine\",\n", + " \"initial_lr\": 0.01,\n", + " \"lr_warmup_epochs\": 5,\n", + " \"multiply_head_lr\": 10,\n", + " \"optimizer\": \"SGD\",\n", + " \"loss\": \"BCEDiceLoss\",\n", + " \"ema\": True,\n", + " \"ema_params\":\n", + " {\n", + " \"decay\": 0.9999,\n", + " \"decay_type\": \"exp\",\n", + " \"beta\": 15,\n", + " },\n", + "\n", + " \"zero_weight_decay_on_bias_and_bn\": True,\n", + " \"average_best_models\": True,\n", + " \"metric_to_watch\": \"target_IOU\",\n", + " \"greater_metric_to_watch_is_better\": True,\n", + " \"train_metrics_list\": [BinaryIOU()],\n", + " \"valid_metrics_list\": [BinaryIOU()],\n", + " \"loss_logging_items_names\": [\"loss\"]\n", + " }" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 11\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2349\n", - "│ │ ├── Epoch N-1 = 0.2409 (\u001B[32m↘ -0.006\u001B[0m)\n", - "│ │ └── Best until now = 0.2409 (\u001B[32m↘ -0.006\u001B[0m)\n", - "│ ├── Target_iou = 0.7988\n", - "│ │ ├── Epoch N-1 = 0.7962 (\u001B[32m↗ 0.0025\u001B[0m)\n", - "│ │ └── Best until now = 0.7962 (\u001B[32m↗ 0.0025\u001B[0m)\n", - "│ ├── Background_iou = 0.7297\n", - "│ │ ├── Epoch N-1 = 0.7243 (\u001B[32m↗ 0.0053\u001B[0m)\n", - "│ │ └── Best until now = 0.7243 (\u001B[32m↗ 0.0053\u001B[0m)\n", - "│ └── Mean_iou = 0.7642\n", - "│ ├── Epoch N-1 = 0.7603 (\u001B[32m↗ 0.0039\u001B[0m)\n", - "│ └── Best until now = 0.7603 (\u001B[32m↗ 0.0039\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3118\n", - " │ ├── Epoch N-1 = 0.3126 (\u001B[32m↘ -0.0008\u001B[0m)\n", - " │ └── Best until now = 0.3126 (\u001B[32m↘ -0.0008\u001B[0m)\n", - " ├── Target_iou = 0.7546\n", - " │ ├── Epoch N-1 = 0.7544 (\u001B[32m↗ 0.0002\u001B[0m)\n", - " │ └── Best until now = 0.7544 (\u001B[32m↗ 0.0002\u001B[0m)\n", - " ├── Background_iou = 0.5468\n", - " │ ├── Epoch N-1 = 0.5464 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " │ └── Best until now = 0.5464 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " └── Mean_iou = 0.6507\n", - " ├── Epoch N-1 = 0.6504 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " └── Best until now = 0.6504 (\u001B[32m↗ 0.0003\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "qTECVyhcs506" + }, + "source": [ + "# 5. Training and evaluation\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 12: 100%|██████████| 309/309 [01:56<00:00, 2.65it/s, BCEDiceLoss=0.227, background_IOU=0.733, gpu_mem=1.14, mean_IOU=0.769, target_IOU=0.805]\n", - "Validating epoch 12: 100%|██████████| 65/65 [00:16<00:00, 3.95it/s]\n", - "[2023-11-08 11:26:03] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:26:03] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7549036145210266\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "S1K5MU2kmmDb" + }, + "source": [ + "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", + "\n", + "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders.\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 12\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2272\n", - "│ │ ├── Epoch N-1 = 0.2349 (\u001B[32m↘ -0.0077\u001B[0m)\n", - "│ │ └── Best until now = 0.2349 (\u001B[32m↘ -0.0077\u001B[0m)\n", - "│ ├── Target_iou = 0.8051\n", - "│ │ ├── Epoch N-1 = 0.7988 (\u001B[32m↗ 0.0064\u001B[0m)\n", - "│ │ └── Best until now = 0.7988 (\u001B[32m↗ 0.0064\u001B[0m)\n", - "│ ├── Background_iou = 0.7333\n", - "│ │ ├── Epoch N-1 = 0.7297 (\u001B[32m↗ 0.0036\u001B[0m)\n", - "│ │ └── Best until now = 0.7297 (\u001B[32m↗ 0.0036\u001B[0m)\n", - "│ └── Mean_iou = 0.7692\n", - "│ ├── Epoch N-1 = 0.7642 (\u001B[32m↗ 0.005\u001B[0m)\n", - "│ └── Best until now = 0.7642 (\u001B[32m↗ 0.005\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.311\n", - " │ ├── Epoch N-1 = 0.3118 (\u001B[32m↘ -0.0008\u001B[0m)\n", - " │ └── Best until now = 0.3118 (\u001B[32m↘ -0.0008\u001B[0m)\n", - " ├── Target_iou = 0.7549\n", - " │ ├── Epoch N-1 = 0.7546 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " │ └── Best until now = 0.7546 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " ├── Background_iou = 0.5474\n", - " │ ├── Epoch N-1 = 0.5468 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " │ └── Best until now = 0.5468 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " └── Mean_iou = 0.6512\n", - " ├── Epoch N-1 = 0.6507 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " └── Best until now = 0.6507 (\u001B[32m↗ 0.0005\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "code", + "execution_count": 11, + "metadata": { + "id": "u6roEj9ktFTi", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "4a295f63-f0c4-43a7-c6e8-2f7ffd1b5ce2" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 11:15:07] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231113_111507_197271`\n", + "[2023-11-13 11:15:07] INFO - sg_trainer.py - Checkpoints directory: ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271\n", + "[2023-11-13 11:15:07] INFO - sg_trainer.py - Using EMA with params {'decay': 0.9999, 'decay_type': 'exp', 'beta': 15}\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is now moved to ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/console_Nov13_11_15_07.txt\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 11:15:08] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", + " - Mode: Single GPU\n", + " - Number of GPUs: 1 (1 available on the machine)\n", + " - Full dataset size: 2477 (len(train_set))\n", + " - Batch size per GPU: 8 (batch_size)\n", + " - Batch Accumulate: 1 (batch_accumulate)\n", + " - Total batch size: 8 (num_gpus * batch_size)\n", + " - Effective Batch size: 8 (num_gpus * batch_size * batch_accumulate)\n", + " - Iterations per epoch: 309 (len(train_loader))\n", + " - Gradient updates per epoch: 309 (len(train_loader) / batch_accumulate)\n", + "\n", + "[2023-11-13 11:15:08] INFO - sg_trainer.py - Started training for 15 epochs (0/14)\n", + "\n", + "Train epoch 0: 100%|██████████| 309/309 [02:12<00:00, 2.33it/s, BCEDiceLoss=0.4, background_IOU=0.545, gpu_mem=1.14, mean_IOU=0.609, target_IOU=0.674]\n", + "Validating: 100%|██████████| 65/65 [00:17<00:00, 3.69it/s]\n", + "[2023-11-13 11:17:39] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:17:39] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.6779429912567139\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 0\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.4001\n", + "│ ├── Target_iou = 0.6736\n", + "│ ├── Background_iou = 0.5448\n", + "│ └── Mean_iou = 0.6092\n", + "└── Validation\n", + " ├── Bcediceloss = 0.4166\n", + " ├── Target_iou = 0.6779\n", + " ├── Background_iou = 0.4039\n", + " └── Mean_iou = 0.5409\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 1: 100%|██████████| 309/309 [02:05<00:00, 2.46it/s, BCEDiceLoss=0.338, background_IOU=0.604, gpu_mem=1.14, mean_IOU=0.661, target_IOU=0.719]\n", + "Validating epoch 1: 100%|██████████| 65/65 [00:17<00:00, 3.69it/s]\n", + "[2023-11-13 11:20:05] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:20:05] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7205255031585693\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 1\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.3381\n", + "│ │ ├── Epoch N-1 = 0.4001 (\u001b[32m↘ -0.062\u001b[0m)\n", + "│ │ └── Best until now = 0.4001 (\u001b[32m↘ -0.062\u001b[0m)\n", + "│ ├── Target_iou = 0.7193\n", + "│ │ ├── Epoch N-1 = 0.6736 (\u001b[32m↗ 0.0457\u001b[0m)\n", + "│ │ └── Best until now = 0.6736 (\u001b[32m↗ 0.0457\u001b[0m)\n", + "│ ├── Background_iou = 0.6036\n", + "│ │ ├── Epoch N-1 = 0.5448 (\u001b[32m↗ 0.0587\u001b[0m)\n", + "│ │ └── Best until now = 0.5448 (\u001b[32m↗ 0.0587\u001b[0m)\n", + "│ └── Mean_iou = 0.6614\n", + "│ ├── Epoch N-1 = 0.6092 (\u001b[32m↗ 0.0522\u001b[0m)\n", + "│ └── Best until now = 0.6092 (\u001b[32m↗ 0.0522\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3578\n", + " │ ├── Epoch N-1 = 0.4166 (\u001b[32m↘ -0.0588\u001b[0m)\n", + " │ └── Best until now = 0.4166 (\u001b[32m↘ -0.0588\u001b[0m)\n", + " ├── Target_iou = 0.7205\n", + " │ ├── Epoch N-1 = 0.6779 (\u001b[32m↗ 0.0426\u001b[0m)\n", + " │ └── Best until now = 0.6779 (\u001b[32m↗ 0.0426\u001b[0m)\n", + " ├── Background_iou = 0.4497\n", + " │ ├── Epoch N-1 = 0.4039 (\u001b[32m↗ 0.0458\u001b[0m)\n", + " │ └── Best until now = 0.4039 (\u001b[32m↗ 0.0458\u001b[0m)\n", + " └── Mean_iou = 0.5851\n", + " ├── Epoch N-1 = 0.5409 (\u001b[32m↗ 0.0442\u001b[0m)\n", + " └── Best until now = 0.5409 (\u001b[32m↗ 0.0442\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 2: 100%|██████████| 309/309 [02:00<00:00, 2.55it/s, BCEDiceLoss=0.32, background_IOU=0.634, gpu_mem=1.14, mean_IOU=0.684, target_IOU=0.734]\n", + "Validating epoch 2: 100%|██████████| 65/65 [00:16<00:00, 3.84it/s]\n", + "[2023-11-13 11:22:24] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:22:24] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7300039529800415\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 2\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.3199\n", + "│ │ ├── Epoch N-1 = 0.3381 (\u001b[32m↘ -0.0182\u001b[0m)\n", + "│ │ └── Best until now = 0.3381 (\u001b[32m↘ -0.0182\u001b[0m)\n", + "│ ├── Target_iou = 0.734\n", + "│ │ ├── Epoch N-1 = 0.7193 (\u001b[32m↗ 0.0147\u001b[0m)\n", + "│ │ └── Best until now = 0.7193 (\u001b[32m↗ 0.0147\u001b[0m)\n", + "│ ├── Background_iou = 0.6344\n", + "│ │ ├── Epoch N-1 = 0.6036 (\u001b[32m↗ 0.0308\u001b[0m)\n", + "│ │ └── Best until now = 0.6036 (\u001b[32m↗ 0.0308\u001b[0m)\n", + "│ └── Mean_iou = 0.6842\n", + "│ ├── Epoch N-1 = 0.6614 (\u001b[32m↗ 0.0227\u001b[0m)\n", + "│ └── Best until now = 0.6614 (\u001b[32m↗ 0.0227\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.357\n", + " │ ├── Epoch N-1 = 0.3578 (\u001b[32m↘ -0.0008\u001b[0m)\n", + " │ └── Best until now = 0.3578 (\u001b[32m↘ -0.0008\u001b[0m)\n", + " ├── Target_iou = 0.73\n", + " │ ├── Epoch N-1 = 0.7205 (\u001b[32m↗ 0.0095\u001b[0m)\n", + " │ └── Best until now = 0.7205 (\u001b[32m↗ 0.0095\u001b[0m)\n", + " ├── Background_iou = 0.4503\n", + " │ ├── Epoch N-1 = 0.4497 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " │ └── Best until now = 0.4497 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " └── Mean_iou = 0.5902\n", + " ├── Epoch N-1 = 0.5851 (\u001b[32m↗ 0.0051\u001b[0m)\n", + " └── Best until now = 0.5851 (\u001b[32m↗ 0.0051\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 3: 100%|██████████| 309/309 [01:59<00:00, 2.58it/s, BCEDiceLoss=0.302, background_IOU=0.645, gpu_mem=1.14, mean_IOU=0.697, target_IOU=0.75]\n", + "Validating epoch 3: 100%|██████████| 65/65 [00:16<00:00, 3.84it/s]\n", + "[2023-11-13 11:24:43] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:24:43] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7432040572166443\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 3\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.3022\n", + "│ │ ├── Epoch N-1 = 0.3199 (\u001b[32m↘ -0.0177\u001b[0m)\n", + "│ │ └── Best until now = 0.3199 (\u001b[32m↘ -0.0177\u001b[0m)\n", + "│ ├── Target_iou = 0.7501\n", + "│ │ ├── Epoch N-1 = 0.734 (\u001b[32m↗ 0.0161\u001b[0m)\n", + "│ │ └── Best until now = 0.734 (\u001b[32m↗ 0.0161\u001b[0m)\n", + "│ ├── Background_iou = 0.6447\n", + "│ │ ├── Epoch N-1 = 0.6344 (\u001b[32m↗ 0.0103\u001b[0m)\n", + "│ │ └── Best until now = 0.6344 (\u001b[32m↗ 0.0103\u001b[0m)\n", + "│ └── Mean_iou = 0.6974\n", + "│ ├── Epoch N-1 = 0.6842 (\u001b[32m↗ 0.0132\u001b[0m)\n", + "│ └── Best until now = 0.6842 (\u001b[32m↗ 0.0132\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3307\n", + " │ ├── Epoch N-1 = 0.357 (\u001b[32m↘ -0.0263\u001b[0m)\n", + " │ └── Best until now = 0.357 (\u001b[32m↘ -0.0263\u001b[0m)\n", + " ├── Target_iou = 0.7432\n", + " │ ├── Epoch N-1 = 0.73 (\u001b[32m↗ 0.0132\u001b[0m)\n", + " │ └── Best until now = 0.73 (\u001b[32m↗ 0.0132\u001b[0m)\n", + " ├── Background_iou = 0.4794\n", + " │ ├── Epoch N-1 = 0.4503 (\u001b[32m↗ 0.0291\u001b[0m)\n", + " │ └── Best until now = 0.4503 (\u001b[32m↗ 0.0291\u001b[0m)\n", + " └── Mean_iou = 0.6113\n", + " ├── Epoch N-1 = 0.5902 (\u001b[32m↗ 0.0212\u001b[0m)\n", + " └── Best until now = 0.5902 (\u001b[32m↗ 0.0212\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 4: 100%|██████████| 309/309 [02:00<00:00, 2.56it/s, BCEDiceLoss=0.287, background_IOU=0.67, gpu_mem=1.14, mean_IOU=0.715, target_IOU=0.76]\n", + "Validating epoch 4: 100%|██████████| 65/65 [00:17<00:00, 3.79it/s]\n", + "[2023-11-13 11:27:02] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:27:02] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7445915341377258\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 4\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2867\n", + "│ │ ├── Epoch N-1 = 0.3022 (\u001b[32m↘ -0.0155\u001b[0m)\n", + "│ │ └── Best until now = 0.3022 (\u001b[32m↘ -0.0155\u001b[0m)\n", + "│ ├── Target_iou = 0.7604\n", + "│ │ ├── Epoch N-1 = 0.7501 (\u001b[32m↗ 0.0103\u001b[0m)\n", + "│ │ └── Best until now = 0.7501 (\u001b[32m↗ 0.0103\u001b[0m)\n", + "│ ├── Background_iou = 0.6697\n", + "│ │ ├── Epoch N-1 = 0.6447 (\u001b[32m↗ 0.0251\u001b[0m)\n", + "│ │ └── Best until now = 0.6447 (\u001b[32m↗ 0.0251\u001b[0m)\n", + "│ └── Mean_iou = 0.715\n", + "│ ├── Epoch N-1 = 0.6974 (\u001b[32m↗ 0.0177\u001b[0m)\n", + "│ └── Best until now = 0.6974 (\u001b[32m↗ 0.0177\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3281\n", + " │ ├── Epoch N-1 = 0.3307 (\u001b[32m↘ -0.0026\u001b[0m)\n", + " │ └── Best until now = 0.3307 (\u001b[32m↘ -0.0026\u001b[0m)\n", + " ├── Target_iou = 0.7446\n", + " │ ├── Epoch N-1 = 0.7432 (\u001b[32m↗ 0.0014\u001b[0m)\n", + " │ └── Best until now = 0.7432 (\u001b[32m↗ 0.0014\u001b[0m)\n", + " ├── Background_iou = 0.4869\n", + " │ ├── Epoch N-1 = 0.4794 (\u001b[32m↗ 0.0074\u001b[0m)\n", + " │ └── Best until now = 0.4794 (\u001b[32m↗ 0.0074\u001b[0m)\n", + " └── Mean_iou = 0.6157\n", + " ├── Epoch N-1 = 0.6113 (\u001b[32m↗ 0.0044\u001b[0m)\n", + " └── Best until now = 0.6113 (\u001b[32m↗ 0.0044\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 5: 100%|██████████| 309/309 [02:02<00:00, 2.53it/s, BCEDiceLoss=0.287, background_IOU=0.664, gpu_mem=1.14, mean_IOU=0.712, target_IOU=0.761]\n", + "Validating epoch 5: 100%|██████████| 65/65 [00:17<00:00, 3.75it/s]\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 5\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2869\n", + "│ │ ├── Epoch N-1 = 0.2867 (\u001b[31m↗ 1e-04\u001b[0m)\n", + "│ │ └── Best until now = 0.2867 (\u001b[31m↗ 1e-04\u001b[0m)\n", + "│ ├── Target_iou = 0.7606\n", + "│ │ ├── Epoch N-1 = 0.7604 (\u001b[32m↗ 0.0002\u001b[0m)\n", + "│ │ └── Best until now = 0.7604 (\u001b[32m↗ 0.0002\u001b[0m)\n", + "│ ├── Background_iou = 0.6637\n", + "│ │ ├── Epoch N-1 = 0.6697 (\u001b[31m↘ -0.0061\u001b[0m)\n", + "│ │ └── Best until now = 0.6697 (\u001b[31m↘ -0.0061\u001b[0m)\n", + "│ └── Mean_iou = 0.7121\n", + "│ ├── Epoch N-1 = 0.715 (\u001b[31m↘ -0.0029\u001b[0m)\n", + "│ └── Best until now = 0.715 (\u001b[31m↘ -0.0029\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3339\n", + " │ ├── Epoch N-1 = 0.3281 (\u001b[31m↗ 0.0059\u001b[0m)\n", + " │ └── Best until now = 0.3281 (\u001b[31m↗ 0.0059\u001b[0m)\n", + " ├── Target_iou = 0.7402\n", + " │ ├── Epoch N-1 = 0.7446 (\u001b[31m↘ -0.0044\u001b[0m)\n", + " │ └── Best until now = 0.7446 (\u001b[31m↘ -0.0044\u001b[0m)\n", + " ├── Background_iou = 0.4593\n", + " │ ├── Epoch N-1 = 0.4869 (\u001b[31m↘ -0.0276\u001b[0m)\n", + " │ └── Best until now = 0.4869 (\u001b[31m↘ -0.0276\u001b[0m)\n", + " └── Mean_iou = 0.5997\n", + " ├── Epoch N-1 = 0.6157 (\u001b[31m↘ -0.016\u001b[0m)\n", + " └── Best until now = 0.6157 (\u001b[31m↘ -0.016\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 6: 100%|██████████| 309/309 [02:03<00:00, 2.50it/s, BCEDiceLoss=0.269, background_IOU=0.689, gpu_mem=1.14, mean_IOU=0.731, target_IOU=0.772]\n", + "Validating epoch 6: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 6\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2686\n", + "│ │ ├── Epoch N-1 = 0.2869 (\u001b[32m↘ -0.0183\u001b[0m)\n", + "│ │ └── Best until now = 0.2867 (\u001b[32m↘ -0.0181\u001b[0m)\n", + "│ ├── Target_iou = 0.7721\n", + "│ │ ├── Epoch N-1 = 0.7606 (\u001b[32m↗ 0.0115\u001b[0m)\n", + "│ │ └── Best until now = 0.7606 (\u001b[32m↗ 0.0115\u001b[0m)\n", + "│ ├── Background_iou = 0.6892\n", + "│ │ ├── Epoch N-1 = 0.6637 (\u001b[32m↗ 0.0255\u001b[0m)\n", + "│ │ └── Best until now = 0.6697 (\u001b[32m↗ 0.0194\u001b[0m)\n", + "│ └── Mean_iou = 0.7306\n", + "│ ├── Epoch N-1 = 0.7121 (\u001b[32m↗ 0.0185\u001b[0m)\n", + "│ └── Best until now = 0.715 (\u001b[32m↗ 0.0156\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3278\n", + " │ ├── Epoch N-1 = 0.3339 (\u001b[32m↘ -0.0061\u001b[0m)\n", + " │ └── Best until now = 0.3281 (\u001b[32m↘ -0.0003\u001b[0m)\n", + " ├── Target_iou = 0.7431\n", + " │ ├── Epoch N-1 = 0.7402 (\u001b[32m↗ 0.003\u001b[0m)\n", + " │ └── Best until now = 0.7446 (\u001b[31m↘ -0.0015\u001b[0m)\n", + " ├── Background_iou = 0.4733\n", + " │ ├── Epoch N-1 = 0.4593 (\u001b[32m↗ 0.0139\u001b[0m)\n", + " │ └── Best until now = 0.4869 (\u001b[31m↘ -0.0136\u001b[0m)\n", + " └── Mean_iou = 0.6082\n", + " ├── Epoch N-1 = 0.5997 (\u001b[32m↗ 0.0085\u001b[0m)\n", + " └── Best until now = 0.6157 (\u001b[31m↘ -0.0075\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 7: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.259, background_IOU=0.701, gpu_mem=1.14, mean_IOU=0.741, target_IOU=0.781]\n", + "Validating epoch 7: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", + "[2023-11-13 11:34:05] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:34:05] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7548585534095764\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 7\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.259\n", + "│ │ ├── Epoch N-1 = 0.2686 (\u001b[32m↘ -0.0096\u001b[0m)\n", + "│ │ └── Best until now = 0.2686 (\u001b[32m↘ -0.0096\u001b[0m)\n", + "│ ├── Target_iou = 0.7808\n", + "│ │ ├── Epoch N-1 = 0.7721 (\u001b[32m↗ 0.0087\u001b[0m)\n", + "│ │ └── Best until now = 0.7721 (\u001b[32m↗ 0.0087\u001b[0m)\n", + "│ ├── Background_iou = 0.7009\n", + "│ │ ├── Epoch N-1 = 0.6892 (\u001b[32m↗ 0.0117\u001b[0m)\n", + "│ │ └── Best until now = 0.6892 (\u001b[32m↗ 0.0117\u001b[0m)\n", + "│ └── Mean_iou = 0.7409\n", + "│ ├── Epoch N-1 = 0.7306 (\u001b[32m↗ 0.0102\u001b[0m)\n", + "│ └── Best until now = 0.7306 (\u001b[32m↗ 0.0102\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3129\n", + " │ ├── Epoch N-1 = 0.3278 (\u001b[32m↘ -0.0149\u001b[0m)\n", + " │ └── Best until now = 0.3278 (\u001b[32m↘ -0.0149\u001b[0m)\n", + " ├── Target_iou = 0.7549\n", + " │ ├── Epoch N-1 = 0.7431 (\u001b[32m↗ 0.0117\u001b[0m)\n", + " │ └── Best until now = 0.7446 (\u001b[32m↗ 0.0103\u001b[0m)\n", + " ├── Background_iou = 0.5241\n", + " │ ├── Epoch N-1 = 0.4733 (\u001b[32m↗ 0.0508\u001b[0m)\n", + " │ └── Best until now = 0.4869 (\u001b[32m↗ 0.0372\u001b[0m)\n", + " └── Mean_iou = 0.6395\n", + " ├── Epoch N-1 = 0.6082 (\u001b[32m↗ 0.0313\u001b[0m)\n", + " └── Best until now = 0.6157 (\u001b[32m↗ 0.0238\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 8: 100%|██████████| 309/309 [02:05<00:00, 2.47it/s, BCEDiceLoss=0.251, background_IOU=0.713, gpu_mem=1.14, mean_IOU=0.749, target_IOU=0.786]\n", + "Validating epoch 8: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", + "[2023-11-13 11:36:30] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:36:30] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7585687637329102\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 8\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.251\n", + "│ │ ├── Epoch N-1 = 0.259 (\u001b[32m↘ -0.008\u001b[0m)\n", + "│ │ └── Best until now = 0.259 (\u001b[32m↘ -0.008\u001b[0m)\n", + "│ ├── Target_iou = 0.786\n", + "│ │ ├── Epoch N-1 = 0.7808 (\u001b[32m↗ 0.0052\u001b[0m)\n", + "│ │ └── Best until now = 0.7808 (\u001b[32m↗ 0.0052\u001b[0m)\n", + "│ ├── Background_iou = 0.7125\n", + "│ │ ├── Epoch N-1 = 0.7009 (\u001b[32m↗ 0.0116\u001b[0m)\n", + "│ │ └── Best until now = 0.7009 (\u001b[32m↗ 0.0116\u001b[0m)\n", + "│ └── Mean_iou = 0.7493\n", + "│ ├── Epoch N-1 = 0.7409 (\u001b[32m↗ 0.0084\u001b[0m)\n", + "│ └── Best until now = 0.7409 (\u001b[32m↗ 0.0084\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3091\n", + " │ ├── Epoch N-1 = 0.3129 (\u001b[32m↘ -0.0039\u001b[0m)\n", + " │ └── Best until now = 0.3129 (\u001b[32m↘ -0.0039\u001b[0m)\n", + " ├── Target_iou = 0.7586\n", + " │ ├── Epoch N-1 = 0.7549 (\u001b[32m↗ 0.0037\u001b[0m)\n", + " │ └── Best until now = 0.7549 (\u001b[32m↗ 0.0037\u001b[0m)\n", + " ├── Background_iou = 0.5411\n", + " │ ├── Epoch N-1 = 0.5241 (\u001b[32m↗ 0.017\u001b[0m)\n", + " │ └── Best until now = 0.5241 (\u001b[32m↗ 0.017\u001b[0m)\n", + " └── Mean_iou = 0.6498\n", + " ├── Epoch N-1 = 0.6395 (\u001b[32m↗ 0.0103\u001b[0m)\n", + " └── Best until now = 0.6395 (\u001b[32m↗ 0.0103\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 9: 100%|██████████| 309/309 [02:02<00:00, 2.53it/s, BCEDiceLoss=0.246, background_IOU=0.713, gpu_mem=1.14, mean_IOU=0.752, target_IOU=0.791]\n", + "Validating epoch 9: 100%|██████████| 65/65 [00:17<00:00, 3.72it/s]\n", + "[2023-11-13 11:38:53] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:38:53] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.759834885597229\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 9\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2465\n", + "│ │ ├── Epoch N-1 = 0.251 (\u001b[32m↘ -0.0045\u001b[0m)\n", + "│ │ └── Best until now = 0.251 (\u001b[32m↘ -0.0045\u001b[0m)\n", + "│ ├── Target_iou = 0.7905\n", + "│ │ ├── Epoch N-1 = 0.786 (\u001b[32m↗ 0.0045\u001b[0m)\n", + "│ │ └── Best until now = 0.786 (\u001b[32m↗ 0.0045\u001b[0m)\n", + "│ ├── Background_iou = 0.7133\n", + "│ │ ├── Epoch N-1 = 0.7125 (\u001b[32m↗ 0.0008\u001b[0m)\n", + "│ │ └── Best until now = 0.7125 (\u001b[32m↗ 0.0008\u001b[0m)\n", + "│ └── Mean_iou = 0.7519\n", + "│ ├── Epoch N-1 = 0.7493 (\u001b[32m↗ 0.0026\u001b[0m)\n", + "│ └── Best until now = 0.7493 (\u001b[32m↗ 0.0026\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3072\n", + " │ ├── Epoch N-1 = 0.3091 (\u001b[32m↘ -0.0018\u001b[0m)\n", + " │ └── Best until now = 0.3091 (\u001b[32m↘ -0.0018\u001b[0m)\n", + " ├── Target_iou = 0.7598\n", + " │ ├── Epoch N-1 = 0.7586 (\u001b[32m↗ 0.0013\u001b[0m)\n", + " │ └── Best until now = 0.7586 (\u001b[32m↗ 0.0013\u001b[0m)\n", + " ├── Background_iou = 0.5481\n", + " │ ├── Epoch N-1 = 0.5411 (\u001b[32m↗ 0.007\u001b[0m)\n", + " │ └── Best until now = 0.5411 (\u001b[32m↗ 0.007\u001b[0m)\n", + " └── Mean_iou = 0.6539\n", + " ├── Epoch N-1 = 0.6498 (\u001b[32m↗ 0.0041\u001b[0m)\n", + " └── Best until now = 0.6498 (\u001b[32m↗ 0.0041\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 10: 100%|██████████| 309/309 [02:03<00:00, 2.50it/s, BCEDiceLoss=0.24, background_IOU=0.723, gpu_mem=1.14, mean_IOU=0.759, target_IOU=0.796]\n", + "Validating epoch 10: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", + "[2023-11-13 11:41:16] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:41:16] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7605207562446594\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 10\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2399\n", + "│ │ ├── Epoch N-1 = 0.2465 (\u001b[32m↘ -0.0066\u001b[0m)\n", + "│ │ └── Best until now = 0.2465 (\u001b[32m↘ -0.0066\u001b[0m)\n", + "│ ├── Target_iou = 0.7956\n", + "│ │ ├── Epoch N-1 = 0.7905 (\u001b[32m↗ 0.0051\u001b[0m)\n", + "│ │ └── Best until now = 0.7905 (\u001b[32m↗ 0.0051\u001b[0m)\n", + "│ ├── Background_iou = 0.7229\n", + "│ │ ├── Epoch N-1 = 0.7133 (\u001b[32m↗ 0.0096\u001b[0m)\n", + "│ │ └── Best until now = 0.7133 (\u001b[32m↗ 0.0096\u001b[0m)\n", + "│ └── Mean_iou = 0.7593\n", + "│ ├── Epoch N-1 = 0.7519 (\u001b[32m↗ 0.0074\u001b[0m)\n", + "│ └── Best until now = 0.7519 (\u001b[32m↗ 0.0074\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3059\n", + " │ ├── Epoch N-1 = 0.3072 (\u001b[32m↘ -0.0014\u001b[0m)\n", + " │ └── Best until now = 0.3072 (\u001b[32m↘ -0.0014\u001b[0m)\n", + " ├── Target_iou = 0.7605\n", + " │ ├── Epoch N-1 = 0.7598 (\u001b[32m↗ 0.0007\u001b[0m)\n", + " │ └── Best until now = 0.7598 (\u001b[32m↗ 0.0007\u001b[0m)\n", + " ├── Background_iou = 0.5517\n", + " │ ├── Epoch N-1 = 0.5481 (\u001b[32m↗ 0.0037\u001b[0m)\n", + " │ └── Best until now = 0.5481 (\u001b[32m↗ 0.0037\u001b[0m)\n", + " └── Mean_iou = 0.6561\n", + " ├── Epoch N-1 = 0.6539 (\u001b[32m↗ 0.0022\u001b[0m)\n", + " └── Best until now = 0.6539 (\u001b[32m↗ 0.0022\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 11: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.231, background_IOU=0.733, gpu_mem=1.14, mean_IOU=0.767, target_IOU=0.801]\n", + "Validating epoch 11: 100%|██████████| 65/65 [00:17<00:00, 3.76it/s]\n", + "[2023-11-13 11:43:37] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:43:37] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7611058950424194\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 11\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2309\n", + "│ │ ├── Epoch N-1 = 0.2399 (\u001b[32m↘ -0.009\u001b[0m)\n", + "│ │ └── Best until now = 0.2399 (\u001b[32m↘ -0.009\u001b[0m)\n", + "│ ├── Target_iou = 0.8015\n", + "│ │ ├── Epoch N-1 = 0.7956 (\u001b[32m↗ 0.0059\u001b[0m)\n", + "│ │ └── Best until now = 0.7956 (\u001b[32m↗ 0.0059\u001b[0m)\n", + "│ ├── Background_iou = 0.7333\n", + "│ │ ├── Epoch N-1 = 0.7229 (\u001b[32m↗ 0.0104\u001b[0m)\n", + "│ │ └── Best until now = 0.7229 (\u001b[32m↗ 0.0104\u001b[0m)\n", + "│ └── Mean_iou = 0.7674\n", + "│ ├── Epoch N-1 = 0.7593 (\u001b[32m↗ 0.0081\u001b[0m)\n", + "│ └── Best until now = 0.7593 (\u001b[32m↗ 0.0081\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3046\n", + " │ ├── Epoch N-1 = 0.3059 (\u001b[32m↘ -0.0012\u001b[0m)\n", + " │ └── Best until now = 0.3059 (\u001b[32m↘ -0.0012\u001b[0m)\n", + " ├── Target_iou = 0.7611\n", + " │ ├── Epoch N-1 = 0.7605 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " │ └── Best until now = 0.7605 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " ├── Background_iou = 0.5546\n", + " │ ├── Epoch N-1 = 0.5517 (\u001b[32m↗ 0.0029\u001b[0m)\n", + " │ └── Best until now = 0.5517 (\u001b[32m↗ 0.0029\u001b[0m)\n", + " └── Mean_iou = 0.6579\n", + " ├── Epoch N-1 = 0.6561 (\u001b[32m↗ 0.0017\u001b[0m)\n", + " └── Best until now = 0.6561 (\u001b[32m↗ 0.0017\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 12: 100%|██████████| 309/309 [02:03<00:00, 2.51it/s, BCEDiceLoss=0.224, background_IOU=0.736, gpu_mem=1.14, mean_IOU=0.771, target_IOU=0.807]\n", + "Validating epoch 12: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", + "[2023-11-13 11:46:00] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:46:00] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7616798877716064\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 12\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2243\n", + "│ │ ├── Epoch N-1 = 0.2309 (\u001b[32m↘ -0.0066\u001b[0m)\n", + "│ │ └── Best until now = 0.2309 (\u001b[32m↘ -0.0066\u001b[0m)\n", + "│ ├── Target_iou = 0.8068\n", + "│ │ ├── Epoch N-1 = 0.8015 (\u001b[32m↗ 0.0053\u001b[0m)\n", + "│ │ └── Best until now = 0.8015 (\u001b[32m↗ 0.0053\u001b[0m)\n", + "│ ├── Background_iou = 0.736\n", + "│ │ ├── Epoch N-1 = 0.7333 (\u001b[32m↗ 0.0027\u001b[0m)\n", + "│ │ └── Best until now = 0.7333 (\u001b[32m↗ 0.0027\u001b[0m)\n", + "│ └── Mean_iou = 0.7714\n", + "│ ├── Epoch N-1 = 0.7674 (\u001b[32m↗ 0.004\u001b[0m)\n", + "│ └── Best until now = 0.7674 (\u001b[32m↗ 0.004\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3035\n", + " │ ├── Epoch N-1 = 0.3046 (\u001b[32m↘ -0.0012\u001b[0m)\n", + " │ └── Best until now = 0.3046 (\u001b[32m↘ -0.0012\u001b[0m)\n", + " ├── Target_iou = 0.7617\n", + " │ ├── Epoch N-1 = 0.7611 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " │ └── Best until now = 0.7611 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " ├── Background_iou = 0.5569\n", + " │ ├── Epoch N-1 = 0.5546 (\u001b[32m↗ 0.0023\u001b[0m)\n", + " │ └── Best until now = 0.5546 (\u001b[32m↗ 0.0023\u001b[0m)\n", + " └── Mean_iou = 0.6593\n", + " ├── Epoch N-1 = 0.6579 (\u001b[32m↗ 0.0014\u001b[0m)\n", + " └── Best until now = 0.6579 (\u001b[32m↗ 0.0014\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 13: 100%|██████████| 309/309 [02:01<00:00, 2.55it/s, BCEDiceLoss=0.219, background_IOU=0.745, gpu_mem=1.14, mean_IOU=0.777, target_IOU=0.81]\n", + "Validating epoch 13: 100%|██████████| 65/65 [00:17<00:00, 3.81it/s]\n", + "[2023-11-13 11:48:23] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:48:23] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7624021172523499\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 13\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2194\n", + "│ │ ├── Epoch N-1 = 0.2243 (\u001b[32m↘ -0.0049\u001b[0m)\n", + "│ │ └── Best until now = 0.2243 (\u001b[32m↘ -0.0049\u001b[0m)\n", + "│ ├── Target_iou = 0.8097\n", + "│ │ ├── Epoch N-1 = 0.8068 (\u001b[32m↗ 0.0029\u001b[0m)\n", + "│ │ └── Best until now = 0.8068 (\u001b[32m↗ 0.0029\u001b[0m)\n", + "│ ├── Background_iou = 0.7447\n", + "│ │ ├── Epoch N-1 = 0.736 (\u001b[32m↗ 0.0086\u001b[0m)\n", + "│ │ └── Best until now = 0.736 (\u001b[32m↗ 0.0086\u001b[0m)\n", + "│ └── Mean_iou = 0.7772\n", + "│ ├── Epoch N-1 = 0.7714 (\u001b[32m↗ 0.0058\u001b[0m)\n", + "│ └── Best until now = 0.7714 (\u001b[32m↗ 0.0058\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3024\n", + " │ ├── Epoch N-1 = 0.3035 (\u001b[32m↘ -0.0011\u001b[0m)\n", + " │ └── Best until now = 0.3035 (\u001b[32m↘ -0.0011\u001b[0m)\n", + " ├── Target_iou = 0.7624\n", + " │ ├── Epoch N-1 = 0.7617 (\u001b[32m↗ 0.0007\u001b[0m)\n", + " │ └── Best until now = 0.7617 (\u001b[32m↗ 0.0007\u001b[0m)\n", + " ├── Background_iou = 0.5596\n", + " │ ├── Epoch N-1 = 0.5569 (\u001b[32m↗ 0.0027\u001b[0m)\n", + " │ └── Best until now = 0.5569 (\u001b[32m↗ 0.0027\u001b[0m)\n", + " └── Mean_iou = 0.661\n", + " ├── Epoch N-1 = 0.6593 (\u001b[32m↗ 0.0017\u001b[0m)\n", + " └── Best until now = 0.6593 (\u001b[32m↗ 0.0017\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 14: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.215, background_IOU=0.748, gpu_mem=1.14, mean_IOU=0.781, target_IOU=0.813]\n", + "Validating epoch 14: 100%|██████████| 65/65 [00:17<00:00, 3.78it/s]\n", + "[2023-11-13 11:50:45] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:50:45] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.763008713722229\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 14\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2155\n", + "│ │ ├── Epoch N-1 = 0.2194 (\u001b[32m↘ -0.0039\u001b[0m)\n", + "│ │ └── Best until now = 0.2194 (\u001b[32m↘ -0.0039\u001b[0m)\n", + "│ ├── Target_iou = 0.8134\n", + "│ │ ├── Epoch N-1 = 0.8097 (\u001b[32m↗ 0.0037\u001b[0m)\n", + "│ │ └── Best until now = 0.8097 (\u001b[32m↗ 0.0037\u001b[0m)\n", + "│ ├── Background_iou = 0.7484\n", + "│ │ ├── Epoch N-1 = 0.7447 (\u001b[32m↗ 0.0038\u001b[0m)\n", + "│ │ └── Best until now = 0.7447 (\u001b[32m↗ 0.0038\u001b[0m)\n", + "│ └── Mean_iou = 0.7809\n", + "│ ├── Epoch N-1 = 0.7772 (\u001b[32m↗ 0.0037\u001b[0m)\n", + "│ └── Best until now = 0.7772 (\u001b[32m↗ 0.0037\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3015\n", + " │ ├── Epoch N-1 = 0.3024 (\u001b[32m↘ -0.0009\u001b[0m)\n", + " │ └── Best until now = 0.3024 (\u001b[32m↘ -0.0009\u001b[0m)\n", + " ├── Target_iou = 0.763\n", + " │ ├── Epoch N-1 = 0.7624 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " │ └── Best until now = 0.7624 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " ├── Background_iou = 0.5621\n", + " │ ├── Epoch N-1 = 0.5596 (\u001b[32m↗ 0.0025\u001b[0m)\n", + " │ └── Best until now = 0.5596 (\u001b[32m↗ 0.0025\u001b[0m)\n", + " └── Mean_iou = 0.6625\n", + " ├── Epoch N-1 = 0.661 (\u001b[32m↗ 0.0016\u001b[0m)\n", + " └── Best until now = 0.661 (\u001b[32m↗ 0.0016\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 11:50:47] INFO - sg_trainer.py - RUNNING ADDITIONAL TEST ON THE AVERAGED MODEL...\n", + "Validating epoch 15: 98%|█████████▊| 64/65 [00:16<00:00, 3.27it/s]" + ] + } + ], + "source": [ + "trainer.train(model=model, training_params=train_params, train_loader=train_loader, valid_loader=valid_loader)" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 13: 100%|██████████| 309/309 [01:56<00:00, 2.65it/s, BCEDiceLoss=0.223, background_IOU=0.74, gpu_mem=1.14, mean_IOU=0.774, target_IOU=0.807]\n", - "Validating epoch 13: 100%|██████████| 65/65 [00:16<00:00, 3.99it/s]\n", - "[2023-11-08 11:28:18] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:28:18] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7553263306617737\n" - ] + "cell_type": "code", + "execution_count": 12, + "metadata": { + "id": "X8BJq1crcbjl", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "661796b8-431a-4c23-ac57-9bdc579a685d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Best Checkpoint mIoU is: 0.763008713722229\n" + ] + } + ], + "source": [ + "print(\"Best Checkpoint mIoU is: \"+ str(trainer.best_metric))" + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 13\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2229\n", - "│ │ ├── Epoch N-1 = 0.2272 (\u001B[32m↘ -0.0043\u001B[0m)\n", - "│ │ └── Best until now = 0.2272 (\u001B[32m↘ -0.0043\u001B[0m)\n", - "│ ├── Target_iou = 0.8075\n", - "│ │ ├── Epoch N-1 = 0.8051 (\u001B[32m↗ 0.0023\u001B[0m)\n", - "│ │ └── Best until now = 0.8051 (\u001B[32m↗ 0.0023\u001B[0m)\n", - "│ ├── Background_iou = 0.7402\n", - "│ │ ├── Epoch N-1 = 0.7333 (\u001B[32m↗ 0.0068\u001B[0m)\n", - "│ │ └── Best until now = 0.7333 (\u001B[32m↗ 0.0068\u001B[0m)\n", - "│ └── Mean_iou = 0.7738\n", - "│ ├── Epoch N-1 = 0.7692 (\u001B[32m↗ 0.0046\u001B[0m)\n", - "│ └── Best until now = 0.7692 (\u001B[32m↗ 0.0046\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3103\n", - " │ ├── Epoch N-1 = 0.311 (\u001B[32m↘ -0.0007\u001B[0m)\n", - " │ └── Best until now = 0.311 (\u001B[32m↘ -0.0007\u001B[0m)\n", - " ├── Target_iou = 0.7553\n", - " │ ├── Epoch N-1 = 0.7549 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " │ └── Best until now = 0.7549 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " ├── Background_iou = 0.548\n", - " │ ├── Epoch N-1 = 0.5474 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " │ └── Best until now = 0.5474 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " └── Mean_iou = 0.6517\n", - " ├── Epoch N-1 = 0.6512 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " └── Best until now = 0.6512 (\u001B[32m↗ 0.0005\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "3Nybj15cchxd" + }, + "source": [ + "Now you can download your trained weights from this directory" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 14: 100%|██████████| 309/309 [01:56<00:00, 2.65it/s, BCEDiceLoss=0.219, background_IOU=0.744, gpu_mem=1.14, mean_IOU=0.777, target_IOU=0.811]\n", - "Validating epoch 14: 100%|██████████| 65/65 [00:16<00:00, 3.97it/s]\n", - "[2023-11-08 11:30:33] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244/ckpt_best.pth\n", - "[2023-11-08 11:30:33] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7558110952377319\n" - ] + "cell_type": "code", + "execution_count": 13, + "metadata": { + "id": "_iHsFgPSciQh" + }, + "outputs": [], + "source": [ + "print(trainer.checkpoints_dir_path)" + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 14\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2193\n", - "│ │ ├── Epoch N-1 = 0.2229 (\u001B[32m↘ -0.0036\u001B[0m)\n", - "│ │ └── Best until now = 0.2229 (\u001B[32m↘ -0.0036\u001B[0m)\n", - "│ ├── Target_iou = 0.8107\n", - "│ │ ├── Epoch N-1 = 0.8075 (\u001B[32m↗ 0.0033\u001B[0m)\n", - "│ │ └── Best until now = 0.8075 (\u001B[32m↗ 0.0033\u001B[0m)\n", - "│ ├── Background_iou = 0.7443\n", - "│ │ ├── Epoch N-1 = 0.7402 (\u001B[32m↗ 0.0041\u001B[0m)\n", - "│ │ └── Best until now = 0.7402 (\u001B[32m↗ 0.0041\u001B[0m)\n", - "│ └── Mean_iou = 0.7775\n", - "│ ├── Epoch N-1 = 0.7738 (\u001B[32m↗ 0.0037\u001B[0m)\n", - "│ └── Best until now = 0.7738 (\u001B[32m↗ 0.0037\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3095\n", - " │ ├── Epoch N-1 = 0.3103 (\u001B[32m↘ -0.0008\u001B[0m)\n", - " │ └── Best until now = 0.3103 (\u001B[32m↘ -0.0008\u001B[0m)\n", - " ├── Target_iou = 0.7558\n", - " │ ├── Epoch N-1 = 0.7553 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " │ └── Best until now = 0.7553 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " ├── Background_iou = 0.5487\n", - " │ ├── Epoch N-1 = 0.548 (\u001B[32m↗ 0.0007\u001B[0m)\n", - " │ └── Best until now = 0.548 (\u001B[32m↗ 0.0007\u001B[0m)\n", - " └── Mean_iou = 0.6523\n", - " ├── Epoch N-1 = 0.6517 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " └── Best until now = 0.6517 (\u001B[32m↗ 0.0006\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "yuhYeXLA18q5" + }, + "source": [ + "# 6. Predict\n" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-08 11:30:39] INFO - sg_trainer.py - RUNNING ADDITIONAL TEST ON THE AVERAGED MODEL...\n", - "Validating epoch 15: 98%|█████████▊| 64/65 [00:16<00:00, 4.31it/s]" - ] - } - ], - "source": [ - "trainer.train(model=model, training_params=train_params, train_loader=train_loader, valid_loader=valid_loader)" - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "metadata": { - "id": "X8BJq1crcbjl" - }, - "outputs": [], - "source": [ - "print(\"Best Checkpoint mIoU is: \"+ str(trainer.best_metric))" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "3Nybj15cchxd" - }, - "source": [ - "Now you can download your trained weights from this directory" - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "metadata": { - "id": "_iHsFgPSciQh" - }, - "outputs": [], - "source": [ - "print(trainer.checkpoints_dir_path)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "yuhYeXLA18q5" - }, - "source": [ - "# 6. Predict\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "VjRA1tu1mvXQ" - }, - "source": [ - "When the training is complete you can use the trained model to get predictions on the validation set, your data or some other image. Let's load some image and\n", - "run a model inference to create a binary segmentation mask." - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "metadata": { - "id": "Ads7RyGN2JwQ", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 1000 + "cell_type": "markdown", + "metadata": { + "id": "VjRA1tu1mvXQ" + }, + "source": [ + "When the training is complete you can use the trained model to get predictions on the validation set, your data or some other image. Let's load some image and\n", + "run a model inference to create a binary segmentation mask." + ] }, - "outputId": "a0da9ef8-2743-46a2-c03a-95875ab80dc8" - }, - "outputs": [ { - "output_type": "display_data", - "data": { - "text/plain": [ - "" + "cell_type": "code", + "execution_count": 17, + "metadata": { + "id": "Ads7RyGN2JwQ", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 977 + }, + "outputId": "c99ede2d-7fdd-428a-95fe-cac9afbf508b" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + } ], - "image/png": "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\n" - }, - "metadata": {} + "source": [ + "from torchvision.transforms import Compose, ToTensor, Resize, Normalize, ToPILImage\n", + "from PIL import Image\n", + "import torch\n", + "\n", + "pre_proccess = Compose([\n", + " ToTensor(),\n", + " Normalize([.485, .456, .406], [.229, .224, .225])\n", + "])\n", + "\n", + "demo_img_path = os.path.join(root_dir, \"images\", \"ache-adult-depression-expression-41253.png\")\n", + "img = Image.open(demo_img_path)\n", + "# Resize the image and display\n", + "img = Resize(size=(480, 320))(img)\n", + "display(img)\n", + "\n", + "# Run pre-proccess - transforms to tensor and apply normalizations.\n", + "img = pre_proccess(img).unsqueeze(0).cuda()\n", + "\n", + "# Run inference\n", + "model = trainer.net\n", + "model = model.eval()\n", + "mask = model(img)\n", + "\n", + "# Run post-proccess - apply sigmoid to output probabilities, then apply hard\n", + "# threshold of 0.5 for binary mask prediction.\n", + "mask = torch.sigmoid(mask).gt(0.5).squeeze()\n", + "mask = ToPILImage()(mask.float())\n", + "display(mask)\n" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "\rValidating epoch 15: 100%|██████████| 65/65 [00:16<00:00, 4.37it/s]\rValidating epoch 15: 100%|██████████| 65/65 [00:16<00:00, 3.91it/s]\n", - "[2023-11-08 11:30:56] INFO - base_sg_logger.py - [CLEANUP] - Successfully stopped system monitoring process\n", - "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py:283: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n", - " and should_run_async(code)\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "-k6ZLKHL1hIM" + }, + "source": [ + "# 7. Convert to ONNX/TensorRT" + ] }, { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "image/png": "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\n" - }, - "metadata": {} + "cell_type": "markdown", + "metadata": { + "id": "br7n55Szm4Nq" + }, + "source": [ + "Let's compile our model to ONNX." + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "Best Checkpoint mIoU is: 0.7558110952377319\n", - "/home/notebook_ckpts/segmentation_quick_start/RUN_20231108_105613_531244\n" - ] + "cell_type": "code", + "execution_count": 22, + "metadata": { + "id": "q0AGQvEf11PT", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "76b54859-3375-4fc4-c7a7-5b86ed3d80fb" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "ONNX successfully created at: /content/model.onnx\n" + ] + } + ], + "source": [ + "from onnxsim import simplify\n", + "import onnx\n", + "\n", + "onnx_path = os.path.join(os.getcwd(), \"model.onnx\")\n", + "\n", + "input_size = [1, 3, 480, 320]\n", + "model.prep_model_for_conversion(input_size=input_size)\n", + "\n", + "torch.onnx.export(model,\n", + " torch.randn(*input_size).cuda(),\n", + " onnx_path)\n", + "\n", + "# onnx simplifier\n", + "model_sim, check = simplify(onnx_path)\n", + "assert check, \"Simplified ONNX model could not be validated\"\n", + "onnx.save_model(model_sim, onnx_path)\n", + "\n", + "print(\"ONNX successfully created at: \", onnx_path)\n", + "\n" + ] } - ], - "source": [ - "from torchvision.transforms import Compose, ToTensor, Resize, Normalize, ToPILImage\n", - "from PIL import Image\n", - "import torch\n", - "\n", - "pre_proccess = Compose([\n", - " ToTensor(),\n", - " Normalize([.485, .456, .406], [.229, .224, .225])\n", - "])\n", - "\n", - "demo_img_path = \"/home/data/supervisely-persons/images/ache-adult-depression-expression-41253.png\"\n", - "img = Image.open(demo_img_path)\n", - "# Resize the image and display\n", - "img = Resize(size=(480, 320))(img)\n", - "display(img)\n", - "\n", - "# Run pre-proccess - transforms to tensor and apply normalizations.\n", - "img = pre_proccess(img).unsqueeze(0).cuda()\n", - "\n", - "# Run inference\n", - "model = trainer.net\n", - "model = model.eval()\n", - "mask = model(img)\n", - "\n", - "# Run post-proccess - apply sigmoid to output probabilities, then apply hard\n", - "# threshold of 0.5 for binary mask prediction.\n", - "mask = torch.sigmoid(mask).gt(0.5).squeeze()\n", - "mask = ToPILImage()(mask.float())\n", - "display(mask)\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "-k6ZLKHL1hIM" - }, - "source": [ - "# 7. Convert to ONNX/TensorRT" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "br7n55Szm4Nq" - }, - "source": [ - "Let's compile our model to ONNX." - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": { - "id": "dsIPbyX6GVKs", + ], + "metadata": { + "accelerator": "GPU", "colab": { - "base_uri": "https://localhost:8080/" + "provenance": [] }, - "outputId": "a2c3f05e-cb84-4fc0-db8a-e59bc72faf80" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "\u001B[31mERROR: Could not find a version that satisfies the requirement infery (from versions: none)\u001B[0m\u001B[31m\n", - "\u001B[0m\u001B[31mERROR: No matching distribution found for infery\u001B[0m\u001B[31m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m776.3/776.3 MB\u001B[0m \u001B[31m1.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", - "torchaudio 2.1.0+cu118 requires torch==2.1.0, but you have torch 1.12.0 which is incompatible.\n", - "torchdata 0.7.0 requires torch==2.1.0, but you have torch 1.12.0 which is incompatible.\n", - "torchtext 0.16.0 requires torch==2.1.0, but you have torch 1.12.0 which is incompatible.\n", - "torchvision 0.16.0+cu118 requires torch==2.1.0, but you have torch 1.12.0 which is incompatible.\u001B[0m\u001B[31m\n", - "\u001B[0m" - ] + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" } - ], - "source": [ - "! pip install -qq onnx-simplifier\n", - "! pip install -qq torch==1.12" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "id": "q0AGQvEf11PT" - }, - "outputs": [], - "source": [ - "from onnxsim import simplify\n", - "import onnx\n", - "\n", - "onnx_path = \"/home/data/model.onnx\"\n", - "\n", - "input_size = [1, 3, 480, 320]\n", - "model.prep_model_for_conversion(input_size=input_size)\n", - "\n", - "torch.onnx.export(model,\n", - " torch.randn(*input_size).cuda(),\n", - " onnx_path,\n", - " opset_version=11)\n", - "\n", - "# onnx simplifier\n", - "model_sim, check = simplify(onnx_path)\n", - "assert check, \"Simplified ONNX model could not be validated\"\n", - "onnx.save_model(model_sim, onnx_path)\n", - "\n", - "print(\"ONNX successfully created at: \", onnx_path)\n", - "\n" - ] - } - ], - "metadata": { - "accelerator": "GPU", - "colab": { - "provenance": [] - }, - "kernelspec": { - "display_name": "Python 3", - "name": "python3" }, - "language_info": { - "name": "python" - } - }, - "nbformat": 4, - "nbformat_minor": 0 + "nbformat": 4, + "nbformat_minor": 0 } diff --git a/notebooks/transfer_learning_semantic_segmentation.ipynb b/notebooks/transfer_learning_semantic_segmentation.ipynb index 3995333ed8..108ea92672 100644 --- a/notebooks/transfer_learning_semantic_segmentation.ipynb +++ b/notebooks/transfer_learning_semantic_segmentation.ipynb @@ -1614,8 +1614,6 @@ "id": "br7n55Szm4Nq" }, "source": [ - "SG is a production ready library. All the models implemented in SG can be compiled to ONNX and TensorRT. Deci also offers the [Infery](https://docs.deci.ai/docs/installing-infery) library that allows to do inference on models saved in various frameworks with the same API regardless of a framework.\n", - "\n", "Let's compile our model to ONNX." ] }, From ad7b32764bc20a58b006dba7c15e58913193269d Mon Sep 17 00:00:00 2001 From: shayaharon Date: Mon, 13 Nov 2023 14:54:40 +0200 Subject: [PATCH 5/5] updated all notebooks --- notebooks/quickstart_segmentation.ipynb | 2518 +++++++------- .../segmentation_connect_custom_dataset.ipynb | 507 +-- ...nsfer_learning_semantic_segmentation.ipynb | 3076 ++++++++--------- 3 files changed, 3016 insertions(+), 3085 deletions(-) diff --git a/notebooks/quickstart_segmentation.ipynb b/notebooks/quickstart_segmentation.ipynb index 003c6b8c7a..04386d76ab 100644 --- a/notebooks/quickstart_segmentation.ipynb +++ b/notebooks/quickstart_segmentation.ipynb @@ -1,1342 +1,1342 @@ { - "cells": [ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "HY_HuQbxn7X0" + }, + "source": [ + "![SG - Horizontal.png](data:image/png;base64,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)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "oA_p5zIsoAJQ" + }, + "source": [ + "# SuperGradients quick start Semantic Segmentation\n", + "\n", + "In this tutorial we will train PPLiteSeg model on Supervisely semantic segmentation dataset\n", + "\n", + "The notebook is divided into 7 sections:\n", + "1. Experiment setup\n", + "2. Dataset definition\n", + "3. Architecture definition\n", + "4. Training setup\n", + "5. Training and Evaluation\n", + "6. Predict\n", + "7. Convert to ONNX\\TensorRT" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "GqH4VGMroWec" + }, + "source": [ + "#Install SG" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Q8uA6AWEhHN6" + }, + "source": [ + "The cell below will install **super_gradients** which will automatically get all its dependencies." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "id": "-mm-E4xRoNEm" + }, + "outputs": [], + "source": [ + "! pip install -qq super-gradients==3.4.1" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "892xArqDsGsQ" + }, + "source": [ + "# 1. Experiment setup\n", + "We will initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", + "\n", + "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", + "\n", + "```\n", + "ckpt_root_dir\n", + "|─── experiment_name_1\n", + "│ ckpt_best.pth # Model checkpoint on best epoch\n", + "│ ckpt_latest.pth # Model checkpoint on last epoch\n", + "│ average_model.pth # Model checkpoint averaged over epochs\n", + "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", + "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", + "└─── experiment_name_2\n", + " ...\n", + "```\n", + "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", + " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "pl0WPz1HisFz" + }, + "source": [ + "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "HAff--HysJmP", + "outputId": "63e96426-a29b-4cdc-9a72-60da27d6aaa7" + }, + "outputs": [ { - "cell_type": "markdown", - "metadata": { - "id": "HY_HuQbxn7X0" - }, - "source": [ - "![SG - Horizontal.png](data:image/png;base64,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)" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is logged into /root/sg_logs/console.log\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "oA_p5zIsoAJQ" - }, - "source": [ - "# SuperGradients quick start Semantic Segmentation\n", - "\n", - "In this tutorial we will train PPLiteSeg model on Supervisely semantic segmentation dataset\n", - "\n", - "The notebook is divided into 7 sections:\n", - "1. Experiment setup\n", - "2. Dataset definition\n", - "3. Architecture definition\n", - "4. Training setup\n", - "5. Training and Evaluation\n", - "6. Predict\n", - "7. Convert to ONNX\\TensorRT" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 11:11:11] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n", + "[2023-11-13 11:11:11] WARNING - __init__.py - Failed to import pytorch_quantization\n", + "[2023-11-13 11:11:11] INFO - utils.py - NumExpr defaulting to 2 threads.\n", + "[2023-11-13 11:11:23] WARNING - calibrator.py - Failed to import pytorch_quantization\n", + "[2023-11-13 11:11:23] WARNING - export.py - Failed to import pytorch_quantization\n", + "[2023-11-13 11:11:23] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n" + ] + } + ], + "source": [ + "from super_gradients import Trainer\n", + "\n", + "CHECKPOINT_DIR = './notebook_ckpts/'\n", + "trainer = Trainer(experiment_name=\"segmentation_quick_start\", ckpt_root_dir=CHECKPOINT_DIR)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "dwVMY4gMjQSL" + }, + "source": [ + "# 2. Dataset definition\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fpIWhnR9j2rm" + }, + "source": [ + "\n", + "For the sake of this presentation, we'll use **Supervisely** semantic segmentation dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ZACgRb-qjzDJ" + }, + "source": [ + "SG trainer is fully compatible with PyTorch data loaders, so you can definitely use your own data for the experiment below if you prefer." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6ulV6Hpao3IN" + }, + "source": [ + "## 2.1 Download data\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mVwslNv-j-2C" + }, + "source": [ + "Feel free to change the download path by editing SUPERVISELY_DATASET_DOWNLOAD_PATH" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "id": "dfR18Rmbo00y" + }, + "outputs": [], + "source": [ + "import os\n", + "\n", + "SUPERVISELY_DATASET_DOWNLOAD_PATH=os.path.join(os.getcwd(),\"data\")\n", + "\n", + "supervisely_dataset_dir_path = os.path.join(SUPERVISELY_DATASET_DOWNLOAD_PATH, 'supervisely-persons')\n", + "\n", + "if os.path.isdir(supervisely_dataset_dir_path):\n", + " print('supervisely dataset already downloaded...')\n", + "else:\n", + " print('Downloading and extracting supervisely dataset to: ' + SUPERVISELY_DATASET_DOWNLOAD_PATH)\n", + " ! mkdir $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " %cd $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " ! wget https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", + " ! unzip --qq supervisely-persons.zip" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": false, + "id": "V9ZcklupX8Qx" + }, + "source": [ + "## 2.2 Create data loaders\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3Mk_YixjlEhj" + }, + "source": [ + "The dataloaders are initiated with the default parameters defined in the [yaml](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/recipes/dataset_params/supervisely_persons_dataset_params.yaml)\n", + "file. Parameters as batch_size, transforms, root_dir and others can be overridden by passing as `dataset_params` and\n", + "`dataloader_params`, as implemented bellow." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "S3BzMRhSX8Qx", + "colab": { + "base_uri": "https://localhost:8080/" }, + "outputId": "87b5092d-fe93-4c0a-8b2e-febe215b52bd" + }, + "outputs": [ { - "cell_type": "markdown", - "metadata": { - "id": "GqH4VGMroWec" - }, - "source": [ - "#Install SG" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "supervisely dataset already downloaded...\n" + ] + } + ], + "source": [ + "from super_gradients.training import dataloaders\n", + "root_dir = supervisely_dataset_dir_path\n", + "batch_size = 8\n", + "\n", + "train_loader = dataloaders.supervisely_persons_train(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})\n", + "valid_loader = dataloaders.supervisely_persons_val(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6dHIwvs46-dk" + }, + "source": [ + "As you can see, we didn't have to pass many parameters into the dataloaders construction. That's because defaults are pre-defined for your convenience, and you might be curious to know what they are. Let's print them and see which resolution and transformations are defined." + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "id": "76tzhKxi6aS-", + "colab": { + "base_uri": "https://localhost:8080/" }, + "outputId": "3b5c8f34-673c-4f4c-d243-80e82c347f3d" + }, + "outputs": [ { - "cell_type": "markdown", - "metadata": { - "id": "Q8uA6AWEhHN6" - }, - "source": [ - "The cell below will install **super_gradients** which will automatically get all its dependencies." - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "Dataloader parameters:\n", + "{'batch_size': 8, 'shuffle': True, 'drop_last': True}\n", + "Dataset parameters\n", + "{'root_dir': '/content/data/supervisely-persons', 'list_file': 'train.csv', 'cache_labels': False, 'cache_images': False, 'transforms': [{'SegRandomRescale': {'scales': [0.25, 1.0]}}, {'SegColorJitter': {'brightness': 0.5, 'contrast': 0.5, 'saturation': 0.5}}, {'SegRandomFlip': {'prob': 0.5}}, {'SegPadShortToCropSize': {'crop_size': [320, 480], 'fill_mask': 0}}, {'SegCropImageAndMask': {'crop_size': [320, 480], 'mode': 'random'}}]}\n" + ] + } + ], + "source": [ + "print('Dataloader parameters:')\n", + "print(train_loader.dataloader_params)\n", + "print('Dataset parameters')\n", + "print(train_loader.dataset.dataset_params)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "l5GcDAg_pUGJ" + }, + "source": [ + "# 3. Architecture definition\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "id": "xXPMJQCJzmb4" + }, + "outputs": [], + "source": [ + "from super_gradients.training import models\n", + "from super_gradients.common.object_names import Models\n", + "\n", + "model = models.get(model_name=Models.PP_LITE_T_SEG,\n", + " arch_params={\"use_aux_heads\": False},\n", + " num_classes=1)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fU8orO7wlwIK" + }, + "source": [ + "SG includes implementations of many different architectures for semantic segmentation tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-oGSU3V8lqcm" + }, + "source": [ + "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", + "and extra Auxiliary heads aren't used for training.\n", + "\n", + "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "X-_dBewgr1dG" + }, + "source": [ + "# 4. Training setup\n", + "\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "H1Rll8Orl-Dy" + }, + "source": [ + "\n", + "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n", + "\n", + "We will be using an average of BCE and Dice loss for segmentation, with different learning rates for the replaced segmentation head layer, and the rest of the network- this is controlled by the `multiply_head_lr` parameter which is the multiplication factor of the learning rate for the newly replaced layer.\n", + "\n", + "As our `metric_to_watch`, we will be monitoring the `target_IOU` which is one of the components of `BinaryIOU` torchmetrics object (the other components are `mean_IOU` which is the mean of the background and target IOUs, and `background_IOU`)." + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "id": "NShu3zLgr5qD" + }, + "outputs": [], + "source": [ + "from super_gradients.training.metrics.segmentation_metrics import BinaryIOU\n", + "\n", + "train_params = {\"max_epochs\": 15,\n", + " \"lr_mode\": \"cosine\",\n", + " \"initial_lr\": 0.01,\n", + " \"lr_warmup_epochs\": 5,\n", + " \"multiply_head_lr\": 10,\n", + " \"optimizer\": \"SGD\",\n", + " \"loss\": \"BCEDiceLoss\",\n", + " \"ema\": True,\n", + " \"ema_params\":\n", + " {\n", + " \"decay\": 0.9999,\n", + " \"decay_type\": \"exp\",\n", + " \"beta\": 15,\n", + " },\n", + "\n", + " \"zero_weight_decay_on_bias_and_bn\": True,\n", + " \"average_best_models\": True,\n", + " \"metric_to_watch\": \"target_IOU\",\n", + " \"greater_metric_to_watch_is_better\": True,\n", + " \"train_metrics_list\": [BinaryIOU()],\n", + " \"valid_metrics_list\": [BinaryIOU()],\n", + " \"loss_logging_items_names\": [\"loss\"]\n", + " }" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qTECVyhcs506" + }, + "source": [ + "# 5. Training and evaluation\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "S1K5MU2kmmDb" + }, + "source": [ + "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", + "\n", + "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "id": "u6roEj9ktFTi", + "colab": { + "base_uri": "https://localhost:8080/" }, + "outputId": "4a295f63-f0c4-43a7-c6e8-2f7ffd1b5ce2" + }, + "outputs": [ { - "cell_type": "code", - "execution_count": 1, - "metadata": { - "id": "-mm-E4xRoNEm" - }, - "outputs": [], - "source": [ - "! pip install -qq super-gradients==3.4.1" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 11:15:07] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231113_111507_197271`\n", + "[2023-11-13 11:15:07] INFO - sg_trainer.py - Checkpoints directory: ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271\n", + "[2023-11-13 11:15:07] INFO - sg_trainer.py - Using EMA with params {'decay': 0.9999, 'decay_type': 'exp', 'beta': 15}\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "892xArqDsGsQ" - }, - "source": [ - "# 1. Experiment setup\n", - "We will initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", - "\n", - "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", - "\n", - "```\n", - "ckpt_root_dir\n", - "|─── experiment_name_1\n", - "│ ckpt_best.pth # Model checkpoint on best epoch\n", - "│ ckpt_latest.pth # Model checkpoint on last epoch\n", - "│ average_model.pth # Model checkpoint averaged over epochs\n", - "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", - "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", - "└─── experiment_name_2\n", - " ...\n", - "```\n", - "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", - " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`." - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is now moved to ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/console_Nov13_11_15_07.txt\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "pl0WPz1HisFz" - }, - "source": [ - "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 11:15:08] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", + " - Mode: Single GPU\n", + " - Number of GPUs: 1 (1 available on the machine)\n", + " - Full dataset size: 2477 (len(train_set))\n", + " - Batch size per GPU: 8 (batch_size)\n", + " - Batch Accumulate: 1 (batch_accumulate)\n", + " - Total batch size: 8 (num_gpus * batch_size)\n", + " - Effective Batch size: 8 (num_gpus * batch_size * batch_accumulate)\n", + " - Iterations per epoch: 309 (len(train_loader))\n", + " - Gradient updates per epoch: 309 (len(train_loader) / batch_accumulate)\n", + "\n", + "[2023-11-13 11:15:08] INFO - sg_trainer.py - Started training for 15 epochs (0/14)\n", + "\n", + "Train epoch 0: 100%|██████████| 309/309 [02:12<00:00, 2.33it/s, BCEDiceLoss=0.4, background_IOU=0.545, gpu_mem=1.14, mean_IOU=0.609, target_IOU=0.674]\n", + "Validating: 100%|██████████| 65/65 [00:17<00:00, 3.69it/s]\n", + "[2023-11-13 11:17:39] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:17:39] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.6779429912567139\n" + ] }, { - "cell_type": "code", - "execution_count": 2, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "HAff--HysJmP", - "outputId": "63e96426-a29b-4cdc-9a72-60da27d6aaa7" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "The console stream is logged into /root/sg_logs/console.log\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-13 11:11:11] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n", - "[2023-11-13 11:11:11] WARNING - __init__.py - Failed to import pytorch_quantization\n", - "[2023-11-13 11:11:11] INFO - utils.py - NumExpr defaulting to 2 threads.\n", - "[2023-11-13 11:11:23] WARNING - calibrator.py - Failed to import pytorch_quantization\n", - "[2023-11-13 11:11:23] WARNING - export.py - Failed to import pytorch_quantization\n", - "[2023-11-13 11:11:23] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n" - ] - } - ], - "source": [ - "from super_gradients import Trainer\n", - "\n", - "CHECKPOINT_DIR = './notebook_ckpts/'\n", - "trainer = Trainer(experiment_name=\"segmentation_quick_start\", ckpt_root_dir=CHECKPOINT_DIR)" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 0\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.4001\n", + "│ ├── Target_iou = 0.6736\n", + "│ ├── Background_iou = 0.5448\n", + "│ └── Mean_iou = 0.6092\n", + "└── Validation\n", + " ├── Bcediceloss = 0.4166\n", + " ├── Target_iou = 0.6779\n", + " ├── Background_iou = 0.4039\n", + " └── Mean_iou = 0.5409\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "dwVMY4gMjQSL" - }, - "source": [ - "# 2. Dataset definition\n" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 1: 100%|██████████| 309/309 [02:05<00:00, 2.46it/s, BCEDiceLoss=0.338, background_IOU=0.604, gpu_mem=1.14, mean_IOU=0.661, target_IOU=0.719]\n", + "Validating epoch 1: 100%|██████████| 65/65 [00:17<00:00, 3.69it/s]\n", + "[2023-11-13 11:20:05] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:20:05] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7205255031585693\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "fpIWhnR9j2rm" - }, - "source": [ - "\n", - "For the sake of this presentation, we'll use **Supervisely** semantic segmentation dataset." - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 1\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.3381\n", + "│ │ ├── Epoch N-1 = 0.4001 (\u001B[32m↘ -0.062\u001B[0m)\n", + "│ │ └── Best until now = 0.4001 (\u001B[32m↘ -0.062\u001B[0m)\n", + "│ ├── Target_iou = 0.7193\n", + "│ │ ├── Epoch N-1 = 0.6736 (\u001B[32m↗ 0.0457\u001B[0m)\n", + "│ │ └── Best until now = 0.6736 (\u001B[32m↗ 0.0457\u001B[0m)\n", + "│ ├── Background_iou = 0.6036\n", + "│ │ ├── Epoch N-1 = 0.5448 (\u001B[32m↗ 0.0587\u001B[0m)\n", + "│ │ └── Best until now = 0.5448 (\u001B[32m↗ 0.0587\u001B[0m)\n", + "│ └── Mean_iou = 0.6614\n", + "│ ├── Epoch N-1 = 0.6092 (\u001B[32m↗ 0.0522\u001B[0m)\n", + "│ └── Best until now = 0.6092 (\u001B[32m↗ 0.0522\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3578\n", + " │ ├── Epoch N-1 = 0.4166 (\u001B[32m↘ -0.0588\u001B[0m)\n", + " │ └── Best until now = 0.4166 (\u001B[32m↘ -0.0588\u001B[0m)\n", + " ├── Target_iou = 0.7205\n", + " │ ├── Epoch N-1 = 0.6779 (\u001B[32m↗ 0.0426\u001B[0m)\n", + " │ └── Best until now = 0.6779 (\u001B[32m↗ 0.0426\u001B[0m)\n", + " ├── Background_iou = 0.4497\n", + " │ ├── Epoch N-1 = 0.4039 (\u001B[32m↗ 0.0458\u001B[0m)\n", + " │ └── Best until now = 0.4039 (\u001B[32m↗ 0.0458\u001B[0m)\n", + " └── Mean_iou = 0.5851\n", + " ├── Epoch N-1 = 0.5409 (\u001B[32m↗ 0.0442\u001B[0m)\n", + " └── Best until now = 0.5409 (\u001B[32m↗ 0.0442\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "ZACgRb-qjzDJ" - }, - "source": [ - "SG trainer is fully compatible with PyTorch data loaders, so you can definitely use your own data for the experiment below if you prefer." - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 2: 100%|██████████| 309/309 [02:00<00:00, 2.55it/s, BCEDiceLoss=0.32, background_IOU=0.634, gpu_mem=1.14, mean_IOU=0.684, target_IOU=0.734]\n", + "Validating epoch 2: 100%|██████████| 65/65 [00:16<00:00, 3.84it/s]\n", + "[2023-11-13 11:22:24] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:22:24] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7300039529800415\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "6ulV6Hpao3IN" - }, - "source": [ - "## 2.A. Download data\n" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 2\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.3199\n", + "│ │ ├── Epoch N-1 = 0.3381 (\u001B[32m↘ -0.0182\u001B[0m)\n", + "│ │ └── Best until now = 0.3381 (\u001B[32m↘ -0.0182\u001B[0m)\n", + "│ ├── Target_iou = 0.734\n", + "│ │ ├── Epoch N-1 = 0.7193 (\u001B[32m↗ 0.0147\u001B[0m)\n", + "│ │ └── Best until now = 0.7193 (\u001B[32m↗ 0.0147\u001B[0m)\n", + "│ ├── Background_iou = 0.6344\n", + "│ │ ├── Epoch N-1 = 0.6036 (\u001B[32m↗ 0.0308\u001B[0m)\n", + "│ │ └── Best until now = 0.6036 (\u001B[32m↗ 0.0308\u001B[0m)\n", + "│ └── Mean_iou = 0.6842\n", + "│ ├── Epoch N-1 = 0.6614 (\u001B[32m↗ 0.0227\u001B[0m)\n", + "│ └── Best until now = 0.6614 (\u001B[32m↗ 0.0227\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.357\n", + " │ ├── Epoch N-1 = 0.3578 (\u001B[32m↘ -0.0008\u001B[0m)\n", + " │ └── Best until now = 0.3578 (\u001B[32m↘ -0.0008\u001B[0m)\n", + " ├── Target_iou = 0.73\n", + " │ ├── Epoch N-1 = 0.7205 (\u001B[32m↗ 0.0095\u001B[0m)\n", + " │ └── Best until now = 0.7205 (\u001B[32m↗ 0.0095\u001B[0m)\n", + " ├── Background_iou = 0.4503\n", + " │ ├── Epoch N-1 = 0.4497 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.4497 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " └── Mean_iou = 0.5902\n", + " ├── Epoch N-1 = 0.5851 (\u001B[32m↗ 0.0051\u001B[0m)\n", + " └── Best until now = 0.5851 (\u001B[32m↗ 0.0051\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "mVwslNv-j-2C" - }, - "source": [ - "Feel free to change the download path by editing SUPERVISELY_DATASET_DOWNLOAD_PATH" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 3: 100%|██████████| 309/309 [01:59<00:00, 2.58it/s, BCEDiceLoss=0.302, background_IOU=0.645, gpu_mem=1.14, mean_IOU=0.697, target_IOU=0.75]\n", + "Validating epoch 3: 100%|██████████| 65/65 [00:16<00:00, 3.84it/s]\n", + "[2023-11-13 11:24:43] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:24:43] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7432040572166443\n" + ] }, { - "cell_type": "code", - "execution_count": 3, - "metadata": { - "id": "dfR18Rmbo00y" - }, - "outputs": [], - "source": [ - "import os\n", - "\n", - "SUPERVISELY_DATASET_DOWNLOAD_PATH=os.path.join(os.getcwd(),\"data\")\n", - "\n", - "supervisely_dataset_dir_path = os.path.join(SUPERVISELY_DATASET_DOWNLOAD_PATH, 'supervisely-persons')\n", - "\n", - "if os.path.isdir(supervisely_dataset_dir_path):\n", - " print('supervisely dataset already downloaded...')\n", - "else:\n", - " print('Downloading and extracting supervisely dataset to: ' + SUPERVISELY_DATASET_DOWNLOAD_PATH)\n", - " ! mkdir $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", - " %cd $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", - " ! wget https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", - " ! unzip --qq supervisely-persons.zip" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 3\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.3022\n", + "│ │ ├── Epoch N-1 = 0.3199 (\u001B[32m↘ -0.0177\u001B[0m)\n", + "│ │ └── Best until now = 0.3199 (\u001B[32m↘ -0.0177\u001B[0m)\n", + "│ ├── Target_iou = 0.7501\n", + "│ │ ├── Epoch N-1 = 0.734 (\u001B[32m↗ 0.0161\u001B[0m)\n", + "│ │ └── Best until now = 0.734 (\u001B[32m↗ 0.0161\u001B[0m)\n", + "│ ├── Background_iou = 0.6447\n", + "│ │ ├── Epoch N-1 = 0.6344 (\u001B[32m↗ 0.0103\u001B[0m)\n", + "│ │ └── Best until now = 0.6344 (\u001B[32m↗ 0.0103\u001B[0m)\n", + "│ └── Mean_iou = 0.6974\n", + "│ ├── Epoch N-1 = 0.6842 (\u001B[32m↗ 0.0132\u001B[0m)\n", + "│ └── Best until now = 0.6842 (\u001B[32m↗ 0.0132\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3307\n", + " │ ├── Epoch N-1 = 0.357 (\u001B[32m↘ -0.0263\u001B[0m)\n", + " │ └── Best until now = 0.357 (\u001B[32m↘ -0.0263\u001B[0m)\n", + " ├── Target_iou = 0.7432\n", + " │ ├── Epoch N-1 = 0.73 (\u001B[32m↗ 0.0132\u001B[0m)\n", + " │ └── Best until now = 0.73 (\u001B[32m↗ 0.0132\u001B[0m)\n", + " ├── Background_iou = 0.4794\n", + " │ ├── Epoch N-1 = 0.4503 (\u001B[32m↗ 0.0291\u001B[0m)\n", + " │ └── Best until now = 0.4503 (\u001B[32m↗ 0.0291\u001B[0m)\n", + " └── Mean_iou = 0.6113\n", + " ├── Epoch N-1 = 0.5902 (\u001B[32m↗ 0.0212\u001B[0m)\n", + " └── Best until now = 0.5902 (\u001B[32m↗ 0.0212\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "collapsed": false, - "id": "V9ZcklupX8Qx" - }, - "source": [ - "## 2.B. Create data loaders\n" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 4: 100%|██████████| 309/309 [02:00<00:00, 2.56it/s, BCEDiceLoss=0.287, background_IOU=0.67, gpu_mem=1.14, mean_IOU=0.715, target_IOU=0.76]\n", + "Validating epoch 4: 100%|██████████| 65/65 [00:17<00:00, 3.79it/s]\n", + "[2023-11-13 11:27:02] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:27:02] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7445915341377258\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "3Mk_YixjlEhj" - }, - "source": [ - "The dataloaders are initiated with the default parameters defined in the [yaml](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/recipes/dataset_params/supervisely_persons_dataset_params.yaml)\n", - "file. Parameters as batch_size, transforms, root_dir and others can be overridden by passing as `dataset_params` and\n", - "`dataloader_params`, as implemented bellow." - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 4\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2867\n", + "│ │ ├── Epoch N-1 = 0.3022 (\u001B[32m↘ -0.0155\u001B[0m)\n", + "│ │ └── Best until now = 0.3022 (\u001B[32m↘ -0.0155\u001B[0m)\n", + "│ ├── Target_iou = 0.7604\n", + "│ │ ├── Epoch N-1 = 0.7501 (\u001B[32m↗ 0.0103\u001B[0m)\n", + "│ │ └── Best until now = 0.7501 (\u001B[32m↗ 0.0103\u001B[0m)\n", + "│ ├── Background_iou = 0.6697\n", + "│ │ ├── Epoch N-1 = 0.6447 (\u001B[32m↗ 0.0251\u001B[0m)\n", + "│ │ └── Best until now = 0.6447 (\u001B[32m↗ 0.0251\u001B[0m)\n", + "│ └── Mean_iou = 0.715\n", + "│ ├── Epoch N-1 = 0.6974 (\u001B[32m↗ 0.0177\u001B[0m)\n", + "│ └── Best until now = 0.6974 (\u001B[32m↗ 0.0177\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3281\n", + " │ ├── Epoch N-1 = 0.3307 (\u001B[32m↘ -0.0026\u001B[0m)\n", + " │ └── Best until now = 0.3307 (\u001B[32m↘ -0.0026\u001B[0m)\n", + " ├── Target_iou = 0.7446\n", + " │ ├── Epoch N-1 = 0.7432 (\u001B[32m↗ 0.0014\u001B[0m)\n", + " │ └── Best until now = 0.7432 (\u001B[32m↗ 0.0014\u001B[0m)\n", + " ├── Background_iou = 0.4869\n", + " │ ├── Epoch N-1 = 0.4794 (\u001B[32m↗ 0.0074\u001B[0m)\n", + " │ └── Best until now = 0.4794 (\u001B[32m↗ 0.0074\u001B[0m)\n", + " └── Mean_iou = 0.6157\n", + " ├── Epoch N-1 = 0.6113 (\u001B[32m↗ 0.0044\u001B[0m)\n", + " └── Best until now = 0.6113 (\u001B[32m↗ 0.0044\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "code", - "execution_count": 4, - "metadata": { - "id": "S3BzMRhSX8Qx", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "87b5092d-fe93-4c0a-8b2e-febe215b52bd" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "supervisely dataset already downloaded...\n" - ] - } - ], - "source": [ - "from super_gradients.training import dataloaders\n", - "root_dir = supervisely_dataset_dir_path\n", - "batch_size = 8\n", - "\n", - "train_loader = dataloaders.supervisely_persons_train(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})\n", - "valid_loader = dataloaders.supervisely_persons_val(dataset_params={\"root_dir\": root_dir}, dataloader_params={\"batch_size\": batch_size})" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 5: 100%|██████████| 309/309 [02:02<00:00, 2.53it/s, BCEDiceLoss=0.287, background_IOU=0.664, gpu_mem=1.14, mean_IOU=0.712, target_IOU=0.761]\n", + "Validating epoch 5: 100%|██████████| 65/65 [00:17<00:00, 3.75it/s]\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "6dHIwvs46-dk" - }, - "source": [ - "As you can see, we didn't have to pass many parameters into the dataloaders construction. That's because defaults are pre-defined for your convenience, and you might be curious to know what they are. Let's print them and see which resolution and transformations are defined." - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 5\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2869\n", + "│ │ ├── Epoch N-1 = 0.2867 (\u001B[31m↗ 1e-04\u001B[0m)\n", + "│ │ └── Best until now = 0.2867 (\u001B[31m↗ 1e-04\u001B[0m)\n", + "│ ├── Target_iou = 0.7606\n", + "│ │ ├── Epoch N-1 = 0.7604 (\u001B[32m↗ 0.0002\u001B[0m)\n", + "│ │ └── Best until now = 0.7604 (\u001B[32m↗ 0.0002\u001B[0m)\n", + "│ ├── Background_iou = 0.6637\n", + "│ │ ├── Epoch N-1 = 0.6697 (\u001B[31m↘ -0.0061\u001B[0m)\n", + "│ │ └── Best until now = 0.6697 (\u001B[31m↘ -0.0061\u001B[0m)\n", + "│ └── Mean_iou = 0.7121\n", + "│ ├── Epoch N-1 = 0.715 (\u001B[31m↘ -0.0029\u001B[0m)\n", + "│ └── Best until now = 0.715 (\u001B[31m↘ -0.0029\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3339\n", + " │ ├── Epoch N-1 = 0.3281 (\u001B[31m↗ 0.0059\u001B[0m)\n", + " │ └── Best until now = 0.3281 (\u001B[31m↗ 0.0059\u001B[0m)\n", + " ├── Target_iou = 0.7402\n", + " │ ├── Epoch N-1 = 0.7446 (\u001B[31m↘ -0.0044\u001B[0m)\n", + " │ └── Best until now = 0.7446 (\u001B[31m↘ -0.0044\u001B[0m)\n", + " ├── Background_iou = 0.4593\n", + " │ ├── Epoch N-1 = 0.4869 (\u001B[31m↘ -0.0276\u001B[0m)\n", + " │ └── Best until now = 0.4869 (\u001B[31m↘ -0.0276\u001B[0m)\n", + " └── Mean_iou = 0.5997\n", + " ├── Epoch N-1 = 0.6157 (\u001B[31m↘ -0.016\u001B[0m)\n", + " └── Best until now = 0.6157 (\u001B[31m↘ -0.016\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "code", - "execution_count": 7, - "metadata": { - "id": "76tzhKxi6aS-", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "3b5c8f34-673c-4f4c-d243-80e82c347f3d" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Dataloader parameters:\n", - "{'batch_size': 8, 'shuffle': True, 'drop_last': True}\n", - "Dataset parameters\n", - "{'root_dir': '/content/data/supervisely-persons', 'list_file': 'train.csv', 'cache_labels': False, 'cache_images': False, 'transforms': [{'SegRandomRescale': {'scales': [0.25, 1.0]}}, {'SegColorJitter': {'brightness': 0.5, 'contrast': 0.5, 'saturation': 0.5}}, {'SegRandomFlip': {'prob': 0.5}}, {'SegPadShortToCropSize': {'crop_size': [320, 480], 'fill_mask': 0}}, {'SegCropImageAndMask': {'crop_size': [320, 480], 'mode': 'random'}}]}\n" - ] - } - ], - "source": [ - "print('Dataloader parameters:')\n", - "print(train_loader.dataloader_params)\n", - "print('Dataset parameters')\n", - "print(train_loader.dataset.dataset_params)" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 6: 100%|██████████| 309/309 [02:03<00:00, 2.50it/s, BCEDiceLoss=0.269, background_IOU=0.689, gpu_mem=1.14, mean_IOU=0.731, target_IOU=0.772]\n", + "Validating epoch 6: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "l5GcDAg_pUGJ" - }, - "source": [ - "# 3. Architecture definition\n", - "\n" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 6\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2686\n", + "│ │ ├── Epoch N-1 = 0.2869 (\u001B[32m↘ -0.0183\u001B[0m)\n", + "│ │ └── Best until now = 0.2867 (\u001B[32m↘ -0.0181\u001B[0m)\n", + "│ ├── Target_iou = 0.7721\n", + "│ │ ├── Epoch N-1 = 0.7606 (\u001B[32m↗ 0.0115\u001B[0m)\n", + "│ │ └── Best until now = 0.7606 (\u001B[32m↗ 0.0115\u001B[0m)\n", + "│ ├── Background_iou = 0.6892\n", + "│ │ ├── Epoch N-1 = 0.6637 (\u001B[32m↗ 0.0255\u001B[0m)\n", + "│ │ └── Best until now = 0.6697 (\u001B[32m↗ 0.0194\u001B[0m)\n", + "│ └── Mean_iou = 0.7306\n", + "│ ├── Epoch N-1 = 0.7121 (\u001B[32m↗ 0.0185\u001B[0m)\n", + "│ └── Best until now = 0.715 (\u001B[32m↗ 0.0156\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3278\n", + " │ ├── Epoch N-1 = 0.3339 (\u001B[32m↘ -0.0061\u001B[0m)\n", + " │ └── Best until now = 0.3281 (\u001B[32m↘ -0.0003\u001B[0m)\n", + " ├── Target_iou = 0.7431\n", + " │ ├── Epoch N-1 = 0.7402 (\u001B[32m↗ 0.003\u001B[0m)\n", + " │ └── Best until now = 0.7446 (\u001B[31m↘ -0.0015\u001B[0m)\n", + " ├── Background_iou = 0.4733\n", + " │ ├── Epoch N-1 = 0.4593 (\u001B[32m↗ 0.0139\u001B[0m)\n", + " │ └── Best until now = 0.4869 (\u001B[31m↘ -0.0136\u001B[0m)\n", + " └── Mean_iou = 0.6082\n", + " ├── Epoch N-1 = 0.5997 (\u001B[32m↗ 0.0085\u001B[0m)\n", + " └── Best until now = 0.6157 (\u001B[31m↘ -0.0075\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "code", - "execution_count": 8, - "metadata": { - "id": "xXPMJQCJzmb4" - }, - "outputs": [], - "source": [ - "from super_gradients.training import models\n", - "from super_gradients.common.object_names import Models\n", - "\n", - "model = models.get(model_name=Models.PP_LITE_T_SEG,\n", - " arch_params={\"use_aux_heads\": False},\n", - " num_classes=1)" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 7: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.259, background_IOU=0.701, gpu_mem=1.14, mean_IOU=0.741, target_IOU=0.781]\n", + "Validating epoch 7: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", + "[2023-11-13 11:34:05] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:34:05] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7548585534095764\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "fU8orO7wlwIK" - }, - "source": [ - "SG includes implementations of many different architectures for semantic segmentation tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 7\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.259\n", + "│ │ ├── Epoch N-1 = 0.2686 (\u001B[32m↘ -0.0096\u001B[0m)\n", + "│ │ └── Best until now = 0.2686 (\u001B[32m↘ -0.0096\u001B[0m)\n", + "│ ├── Target_iou = 0.7808\n", + "│ │ ├── Epoch N-1 = 0.7721 (\u001B[32m↗ 0.0087\u001B[0m)\n", + "│ │ └── Best until now = 0.7721 (\u001B[32m↗ 0.0087\u001B[0m)\n", + "│ ├── Background_iou = 0.7009\n", + "│ │ ├── Epoch N-1 = 0.6892 (\u001B[32m↗ 0.0117\u001B[0m)\n", + "│ │ └── Best until now = 0.6892 (\u001B[32m↗ 0.0117\u001B[0m)\n", + "│ └── Mean_iou = 0.7409\n", + "│ ├── Epoch N-1 = 0.7306 (\u001B[32m↗ 0.0102\u001B[0m)\n", + "│ └── Best until now = 0.7306 (\u001B[32m↗ 0.0102\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3129\n", + " │ ├── Epoch N-1 = 0.3278 (\u001B[32m↘ -0.0149\u001B[0m)\n", + " │ └── Best until now = 0.3278 (\u001B[32m↘ -0.0149\u001B[0m)\n", + " ├── Target_iou = 0.7549\n", + " │ ├── Epoch N-1 = 0.7431 (\u001B[32m↗ 0.0117\u001B[0m)\n", + " │ └── Best until now = 0.7446 (\u001B[32m↗ 0.0103\u001B[0m)\n", + " ├── Background_iou = 0.5241\n", + " │ ├── Epoch N-1 = 0.4733 (\u001B[32m↗ 0.0508\u001B[0m)\n", + " │ └── Best until now = 0.4869 (\u001B[32m↗ 0.0372\u001B[0m)\n", + " └── Mean_iou = 0.6395\n", + " ├── Epoch N-1 = 0.6082 (\u001B[32m↗ 0.0313\u001B[0m)\n", + " └── Best until now = 0.6157 (\u001B[32m↗ 0.0238\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "-oGSU3V8lqcm" - }, - "source": [ - "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", - "and extra Auxiliary heads aren't used for training.\n", - "\n", - "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 8: 100%|██████████| 309/309 [02:05<00:00, 2.47it/s, BCEDiceLoss=0.251, background_IOU=0.713, gpu_mem=1.14, mean_IOU=0.749, target_IOU=0.786]\n", + "Validating epoch 8: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", + "[2023-11-13 11:36:30] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:36:30] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7585687637329102\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "X-_dBewgr1dG" - }, - "source": [ - "# 4. Training setup\n", - "\n", - "\n" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 8\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.251\n", + "│ │ ├── Epoch N-1 = 0.259 (\u001B[32m↘ -0.008\u001B[0m)\n", + "│ │ └── Best until now = 0.259 (\u001B[32m↘ -0.008\u001B[0m)\n", + "│ ├── Target_iou = 0.786\n", + "│ │ ├── Epoch N-1 = 0.7808 (\u001B[32m↗ 0.0052\u001B[0m)\n", + "│ │ └── Best until now = 0.7808 (\u001B[32m↗ 0.0052\u001B[0m)\n", + "│ ├── Background_iou = 0.7125\n", + "│ │ ├── Epoch N-1 = 0.7009 (\u001B[32m↗ 0.0116\u001B[0m)\n", + "│ │ └── Best until now = 0.7009 (\u001B[32m↗ 0.0116\u001B[0m)\n", + "│ └── Mean_iou = 0.7493\n", + "│ ├── Epoch N-1 = 0.7409 (\u001B[32m↗ 0.0084\u001B[0m)\n", + "│ └── Best until now = 0.7409 (\u001B[32m↗ 0.0084\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3091\n", + " │ ├── Epoch N-1 = 0.3129 (\u001B[32m↘ -0.0039\u001B[0m)\n", + " │ └── Best until now = 0.3129 (\u001B[32m↘ -0.0039\u001B[0m)\n", + " ├── Target_iou = 0.7586\n", + " │ ├── Epoch N-1 = 0.7549 (\u001B[32m↗ 0.0037\u001B[0m)\n", + " │ └── Best until now = 0.7549 (\u001B[32m↗ 0.0037\u001B[0m)\n", + " ├── Background_iou = 0.5411\n", + " │ ├── Epoch N-1 = 0.5241 (\u001B[32m↗ 0.017\u001B[0m)\n", + " │ └── Best until now = 0.5241 (\u001B[32m↗ 0.017\u001B[0m)\n", + " └── Mean_iou = 0.6498\n", + " ├── Epoch N-1 = 0.6395 (\u001B[32m↗ 0.0103\u001B[0m)\n", + " └── Best until now = 0.6395 (\u001B[32m↗ 0.0103\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "H1Rll8Orl-Dy" - }, - "source": [ - "\n", - "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n", - "\n", - "We will be using an average of BCE and Dice loss for segmentation, with different learning rates for the replaced segmentation head layer, and the rest of the network- this is controlled by the `multiply_head_lr` parameter which is the multiplication factor of the learning rate for the newly replaced layer.\n", - "\n", - "As our `metric_to_watch`, we will be monitoring the `target_IOU` which is one of the components of `BinaryIOU` torchmetrics object (the other components are `mean_IOU` which is the mean of the background and target IOUs, and `background_IOU`)." - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 9: 100%|██████████| 309/309 [02:02<00:00, 2.53it/s, BCEDiceLoss=0.246, background_IOU=0.713, gpu_mem=1.14, mean_IOU=0.752, target_IOU=0.791]\n", + "Validating epoch 9: 100%|██████████| 65/65 [00:17<00:00, 3.72it/s]\n", + "[2023-11-13 11:38:53] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:38:53] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.759834885597229\n" + ] }, { - "cell_type": "code", - "execution_count": 10, - "metadata": { - "id": "NShu3zLgr5qD" - }, - "outputs": [], - "source": [ - "from super_gradients.training.metrics.segmentation_metrics import BinaryIOU\n", - "\n", - "train_params = {\"max_epochs\": 15,\n", - " \"lr_mode\": \"cosine\",\n", - " \"initial_lr\": 0.01,\n", - " \"lr_warmup_epochs\": 5,\n", - " \"multiply_head_lr\": 10,\n", - " \"optimizer\": \"SGD\",\n", - " \"loss\": \"BCEDiceLoss\",\n", - " \"ema\": True,\n", - " \"ema_params\":\n", - " {\n", - " \"decay\": 0.9999,\n", - " \"decay_type\": \"exp\",\n", - " \"beta\": 15,\n", - " },\n", - "\n", - " \"zero_weight_decay_on_bias_and_bn\": True,\n", - " \"average_best_models\": True,\n", - " \"metric_to_watch\": \"target_IOU\",\n", - " \"greater_metric_to_watch_is_better\": True,\n", - " \"train_metrics_list\": [BinaryIOU()],\n", - " \"valid_metrics_list\": [BinaryIOU()],\n", - " \"loss_logging_items_names\": [\"loss\"]\n", - " }" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 9\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2465\n", + "│ │ ├── Epoch N-1 = 0.251 (\u001B[32m↘ -0.0045\u001B[0m)\n", + "│ │ └── Best until now = 0.251 (\u001B[32m↘ -0.0045\u001B[0m)\n", + "│ ├── Target_iou = 0.7905\n", + "│ │ ├── Epoch N-1 = 0.786 (\u001B[32m↗ 0.0045\u001B[0m)\n", + "│ │ └── Best until now = 0.786 (\u001B[32m↗ 0.0045\u001B[0m)\n", + "│ ├── Background_iou = 0.7133\n", + "│ │ ├── Epoch N-1 = 0.7125 (\u001B[32m↗ 0.0008\u001B[0m)\n", + "│ │ └── Best until now = 0.7125 (\u001B[32m↗ 0.0008\u001B[0m)\n", + "│ └── Mean_iou = 0.7519\n", + "│ ├── Epoch N-1 = 0.7493 (\u001B[32m↗ 0.0026\u001B[0m)\n", + "│ └── Best until now = 0.7493 (\u001B[32m↗ 0.0026\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3072\n", + " │ ├── Epoch N-1 = 0.3091 (\u001B[32m↘ -0.0018\u001B[0m)\n", + " │ └── Best until now = 0.3091 (\u001B[32m↘ -0.0018\u001B[0m)\n", + " ├── Target_iou = 0.7598\n", + " │ ├── Epoch N-1 = 0.7586 (\u001B[32m↗ 0.0013\u001B[0m)\n", + " │ └── Best until now = 0.7586 (\u001B[32m↗ 0.0013\u001B[0m)\n", + " ├── Background_iou = 0.5481\n", + " │ ├── Epoch N-1 = 0.5411 (\u001B[32m↗ 0.007\u001B[0m)\n", + " │ └── Best until now = 0.5411 (\u001B[32m↗ 0.007\u001B[0m)\n", + " └── Mean_iou = 0.6539\n", + " ├── Epoch N-1 = 0.6498 (\u001B[32m↗ 0.0041\u001B[0m)\n", + " └── Best until now = 0.6498 (\u001B[32m↗ 0.0041\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "qTECVyhcs506" - }, - "source": [ - "# 5. Training and evaluation\n" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 10: 100%|██████████| 309/309 [02:03<00:00, 2.50it/s, BCEDiceLoss=0.24, background_IOU=0.723, gpu_mem=1.14, mean_IOU=0.759, target_IOU=0.796]\n", + "Validating epoch 10: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", + "[2023-11-13 11:41:16] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:41:16] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7605207562446594\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "S1K5MU2kmmDb" - }, - "source": [ - "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", - "\n", - "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders.\n" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 10\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2399\n", + "│ │ ├── Epoch N-1 = 0.2465 (\u001B[32m↘ -0.0066\u001B[0m)\n", + "│ │ └── Best until now = 0.2465 (\u001B[32m↘ -0.0066\u001B[0m)\n", + "│ ├── Target_iou = 0.7956\n", + "│ │ ├── Epoch N-1 = 0.7905 (\u001B[32m↗ 0.0051\u001B[0m)\n", + "│ │ └── Best until now = 0.7905 (\u001B[32m↗ 0.0051\u001B[0m)\n", + "│ ├── Background_iou = 0.7229\n", + "│ │ ├── Epoch N-1 = 0.7133 (\u001B[32m↗ 0.0096\u001B[0m)\n", + "│ │ └── Best until now = 0.7133 (\u001B[32m↗ 0.0096\u001B[0m)\n", + "│ └── Mean_iou = 0.7593\n", + "│ ├── Epoch N-1 = 0.7519 (\u001B[32m↗ 0.0074\u001B[0m)\n", + "│ └── Best until now = 0.7519 (\u001B[32m↗ 0.0074\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3059\n", + " │ ├── Epoch N-1 = 0.3072 (\u001B[32m↘ -0.0014\u001B[0m)\n", + " │ └── Best until now = 0.3072 (\u001B[32m↘ -0.0014\u001B[0m)\n", + " ├── Target_iou = 0.7605\n", + " │ ├── Epoch N-1 = 0.7598 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " │ └── Best until now = 0.7598 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " ├── Background_iou = 0.5517\n", + " │ ├── Epoch N-1 = 0.5481 (\u001B[32m↗ 0.0037\u001B[0m)\n", + " │ └── Best until now = 0.5481 (\u001B[32m↗ 0.0037\u001B[0m)\n", + " └── Mean_iou = 0.6561\n", + " ├── Epoch N-1 = 0.6539 (\u001B[32m↗ 0.0022\u001B[0m)\n", + " └── Best until now = 0.6539 (\u001B[32m↗ 0.0022\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "code", - "execution_count": 11, - "metadata": { - "id": "u6roEj9ktFTi", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "4a295f63-f0c4-43a7-c6e8-2f7ffd1b5ce2" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-13 11:15:07] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231113_111507_197271`\n", - "[2023-11-13 11:15:07] INFO - sg_trainer.py - Checkpoints directory: ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271\n", - "[2023-11-13 11:15:07] INFO - sg_trainer.py - Using EMA with params {'decay': 0.9999, 'decay_type': 'exp', 'beta': 15}\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "The console stream is now moved to ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/console_Nov13_11_15_07.txt\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-13 11:15:08] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", - " - Mode: Single GPU\n", - " - Number of GPUs: 1 (1 available on the machine)\n", - " - Full dataset size: 2477 (len(train_set))\n", - " - Batch size per GPU: 8 (batch_size)\n", - " - Batch Accumulate: 1 (batch_accumulate)\n", - " - Total batch size: 8 (num_gpus * batch_size)\n", - " - Effective Batch size: 8 (num_gpus * batch_size * batch_accumulate)\n", - " - Iterations per epoch: 309 (len(train_loader))\n", - " - Gradient updates per epoch: 309 (len(train_loader) / batch_accumulate)\n", - "\n", - "[2023-11-13 11:15:08] INFO - sg_trainer.py - Started training for 15 epochs (0/14)\n", - "\n", - "Train epoch 0: 100%|██████████| 309/309 [02:12<00:00, 2.33it/s, BCEDiceLoss=0.4, background_IOU=0.545, gpu_mem=1.14, mean_IOU=0.609, target_IOU=0.674]\n", - "Validating: 100%|██████████| 65/65 [00:17<00:00, 3.69it/s]\n", - "[2023-11-13 11:17:39] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:17:39] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.6779429912567139\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 0\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.4001\n", - "│ ├── Target_iou = 0.6736\n", - "│ ├── Background_iou = 0.5448\n", - "│ └── Mean_iou = 0.6092\n", - "└── Validation\n", - " ├── Bcediceloss = 0.4166\n", - " ├── Target_iou = 0.6779\n", - " ├── Background_iou = 0.4039\n", - " └── Mean_iou = 0.5409\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 1: 100%|██████████| 309/309 [02:05<00:00, 2.46it/s, BCEDiceLoss=0.338, background_IOU=0.604, gpu_mem=1.14, mean_IOU=0.661, target_IOU=0.719]\n", - "Validating epoch 1: 100%|██████████| 65/65 [00:17<00:00, 3.69it/s]\n", - "[2023-11-13 11:20:05] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:20:05] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7205255031585693\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 1\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.3381\n", - "│ │ ├── Epoch N-1 = 0.4001 (\u001b[32m↘ -0.062\u001b[0m)\n", - "│ │ └── Best until now = 0.4001 (\u001b[32m↘ -0.062\u001b[0m)\n", - "│ ├── Target_iou = 0.7193\n", - "│ │ ├── Epoch N-1 = 0.6736 (\u001b[32m↗ 0.0457\u001b[0m)\n", - "│ │ └── Best until now = 0.6736 (\u001b[32m↗ 0.0457\u001b[0m)\n", - "│ ├── Background_iou = 0.6036\n", - "│ │ ├── Epoch N-1 = 0.5448 (\u001b[32m↗ 0.0587\u001b[0m)\n", - "│ │ └── Best until now = 0.5448 (\u001b[32m↗ 0.0587\u001b[0m)\n", - "│ └── Mean_iou = 0.6614\n", - "│ ├── Epoch N-1 = 0.6092 (\u001b[32m↗ 0.0522\u001b[0m)\n", - "│ └── Best until now = 0.6092 (\u001b[32m↗ 0.0522\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3578\n", - " │ ├── Epoch N-1 = 0.4166 (\u001b[32m↘ -0.0588\u001b[0m)\n", - " │ └── Best until now = 0.4166 (\u001b[32m↘ -0.0588\u001b[0m)\n", - " ├── Target_iou = 0.7205\n", - " │ ├── Epoch N-1 = 0.6779 (\u001b[32m↗ 0.0426\u001b[0m)\n", - " │ └── Best until now = 0.6779 (\u001b[32m↗ 0.0426\u001b[0m)\n", - " ├── Background_iou = 0.4497\n", - " │ ├── Epoch N-1 = 0.4039 (\u001b[32m↗ 0.0458\u001b[0m)\n", - " │ └── Best until now = 0.4039 (\u001b[32m↗ 0.0458\u001b[0m)\n", - " └── Mean_iou = 0.5851\n", - " ├── Epoch N-1 = 0.5409 (\u001b[32m↗ 0.0442\u001b[0m)\n", - " └── Best until now = 0.5409 (\u001b[32m↗ 0.0442\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 2: 100%|██████████| 309/309 [02:00<00:00, 2.55it/s, BCEDiceLoss=0.32, background_IOU=0.634, gpu_mem=1.14, mean_IOU=0.684, target_IOU=0.734]\n", - "Validating epoch 2: 100%|██████████| 65/65 [00:16<00:00, 3.84it/s]\n", - "[2023-11-13 11:22:24] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:22:24] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7300039529800415\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 2\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.3199\n", - "│ │ ├── Epoch N-1 = 0.3381 (\u001b[32m↘ -0.0182\u001b[0m)\n", - "│ │ └── Best until now = 0.3381 (\u001b[32m↘ -0.0182\u001b[0m)\n", - "│ ├── Target_iou = 0.734\n", - "│ │ ├── Epoch N-1 = 0.7193 (\u001b[32m↗ 0.0147\u001b[0m)\n", - "│ │ └── Best until now = 0.7193 (\u001b[32m↗ 0.0147\u001b[0m)\n", - "│ ├── Background_iou = 0.6344\n", - "│ │ ├── Epoch N-1 = 0.6036 (\u001b[32m↗ 0.0308\u001b[0m)\n", - "│ │ └── Best until now = 0.6036 (\u001b[32m↗ 0.0308\u001b[0m)\n", - "│ └── Mean_iou = 0.6842\n", - "│ ├── Epoch N-1 = 0.6614 (\u001b[32m↗ 0.0227\u001b[0m)\n", - "│ └── Best until now = 0.6614 (\u001b[32m↗ 0.0227\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.357\n", - " │ ├── Epoch N-1 = 0.3578 (\u001b[32m↘ -0.0008\u001b[0m)\n", - " │ └── Best until now = 0.3578 (\u001b[32m↘ -0.0008\u001b[0m)\n", - " ├── Target_iou = 0.73\n", - " │ ├── Epoch N-1 = 0.7205 (\u001b[32m↗ 0.0095\u001b[0m)\n", - " │ └── Best until now = 0.7205 (\u001b[32m↗ 0.0095\u001b[0m)\n", - " ├── Background_iou = 0.4503\n", - " │ ├── Epoch N-1 = 0.4497 (\u001b[32m↗ 0.0006\u001b[0m)\n", - " │ └── Best until now = 0.4497 (\u001b[32m↗ 0.0006\u001b[0m)\n", - " └── Mean_iou = 0.5902\n", - " ├── Epoch N-1 = 0.5851 (\u001b[32m↗ 0.0051\u001b[0m)\n", - " └── Best until now = 0.5851 (\u001b[32m↗ 0.0051\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 3: 100%|██████████| 309/309 [01:59<00:00, 2.58it/s, BCEDiceLoss=0.302, background_IOU=0.645, gpu_mem=1.14, mean_IOU=0.697, target_IOU=0.75]\n", - "Validating epoch 3: 100%|██████████| 65/65 [00:16<00:00, 3.84it/s]\n", - "[2023-11-13 11:24:43] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:24:43] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7432040572166443\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 3\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.3022\n", - "│ │ ├── Epoch N-1 = 0.3199 (\u001b[32m↘ -0.0177\u001b[0m)\n", - "│ │ └── Best until now = 0.3199 (\u001b[32m↘ -0.0177\u001b[0m)\n", - "│ ├── Target_iou = 0.7501\n", - "│ │ ├── Epoch N-1 = 0.734 (\u001b[32m↗ 0.0161\u001b[0m)\n", - "│ │ └── Best until now = 0.734 (\u001b[32m↗ 0.0161\u001b[0m)\n", - "│ ├── Background_iou = 0.6447\n", - "│ │ ├── Epoch N-1 = 0.6344 (\u001b[32m↗ 0.0103\u001b[0m)\n", - "│ │ └── Best until now = 0.6344 (\u001b[32m↗ 0.0103\u001b[0m)\n", - "│ └── Mean_iou = 0.6974\n", - "│ ├── Epoch N-1 = 0.6842 (\u001b[32m↗ 0.0132\u001b[0m)\n", - "│ └── Best until now = 0.6842 (\u001b[32m↗ 0.0132\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3307\n", - " │ ├── Epoch N-1 = 0.357 (\u001b[32m↘ -0.0263\u001b[0m)\n", - " │ └── Best until now = 0.357 (\u001b[32m↘ -0.0263\u001b[0m)\n", - " ├── Target_iou = 0.7432\n", - " │ ├── Epoch N-1 = 0.73 (\u001b[32m↗ 0.0132\u001b[0m)\n", - " │ └── Best until now = 0.73 (\u001b[32m↗ 0.0132\u001b[0m)\n", - " ├── Background_iou = 0.4794\n", - " │ ├── Epoch N-1 = 0.4503 (\u001b[32m↗ 0.0291\u001b[0m)\n", - " │ └── Best until now = 0.4503 (\u001b[32m↗ 0.0291\u001b[0m)\n", - " └── Mean_iou = 0.6113\n", - " ├── Epoch N-1 = 0.5902 (\u001b[32m↗ 0.0212\u001b[0m)\n", - " └── Best until now = 0.5902 (\u001b[32m↗ 0.0212\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 4: 100%|██████████| 309/309 [02:00<00:00, 2.56it/s, BCEDiceLoss=0.287, background_IOU=0.67, gpu_mem=1.14, mean_IOU=0.715, target_IOU=0.76]\n", - "Validating epoch 4: 100%|██████████| 65/65 [00:17<00:00, 3.79it/s]\n", - "[2023-11-13 11:27:02] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:27:02] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7445915341377258\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 4\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2867\n", - "│ │ ├── Epoch N-1 = 0.3022 (\u001b[32m↘ -0.0155\u001b[0m)\n", - "│ │ └── Best until now = 0.3022 (\u001b[32m↘ -0.0155\u001b[0m)\n", - "│ ├── Target_iou = 0.7604\n", - "│ │ ├── Epoch N-1 = 0.7501 (\u001b[32m↗ 0.0103\u001b[0m)\n", - "│ │ └── Best until now = 0.7501 (\u001b[32m↗ 0.0103\u001b[0m)\n", - "│ ├── Background_iou = 0.6697\n", - "│ │ ├── Epoch N-1 = 0.6447 (\u001b[32m↗ 0.0251\u001b[0m)\n", - "│ │ └── Best until now = 0.6447 (\u001b[32m↗ 0.0251\u001b[0m)\n", - "│ └── Mean_iou = 0.715\n", - "│ ├── Epoch N-1 = 0.6974 (\u001b[32m↗ 0.0177\u001b[0m)\n", - "│ └── Best until now = 0.6974 (\u001b[32m↗ 0.0177\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3281\n", - " │ ├── Epoch N-1 = 0.3307 (\u001b[32m↘ -0.0026\u001b[0m)\n", - " │ └── Best until now = 0.3307 (\u001b[32m↘ -0.0026\u001b[0m)\n", - " ├── Target_iou = 0.7446\n", - " │ ├── Epoch N-1 = 0.7432 (\u001b[32m↗ 0.0014\u001b[0m)\n", - " │ └── Best until now = 0.7432 (\u001b[32m↗ 0.0014\u001b[0m)\n", - " ├── Background_iou = 0.4869\n", - " │ ├── Epoch N-1 = 0.4794 (\u001b[32m↗ 0.0074\u001b[0m)\n", - " │ └── Best until now = 0.4794 (\u001b[32m↗ 0.0074\u001b[0m)\n", - " └── Mean_iou = 0.6157\n", - " ├── Epoch N-1 = 0.6113 (\u001b[32m↗ 0.0044\u001b[0m)\n", - " └── Best until now = 0.6113 (\u001b[32m↗ 0.0044\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 5: 100%|██████████| 309/309 [02:02<00:00, 2.53it/s, BCEDiceLoss=0.287, background_IOU=0.664, gpu_mem=1.14, mean_IOU=0.712, target_IOU=0.761]\n", - "Validating epoch 5: 100%|██████████| 65/65 [00:17<00:00, 3.75it/s]\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 5\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2869\n", - "│ │ ├── Epoch N-1 = 0.2867 (\u001b[31m↗ 1e-04\u001b[0m)\n", - "│ │ └── Best until now = 0.2867 (\u001b[31m↗ 1e-04\u001b[0m)\n", - "│ ├── Target_iou = 0.7606\n", - "│ │ ├── Epoch N-1 = 0.7604 (\u001b[32m↗ 0.0002\u001b[0m)\n", - "│ │ └── Best until now = 0.7604 (\u001b[32m↗ 0.0002\u001b[0m)\n", - "│ ├── Background_iou = 0.6637\n", - "│ │ ├── Epoch N-1 = 0.6697 (\u001b[31m↘ -0.0061\u001b[0m)\n", - "│ │ └── Best until now = 0.6697 (\u001b[31m↘ -0.0061\u001b[0m)\n", - "│ └── Mean_iou = 0.7121\n", - "│ ├── Epoch N-1 = 0.715 (\u001b[31m↘ -0.0029\u001b[0m)\n", - "│ └── Best until now = 0.715 (\u001b[31m↘ -0.0029\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3339\n", - " │ ├── Epoch N-1 = 0.3281 (\u001b[31m↗ 0.0059\u001b[0m)\n", - " │ └── Best until now = 0.3281 (\u001b[31m↗ 0.0059\u001b[0m)\n", - " ├── Target_iou = 0.7402\n", - " │ ├── Epoch N-1 = 0.7446 (\u001b[31m↘ -0.0044\u001b[0m)\n", - " │ └── Best until now = 0.7446 (\u001b[31m↘ -0.0044\u001b[0m)\n", - " ├── Background_iou = 0.4593\n", - " │ ├── Epoch N-1 = 0.4869 (\u001b[31m↘ -0.0276\u001b[0m)\n", - " │ └── Best until now = 0.4869 (\u001b[31m↘ -0.0276\u001b[0m)\n", - " └── Mean_iou = 0.5997\n", - " ├── Epoch N-1 = 0.6157 (\u001b[31m↘ -0.016\u001b[0m)\n", - " └── Best until now = 0.6157 (\u001b[31m↘ -0.016\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 6: 100%|██████████| 309/309 [02:03<00:00, 2.50it/s, BCEDiceLoss=0.269, background_IOU=0.689, gpu_mem=1.14, mean_IOU=0.731, target_IOU=0.772]\n", - "Validating epoch 6: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 6\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2686\n", - "│ │ ├── Epoch N-1 = 0.2869 (\u001b[32m↘ -0.0183\u001b[0m)\n", - "│ │ └── Best until now = 0.2867 (\u001b[32m↘ -0.0181\u001b[0m)\n", - "│ ├── Target_iou = 0.7721\n", - "│ │ ├── Epoch N-1 = 0.7606 (\u001b[32m↗ 0.0115\u001b[0m)\n", - "│ │ └── Best until now = 0.7606 (\u001b[32m↗ 0.0115\u001b[0m)\n", - "│ ├── Background_iou = 0.6892\n", - "│ │ ├── Epoch N-1 = 0.6637 (\u001b[32m↗ 0.0255\u001b[0m)\n", - "│ │ └── Best until now = 0.6697 (\u001b[32m↗ 0.0194\u001b[0m)\n", - "│ └── Mean_iou = 0.7306\n", - "│ ├── Epoch N-1 = 0.7121 (\u001b[32m↗ 0.0185\u001b[0m)\n", - "│ └── Best until now = 0.715 (\u001b[32m↗ 0.0156\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3278\n", - " │ ├── Epoch N-1 = 0.3339 (\u001b[32m↘ -0.0061\u001b[0m)\n", - " │ └── Best until now = 0.3281 (\u001b[32m↘ -0.0003\u001b[0m)\n", - " ├── Target_iou = 0.7431\n", - " │ ├── Epoch N-1 = 0.7402 (\u001b[32m↗ 0.003\u001b[0m)\n", - " │ └── Best until now = 0.7446 (\u001b[31m↘ -0.0015\u001b[0m)\n", - " ├── Background_iou = 0.4733\n", - " │ ├── Epoch N-1 = 0.4593 (\u001b[32m↗ 0.0139\u001b[0m)\n", - " │ └── Best until now = 0.4869 (\u001b[31m↘ -0.0136\u001b[0m)\n", - " └── Mean_iou = 0.6082\n", - " ├── Epoch N-1 = 0.5997 (\u001b[32m↗ 0.0085\u001b[0m)\n", - " └── Best until now = 0.6157 (\u001b[31m↘ -0.0075\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 7: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.259, background_IOU=0.701, gpu_mem=1.14, mean_IOU=0.741, target_IOU=0.781]\n", - "Validating epoch 7: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", - "[2023-11-13 11:34:05] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:34:05] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7548585534095764\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 7\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.259\n", - "│ │ ├── Epoch N-1 = 0.2686 (\u001b[32m↘ -0.0096\u001b[0m)\n", - "│ │ └── Best until now = 0.2686 (\u001b[32m↘ -0.0096\u001b[0m)\n", - "│ ├── Target_iou = 0.7808\n", - "│ │ ├── Epoch N-1 = 0.7721 (\u001b[32m↗ 0.0087\u001b[0m)\n", - "│ │ └── Best until now = 0.7721 (\u001b[32m↗ 0.0087\u001b[0m)\n", - "│ ├── Background_iou = 0.7009\n", - "│ │ ├── Epoch N-1 = 0.6892 (\u001b[32m↗ 0.0117\u001b[0m)\n", - "│ │ └── Best until now = 0.6892 (\u001b[32m↗ 0.0117\u001b[0m)\n", - "│ └── Mean_iou = 0.7409\n", - "│ ├── Epoch N-1 = 0.7306 (\u001b[32m↗ 0.0102\u001b[0m)\n", - "│ └── Best until now = 0.7306 (\u001b[32m↗ 0.0102\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3129\n", - " │ ├── Epoch N-1 = 0.3278 (\u001b[32m↘ -0.0149\u001b[0m)\n", - " │ └── Best until now = 0.3278 (\u001b[32m↘ -0.0149\u001b[0m)\n", - " ├── Target_iou = 0.7549\n", - " │ ├── Epoch N-1 = 0.7431 (\u001b[32m↗ 0.0117\u001b[0m)\n", - " │ └── Best until now = 0.7446 (\u001b[32m↗ 0.0103\u001b[0m)\n", - " ├── Background_iou = 0.5241\n", - " │ ├── Epoch N-1 = 0.4733 (\u001b[32m↗ 0.0508\u001b[0m)\n", - " │ └── Best until now = 0.4869 (\u001b[32m↗ 0.0372\u001b[0m)\n", - " └── Mean_iou = 0.6395\n", - " ├── Epoch N-1 = 0.6082 (\u001b[32m↗ 0.0313\u001b[0m)\n", - " └── Best until now = 0.6157 (\u001b[32m↗ 0.0238\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 8: 100%|██████████| 309/309 [02:05<00:00, 2.47it/s, BCEDiceLoss=0.251, background_IOU=0.713, gpu_mem=1.14, mean_IOU=0.749, target_IOU=0.786]\n", - "Validating epoch 8: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", - "[2023-11-13 11:36:30] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:36:30] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7585687637329102\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 8\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.251\n", - "│ │ ├── Epoch N-1 = 0.259 (\u001b[32m↘ -0.008\u001b[0m)\n", - "│ │ └── Best until now = 0.259 (\u001b[32m↘ -0.008\u001b[0m)\n", - "│ ├── Target_iou = 0.786\n", - "│ │ ├── Epoch N-1 = 0.7808 (\u001b[32m↗ 0.0052\u001b[0m)\n", - "│ │ └── Best until now = 0.7808 (\u001b[32m↗ 0.0052\u001b[0m)\n", - "│ ├── Background_iou = 0.7125\n", - "│ │ ├── Epoch N-1 = 0.7009 (\u001b[32m↗ 0.0116\u001b[0m)\n", - "│ │ └── Best until now = 0.7009 (\u001b[32m↗ 0.0116\u001b[0m)\n", - "│ └── Mean_iou = 0.7493\n", - "│ ├── Epoch N-1 = 0.7409 (\u001b[32m↗ 0.0084\u001b[0m)\n", - "│ └── Best until now = 0.7409 (\u001b[32m↗ 0.0084\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3091\n", - " │ ├── Epoch N-1 = 0.3129 (\u001b[32m↘ -0.0039\u001b[0m)\n", - " │ └── Best until now = 0.3129 (\u001b[32m↘ -0.0039\u001b[0m)\n", - " ├── Target_iou = 0.7586\n", - " │ ├── Epoch N-1 = 0.7549 (\u001b[32m↗ 0.0037\u001b[0m)\n", - " │ └── Best until now = 0.7549 (\u001b[32m↗ 0.0037\u001b[0m)\n", - " ├── Background_iou = 0.5411\n", - " │ ├── Epoch N-1 = 0.5241 (\u001b[32m↗ 0.017\u001b[0m)\n", - " │ └── Best until now = 0.5241 (\u001b[32m↗ 0.017\u001b[0m)\n", - " └── Mean_iou = 0.6498\n", - " ├── Epoch N-1 = 0.6395 (\u001b[32m↗ 0.0103\u001b[0m)\n", - " └── Best until now = 0.6395 (\u001b[32m↗ 0.0103\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 9: 100%|██████████| 309/309 [02:02<00:00, 2.53it/s, BCEDiceLoss=0.246, background_IOU=0.713, gpu_mem=1.14, mean_IOU=0.752, target_IOU=0.791]\n", - "Validating epoch 9: 100%|██████████| 65/65 [00:17<00:00, 3.72it/s]\n", - "[2023-11-13 11:38:53] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:38:53] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.759834885597229\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 9\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2465\n", - "│ │ ├── Epoch N-1 = 0.251 (\u001b[32m↘ -0.0045\u001b[0m)\n", - "│ │ └── Best until now = 0.251 (\u001b[32m↘ -0.0045\u001b[0m)\n", - "│ ├── Target_iou = 0.7905\n", - "│ │ ├── Epoch N-1 = 0.786 (\u001b[32m↗ 0.0045\u001b[0m)\n", - "│ │ └── Best until now = 0.786 (\u001b[32m↗ 0.0045\u001b[0m)\n", - "│ ├── Background_iou = 0.7133\n", - "│ │ ├── Epoch N-1 = 0.7125 (\u001b[32m↗ 0.0008\u001b[0m)\n", - "│ │ └── Best until now = 0.7125 (\u001b[32m↗ 0.0008\u001b[0m)\n", - "│ └── Mean_iou = 0.7519\n", - "│ ├── Epoch N-1 = 0.7493 (\u001b[32m↗ 0.0026\u001b[0m)\n", - "│ └── Best until now = 0.7493 (\u001b[32m↗ 0.0026\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3072\n", - " │ ├── Epoch N-1 = 0.3091 (\u001b[32m↘ -0.0018\u001b[0m)\n", - " │ └── Best until now = 0.3091 (\u001b[32m↘ -0.0018\u001b[0m)\n", - " ├── Target_iou = 0.7598\n", - " │ ├── Epoch N-1 = 0.7586 (\u001b[32m↗ 0.0013\u001b[0m)\n", - " │ └── Best until now = 0.7586 (\u001b[32m↗ 0.0013\u001b[0m)\n", - " ├── Background_iou = 0.5481\n", - " │ ├── Epoch N-1 = 0.5411 (\u001b[32m↗ 0.007\u001b[0m)\n", - " │ └── Best until now = 0.5411 (\u001b[32m↗ 0.007\u001b[0m)\n", - " └── Mean_iou = 0.6539\n", - " ├── Epoch N-1 = 0.6498 (\u001b[32m↗ 0.0041\u001b[0m)\n", - " └── Best until now = 0.6498 (\u001b[32m↗ 0.0041\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 10: 100%|██████████| 309/309 [02:03<00:00, 2.50it/s, BCEDiceLoss=0.24, background_IOU=0.723, gpu_mem=1.14, mean_IOU=0.759, target_IOU=0.796]\n", - "Validating epoch 10: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", - "[2023-11-13 11:41:16] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:41:16] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7605207562446594\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 10\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2399\n", - "│ │ ├── Epoch N-1 = 0.2465 (\u001b[32m↘ -0.0066\u001b[0m)\n", - "│ │ └── Best until now = 0.2465 (\u001b[32m↘ -0.0066\u001b[0m)\n", - "│ ├── Target_iou = 0.7956\n", - "│ │ ├── Epoch N-1 = 0.7905 (\u001b[32m↗ 0.0051\u001b[0m)\n", - "│ │ └── Best until now = 0.7905 (\u001b[32m↗ 0.0051\u001b[0m)\n", - "│ ├── Background_iou = 0.7229\n", - "│ │ ├── Epoch N-1 = 0.7133 (\u001b[32m↗ 0.0096\u001b[0m)\n", - "│ │ └── Best until now = 0.7133 (\u001b[32m↗ 0.0096\u001b[0m)\n", - "│ └── Mean_iou = 0.7593\n", - "│ ├── Epoch N-1 = 0.7519 (\u001b[32m↗ 0.0074\u001b[0m)\n", - "│ └── Best until now = 0.7519 (\u001b[32m↗ 0.0074\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3059\n", - " │ ├── Epoch N-1 = 0.3072 (\u001b[32m↘ -0.0014\u001b[0m)\n", - " │ └── Best until now = 0.3072 (\u001b[32m↘ -0.0014\u001b[0m)\n", - " ├── Target_iou = 0.7605\n", - " │ ├── Epoch N-1 = 0.7598 (\u001b[32m↗ 0.0007\u001b[0m)\n", - " │ └── Best until now = 0.7598 (\u001b[32m↗ 0.0007\u001b[0m)\n", - " ├── Background_iou = 0.5517\n", - " │ ├── Epoch N-1 = 0.5481 (\u001b[32m↗ 0.0037\u001b[0m)\n", - " │ └── Best until now = 0.5481 (\u001b[32m↗ 0.0037\u001b[0m)\n", - " └── Mean_iou = 0.6561\n", - " ├── Epoch N-1 = 0.6539 (\u001b[32m↗ 0.0022\u001b[0m)\n", - " └── Best until now = 0.6539 (\u001b[32m↗ 0.0022\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 11: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.231, background_IOU=0.733, gpu_mem=1.14, mean_IOU=0.767, target_IOU=0.801]\n", - "Validating epoch 11: 100%|██████████| 65/65 [00:17<00:00, 3.76it/s]\n", - "[2023-11-13 11:43:37] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:43:37] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7611058950424194\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 11\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2309\n", - "│ │ ├── Epoch N-1 = 0.2399 (\u001b[32m↘ -0.009\u001b[0m)\n", - "│ │ └── Best until now = 0.2399 (\u001b[32m↘ -0.009\u001b[0m)\n", - "│ ├── Target_iou = 0.8015\n", - "│ │ ├── Epoch N-1 = 0.7956 (\u001b[32m↗ 0.0059\u001b[0m)\n", - "│ │ └── Best until now = 0.7956 (\u001b[32m↗ 0.0059\u001b[0m)\n", - "│ ├── Background_iou = 0.7333\n", - "│ │ ├── Epoch N-1 = 0.7229 (\u001b[32m↗ 0.0104\u001b[0m)\n", - "│ │ └── Best until now = 0.7229 (\u001b[32m↗ 0.0104\u001b[0m)\n", - "│ └── Mean_iou = 0.7674\n", - "│ ├── Epoch N-1 = 0.7593 (\u001b[32m↗ 0.0081\u001b[0m)\n", - "│ └── Best until now = 0.7593 (\u001b[32m↗ 0.0081\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3046\n", - " │ ├── Epoch N-1 = 0.3059 (\u001b[32m↘ -0.0012\u001b[0m)\n", - " │ └── Best until now = 0.3059 (\u001b[32m↘ -0.0012\u001b[0m)\n", - " ├── Target_iou = 0.7611\n", - " │ ├── Epoch N-1 = 0.7605 (\u001b[32m↗ 0.0006\u001b[0m)\n", - " │ └── Best until now = 0.7605 (\u001b[32m↗ 0.0006\u001b[0m)\n", - " ├── Background_iou = 0.5546\n", - " │ ├── Epoch N-1 = 0.5517 (\u001b[32m↗ 0.0029\u001b[0m)\n", - " │ └── Best until now = 0.5517 (\u001b[32m↗ 0.0029\u001b[0m)\n", - " └── Mean_iou = 0.6579\n", - " ├── Epoch N-1 = 0.6561 (\u001b[32m↗ 0.0017\u001b[0m)\n", - " └── Best until now = 0.6561 (\u001b[32m↗ 0.0017\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 12: 100%|██████████| 309/309 [02:03<00:00, 2.51it/s, BCEDiceLoss=0.224, background_IOU=0.736, gpu_mem=1.14, mean_IOU=0.771, target_IOU=0.807]\n", - "Validating epoch 12: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", - "[2023-11-13 11:46:00] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:46:00] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7616798877716064\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 12\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2243\n", - "│ │ ├── Epoch N-1 = 0.2309 (\u001b[32m↘ -0.0066\u001b[0m)\n", - "│ │ └── Best until now = 0.2309 (\u001b[32m↘ -0.0066\u001b[0m)\n", - "│ ├── Target_iou = 0.8068\n", - "│ │ ├── Epoch N-1 = 0.8015 (\u001b[32m↗ 0.0053\u001b[0m)\n", - "│ │ └── Best until now = 0.8015 (\u001b[32m↗ 0.0053\u001b[0m)\n", - "│ ├── Background_iou = 0.736\n", - "│ │ ├── Epoch N-1 = 0.7333 (\u001b[32m↗ 0.0027\u001b[0m)\n", - "│ │ └── Best until now = 0.7333 (\u001b[32m↗ 0.0027\u001b[0m)\n", - "│ └── Mean_iou = 0.7714\n", - "│ ├── Epoch N-1 = 0.7674 (\u001b[32m↗ 0.004\u001b[0m)\n", - "│ └── Best until now = 0.7674 (\u001b[32m↗ 0.004\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3035\n", - " │ ├── Epoch N-1 = 0.3046 (\u001b[32m↘ -0.0012\u001b[0m)\n", - " │ └── Best until now = 0.3046 (\u001b[32m↘ -0.0012\u001b[0m)\n", - " ├── Target_iou = 0.7617\n", - " │ ├── Epoch N-1 = 0.7611 (\u001b[32m↗ 0.0006\u001b[0m)\n", - " │ └── Best until now = 0.7611 (\u001b[32m↗ 0.0006\u001b[0m)\n", - " ├── Background_iou = 0.5569\n", - " │ ├── Epoch N-1 = 0.5546 (\u001b[32m↗ 0.0023\u001b[0m)\n", - " │ └── Best until now = 0.5546 (\u001b[32m↗ 0.0023\u001b[0m)\n", - " └── Mean_iou = 0.6593\n", - " ├── Epoch N-1 = 0.6579 (\u001b[32m↗ 0.0014\u001b[0m)\n", - " └── Best until now = 0.6579 (\u001b[32m↗ 0.0014\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 13: 100%|██████████| 309/309 [02:01<00:00, 2.55it/s, BCEDiceLoss=0.219, background_IOU=0.745, gpu_mem=1.14, mean_IOU=0.777, target_IOU=0.81]\n", - "Validating epoch 13: 100%|██████████| 65/65 [00:17<00:00, 3.81it/s]\n", - "[2023-11-13 11:48:23] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:48:23] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7624021172523499\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 13\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2194\n", - "│ │ ├── Epoch N-1 = 0.2243 (\u001b[32m↘ -0.0049\u001b[0m)\n", - "│ │ └── Best until now = 0.2243 (\u001b[32m↘ -0.0049\u001b[0m)\n", - "│ ├── Target_iou = 0.8097\n", - "│ │ ├── Epoch N-1 = 0.8068 (\u001b[32m↗ 0.0029\u001b[0m)\n", - "│ │ └── Best until now = 0.8068 (\u001b[32m↗ 0.0029\u001b[0m)\n", - "│ ├── Background_iou = 0.7447\n", - "│ │ ├── Epoch N-1 = 0.736 (\u001b[32m↗ 0.0086\u001b[0m)\n", - "│ │ └── Best until now = 0.736 (\u001b[32m↗ 0.0086\u001b[0m)\n", - "│ └── Mean_iou = 0.7772\n", - "│ ├── Epoch N-1 = 0.7714 (\u001b[32m↗ 0.0058\u001b[0m)\n", - "│ └── Best until now = 0.7714 (\u001b[32m↗ 0.0058\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3024\n", - " │ ├── Epoch N-1 = 0.3035 (\u001b[32m↘ -0.0011\u001b[0m)\n", - " │ └── Best until now = 0.3035 (\u001b[32m↘ -0.0011\u001b[0m)\n", - " ├── Target_iou = 0.7624\n", - " │ ├── Epoch N-1 = 0.7617 (\u001b[32m↗ 0.0007\u001b[0m)\n", - " │ └── Best until now = 0.7617 (\u001b[32m↗ 0.0007\u001b[0m)\n", - " ├── Background_iou = 0.5596\n", - " │ ├── Epoch N-1 = 0.5569 (\u001b[32m↗ 0.0027\u001b[0m)\n", - " │ └── Best until now = 0.5569 (\u001b[32m↗ 0.0027\u001b[0m)\n", - " └── Mean_iou = 0.661\n", - " ├── Epoch N-1 = 0.6593 (\u001b[32m↗ 0.0017\u001b[0m)\n", - " └── Best until now = 0.6593 (\u001b[32m↗ 0.0017\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 14: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.215, background_IOU=0.748, gpu_mem=1.14, mean_IOU=0.781, target_IOU=0.813]\n", - "Validating epoch 14: 100%|██████████| 65/65 [00:17<00:00, 3.78it/s]\n", - "[2023-11-13 11:50:45] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", - "[2023-11-13 11:50:45] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.763008713722229\n" - ] - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 14\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2155\n", - "│ │ ├── Epoch N-1 = 0.2194 (\u001b[32m↘ -0.0039\u001b[0m)\n", - "│ │ └── Best until now = 0.2194 (\u001b[32m↘ -0.0039\u001b[0m)\n", - "│ ├── Target_iou = 0.8134\n", - "│ │ ├── Epoch N-1 = 0.8097 (\u001b[32m↗ 0.0037\u001b[0m)\n", - "│ │ └── Best until now = 0.8097 (\u001b[32m↗ 0.0037\u001b[0m)\n", - "│ ├── Background_iou = 0.7484\n", - "│ │ ├── Epoch N-1 = 0.7447 (\u001b[32m↗ 0.0038\u001b[0m)\n", - "│ │ └── Best until now = 0.7447 (\u001b[32m↗ 0.0038\u001b[0m)\n", - "│ └── Mean_iou = 0.7809\n", - "│ ├── Epoch N-1 = 0.7772 (\u001b[32m↗ 0.0037\u001b[0m)\n", - "│ └── Best until now = 0.7772 (\u001b[32m↗ 0.0037\u001b[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.3015\n", - " │ ├── Epoch N-1 = 0.3024 (\u001b[32m↘ -0.0009\u001b[0m)\n", - " │ └── Best until now = 0.3024 (\u001b[32m↘ -0.0009\u001b[0m)\n", - " ├── Target_iou = 0.763\n", - " │ ├── Epoch N-1 = 0.7624 (\u001b[32m↗ 0.0006\u001b[0m)\n", - " │ └── Best until now = 0.7624 (\u001b[32m↗ 0.0006\u001b[0m)\n", - " ├── Background_iou = 0.5621\n", - " │ ├── Epoch N-1 = 0.5596 (\u001b[32m↗ 0.0025\u001b[0m)\n", - " │ └── Best until now = 0.5596 (\u001b[32m↗ 0.0025\u001b[0m)\n", - " └── Mean_iou = 0.6625\n", - " ├── Epoch N-1 = 0.661 (\u001b[32m↗ 0.0016\u001b[0m)\n", - " └── Best until now = 0.661 (\u001b[32m↗ 0.0016\u001b[0m)\n", - "\n", - "===========================================================\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-13 11:50:47] INFO - sg_trainer.py - RUNNING ADDITIONAL TEST ON THE AVERAGED MODEL...\n", - "Validating epoch 15: 98%|█████████▊| 64/65 [00:16<00:00, 3.27it/s]" - ] - } - ], - "source": [ - "trainer.train(model=model, training_params=train_params, train_loader=train_loader, valid_loader=valid_loader)" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 11: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.231, background_IOU=0.733, gpu_mem=1.14, mean_IOU=0.767, target_IOU=0.801]\n", + "Validating epoch 11: 100%|██████████| 65/65 [00:17<00:00, 3.76it/s]\n", + "[2023-11-13 11:43:37] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:43:37] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7611058950424194\n" + ] }, { - "cell_type": "code", - "execution_count": 12, - "metadata": { - "id": "X8BJq1crcbjl", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "661796b8-431a-4c23-ac57-9bdc579a685d" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Best Checkpoint mIoU is: 0.763008713722229\n" - ] - } - ], - "source": [ - "print(\"Best Checkpoint mIoU is: \"+ str(trainer.best_metric))" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 11\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2309\n", + "│ │ ├── Epoch N-1 = 0.2399 (\u001B[32m↘ -0.009\u001B[0m)\n", + "│ │ └── Best until now = 0.2399 (\u001B[32m↘ -0.009\u001B[0m)\n", + "│ ├── Target_iou = 0.8015\n", + "│ │ ├── Epoch N-1 = 0.7956 (\u001B[32m↗ 0.0059\u001B[0m)\n", + "│ │ └── Best until now = 0.7956 (\u001B[32m↗ 0.0059\u001B[0m)\n", + "│ ├── Background_iou = 0.7333\n", + "│ │ ├── Epoch N-1 = 0.7229 (\u001B[32m↗ 0.0104\u001B[0m)\n", + "│ │ └── Best until now = 0.7229 (\u001B[32m↗ 0.0104\u001B[0m)\n", + "│ └── Mean_iou = 0.7674\n", + "│ ├── Epoch N-1 = 0.7593 (\u001B[32m↗ 0.0081\u001B[0m)\n", + "│ └── Best until now = 0.7593 (\u001B[32m↗ 0.0081\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3046\n", + " │ ├── Epoch N-1 = 0.3059 (\u001B[32m↘ -0.0012\u001B[0m)\n", + " │ └── Best until now = 0.3059 (\u001B[32m↘ -0.0012\u001B[0m)\n", + " ├── Target_iou = 0.7611\n", + " │ ├── Epoch N-1 = 0.7605 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.7605 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " ├── Background_iou = 0.5546\n", + " │ ├── Epoch N-1 = 0.5517 (\u001B[32m↗ 0.0029\u001B[0m)\n", + " │ └── Best until now = 0.5517 (\u001B[32m↗ 0.0029\u001B[0m)\n", + " └── Mean_iou = 0.6579\n", + " ├── Epoch N-1 = 0.6561 (\u001B[32m↗ 0.0017\u001B[0m)\n", + " └── Best until now = 0.6561 (\u001B[32m↗ 0.0017\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "3Nybj15cchxd" - }, - "source": [ - "Now you can download your trained weights from this directory" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 12: 100%|██████████| 309/309 [02:03<00:00, 2.51it/s, BCEDiceLoss=0.224, background_IOU=0.736, gpu_mem=1.14, mean_IOU=0.771, target_IOU=0.807]\n", + "Validating epoch 12: 100%|██████████| 65/65 [00:17<00:00, 3.77it/s]\n", + "[2023-11-13 11:46:00] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:46:00] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7616798877716064\n" + ] }, { - "cell_type": "code", - "execution_count": 13, - "metadata": { - "id": "_iHsFgPSciQh" - }, - "outputs": [], - "source": [ - "print(trainer.checkpoints_dir_path)" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 12\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2243\n", + "│ │ ├── Epoch N-1 = 0.2309 (\u001B[32m↘ -0.0066\u001B[0m)\n", + "│ │ └── Best until now = 0.2309 (\u001B[32m↘ -0.0066\u001B[0m)\n", + "│ ├── Target_iou = 0.8068\n", + "│ │ ├── Epoch N-1 = 0.8015 (\u001B[32m↗ 0.0053\u001B[0m)\n", + "│ │ └── Best until now = 0.8015 (\u001B[32m↗ 0.0053\u001B[0m)\n", + "│ ├── Background_iou = 0.736\n", + "│ │ ├── Epoch N-1 = 0.7333 (\u001B[32m↗ 0.0027\u001B[0m)\n", + "│ │ └── Best until now = 0.7333 (\u001B[32m↗ 0.0027\u001B[0m)\n", + "│ └── Mean_iou = 0.7714\n", + "│ ├── Epoch N-1 = 0.7674 (\u001B[32m↗ 0.004\u001B[0m)\n", + "│ └── Best until now = 0.7674 (\u001B[32m↗ 0.004\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3035\n", + " │ ├── Epoch N-1 = 0.3046 (\u001B[32m↘ -0.0012\u001B[0m)\n", + " │ └── Best until now = 0.3046 (\u001B[32m↘ -0.0012\u001B[0m)\n", + " ├── Target_iou = 0.7617\n", + " │ ├── Epoch N-1 = 0.7611 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.7611 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " ├── Background_iou = 0.5569\n", + " │ ├── Epoch N-1 = 0.5546 (\u001B[32m↗ 0.0023\u001B[0m)\n", + " │ └── Best until now = 0.5546 (\u001B[32m↗ 0.0023\u001B[0m)\n", + " └── Mean_iou = 0.6593\n", + " ├── Epoch N-1 = 0.6579 (\u001B[32m↗ 0.0014\u001B[0m)\n", + " └── Best until now = 0.6579 (\u001B[32m↗ 0.0014\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "yuhYeXLA18q5" - }, - "source": [ - "# 6. Predict\n" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 13: 100%|██████████| 309/309 [02:01<00:00, 2.55it/s, BCEDiceLoss=0.219, background_IOU=0.745, gpu_mem=1.14, mean_IOU=0.777, target_IOU=0.81]\n", + "Validating epoch 13: 100%|██████████| 65/65 [00:17<00:00, 3.81it/s]\n", + "[2023-11-13 11:48:23] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:48:23] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.7624021172523499\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "VjRA1tu1mvXQ" - }, - "source": [ - "When the training is complete you can use the trained model to get predictions on the validation set, your data or some other image. Let's load some image and\n", - "run a model inference to create a binary segmentation mask." - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 13\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2194\n", + "│ │ ├── Epoch N-1 = 0.2243 (\u001B[32m↘ -0.0049\u001B[0m)\n", + "│ │ └── Best until now = 0.2243 (\u001B[32m↘ -0.0049\u001B[0m)\n", + "│ ├── Target_iou = 0.8097\n", + "│ │ ├── Epoch N-1 = 0.8068 (\u001B[32m↗ 0.0029\u001B[0m)\n", + "│ │ └── Best until now = 0.8068 (\u001B[32m↗ 0.0029\u001B[0m)\n", + "│ ├── Background_iou = 0.7447\n", + "│ │ ├── Epoch N-1 = 0.736 (\u001B[32m↗ 0.0086\u001B[0m)\n", + "│ │ └── Best until now = 0.736 (\u001B[32m↗ 0.0086\u001B[0m)\n", + "│ └── Mean_iou = 0.7772\n", + "│ ├── Epoch N-1 = 0.7714 (\u001B[32m↗ 0.0058\u001B[0m)\n", + "│ └── Best until now = 0.7714 (\u001B[32m↗ 0.0058\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3024\n", + " │ ├── Epoch N-1 = 0.3035 (\u001B[32m↘ -0.0011\u001B[0m)\n", + " │ └── Best until now = 0.3035 (\u001B[32m↘ -0.0011\u001B[0m)\n", + " ├── Target_iou = 0.7624\n", + " │ ├── Epoch N-1 = 0.7617 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " │ └── Best until now = 0.7617 (\u001B[32m↗ 0.0007\u001B[0m)\n", + " ├── Background_iou = 0.5596\n", + " │ ├── Epoch N-1 = 0.5569 (\u001B[32m↗ 0.0027\u001B[0m)\n", + " │ └── Best until now = 0.5569 (\u001B[32m↗ 0.0027\u001B[0m)\n", + " └── Mean_iou = 0.661\n", + " ├── Epoch N-1 = 0.6593 (\u001B[32m↗ 0.0017\u001B[0m)\n", + " └── Best until now = 0.6593 (\u001B[32m↗ 0.0017\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "code", - "execution_count": 17, - "metadata": { - "id": "Ads7RyGN2JwQ", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 977 - }, - "outputId": "c99ede2d-7fdd-428a-95fe-cac9afbf508b" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "image/png": "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\n" - }, - "metadata": {} - }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "image/png": "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\n" - }, - "metadata": {} - } - ], - "source": [ - "from torchvision.transforms import Compose, ToTensor, Resize, Normalize, ToPILImage\n", - "from PIL import Image\n", - "import torch\n", - "\n", - "pre_proccess = Compose([\n", - " ToTensor(),\n", - " Normalize([.485, .456, .406], [.229, .224, .225])\n", - "])\n", - "\n", - "demo_img_path = os.path.join(root_dir, \"images\", \"ache-adult-depression-expression-41253.png\")\n", - "img = Image.open(demo_img_path)\n", - "# Resize the image and display\n", - "img = Resize(size=(480, 320))(img)\n", - "display(img)\n", - "\n", - "# Run pre-proccess - transforms to tensor and apply normalizations.\n", - "img = pre_proccess(img).unsqueeze(0).cuda()\n", - "\n", - "# Run inference\n", - "model = trainer.net\n", - "model = model.eval()\n", - "mask = model(img)\n", - "\n", - "# Run post-proccess - apply sigmoid to output probabilities, then apply hard\n", - "# threshold of 0.5 for binary mask prediction.\n", - "mask = torch.sigmoid(mask).gt(0.5).squeeze()\n", - "mask = ToPILImage()(mask.float())\n", - "display(mask)\n" - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 14: 100%|██████████| 309/309 [02:01<00:00, 2.54it/s, BCEDiceLoss=0.215, background_IOU=0.748, gpu_mem=1.14, mean_IOU=0.781, target_IOU=0.813]\n", + "Validating epoch 14: 100%|██████████| 65/65 [00:17<00:00, 3.78it/s]\n", + "[2023-11-13 11:50:45] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_quick_start/RUN_20231113_111507_197271/ckpt_best.pth\n", + "[2023-11-13 11:50:45] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.763008713722229\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "-k6ZLKHL1hIM" - }, - "source": [ - "# 7. Convert to ONNX/TensorRT" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 14\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2155\n", + "│ │ ├── Epoch N-1 = 0.2194 (\u001B[32m↘ -0.0039\u001B[0m)\n", + "│ │ └── Best until now = 0.2194 (\u001B[32m↘ -0.0039\u001B[0m)\n", + "│ ├── Target_iou = 0.8134\n", + "│ │ ├── Epoch N-1 = 0.8097 (\u001B[32m↗ 0.0037\u001B[0m)\n", + "│ │ └── Best until now = 0.8097 (\u001B[32m↗ 0.0037\u001B[0m)\n", + "│ ├── Background_iou = 0.7484\n", + "│ │ ├── Epoch N-1 = 0.7447 (\u001B[32m↗ 0.0038\u001B[0m)\n", + "│ │ └── Best until now = 0.7447 (\u001B[32m↗ 0.0038\u001B[0m)\n", + "│ └── Mean_iou = 0.7809\n", + "│ ├── Epoch N-1 = 0.7772 (\u001B[32m↗ 0.0037\u001B[0m)\n", + "│ └── Best until now = 0.7772 (\u001B[32m↗ 0.0037\u001B[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.3015\n", + " │ ├── Epoch N-1 = 0.3024 (\u001B[32m↘ -0.0009\u001B[0m)\n", + " │ └── Best until now = 0.3024 (\u001B[32m↘ -0.0009\u001B[0m)\n", + " ├── Target_iou = 0.763\n", + " │ ├── Epoch N-1 = 0.7624 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " │ └── Best until now = 0.7624 (\u001B[32m↗ 0.0006\u001B[0m)\n", + " ├── Background_iou = 0.5621\n", + " │ ├── Epoch N-1 = 0.5596 (\u001B[32m↗ 0.0025\u001B[0m)\n", + " │ └── Best until now = 0.5596 (\u001B[32m↗ 0.0025\u001B[0m)\n", + " └── Mean_iou = 0.6625\n", + " ├── Epoch N-1 = 0.661 (\u001B[32m↗ 0.0016\u001B[0m)\n", + " └── Best until now = 0.661 (\u001B[32m↗ 0.0016\u001B[0m)\n", + "\n", + "===========================================================\n" + ] }, { - "cell_type": "markdown", - "metadata": { - "id": "br7n55Szm4Nq" - }, - "source": [ - "Let's compile our model to ONNX." - ] + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 11:50:47] INFO - sg_trainer.py - RUNNING ADDITIONAL TEST ON THE AVERAGED MODEL...\n", + "Validating epoch 15: 98%|█████████▊| 64/65 [00:16<00:00, 3.27it/s]" + ] + } + ], + "source": [ + "trainer.train(model=model, training_params=train_params, train_loader=train_loader, valid_loader=valid_loader)" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "id": "X8BJq1crcbjl", + "colab": { + "base_uri": "https://localhost:8080/" }, + "outputId": "661796b8-431a-4c23-ac57-9bdc579a685d" + }, + "outputs": [ { - "cell_type": "code", - "execution_count": 22, - "metadata": { - "id": "q0AGQvEf11PT", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "76b54859-3375-4fc4-c7a7-5b86ed3d80fb" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "ONNX successfully created at: /content/model.onnx\n" - ] - } - ], - "source": [ - "from onnxsim import simplify\n", - "import onnx\n", - "\n", - "onnx_path = os.path.join(os.getcwd(), \"model.onnx\")\n", - "\n", - "input_size = [1, 3, 480, 320]\n", - "model.prep_model_for_conversion(input_size=input_size)\n", - "\n", - "torch.onnx.export(model,\n", - " torch.randn(*input_size).cuda(),\n", - " onnx_path)\n", - "\n", - "# onnx simplifier\n", - "model_sim, check = simplify(onnx_path)\n", - "assert check, \"Simplified ONNX model could not be validated\"\n", - "onnx.save_model(model_sim, onnx_path)\n", - "\n", - "print(\"ONNX successfully created at: \", onnx_path)\n", - "\n" - ] + "output_type": "stream", + "name": "stdout", + "text": [ + "Best Checkpoint mIoU is: 0.763008713722229\n" + ] } - ], - "metadata": { - "accelerator": "GPU", + ], + "source": [ + "print(\"Best Checkpoint mIoU is: \"+ str(trainer.best_metric))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3Nybj15cchxd" + }, + "source": [ + "Now you can download your trained weights from this directory" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "id": "_iHsFgPSciQh" + }, + "outputs": [], + "source": [ + "print(trainer.checkpoints_dir_path)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "yuhYeXLA18q5" + }, + "source": [ + "# 6. Predict\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "VjRA1tu1mvXQ" + }, + "source": [ + "When the training is complete you can use the trained model to get predictions on the validation set, your data or some other image. Let's load some image and\n", + "run a model inference to create a binary segmentation mask." + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": { + "id": "Ads7RyGN2JwQ", "colab": { - "provenance": [] + "base_uri": "https://localhost:8080/", + "height": 977 }, - "kernelspec": { - "display_name": "Python 3", - "name": "python3" + "outputId": "c99ede2d-7fdd-428a-95fe-cac9afbf508b" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} }, - "language_info": { - "name": "python" + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} } + ], + "source": [ + "from torchvision.transforms import Compose, ToTensor, Resize, Normalize, ToPILImage\n", + "from PIL import Image\n", + "import torch\n", + "\n", + "pre_proccess = Compose([\n", + " ToTensor(),\n", + " Normalize([.485, .456, .406], [.229, .224, .225])\n", + "])\n", + "\n", + "demo_img_path = os.path.join(root_dir, \"images\", \"ache-adult-depression-expression-41253.png\")\n", + "img = Image.open(demo_img_path)\n", + "# Resize the image and display\n", + "img = Resize(size=(480, 320))(img)\n", + "display(img)\n", + "\n", + "# Run pre-proccess - transforms to tensor and apply normalizations.\n", + "img = pre_proccess(img).unsqueeze(0).cuda()\n", + "\n", + "# Run inference\n", + "model = trainer.net\n", + "model = model.eval()\n", + "mask = model(img)\n", + "\n", + "# Run post-proccess - apply sigmoid to output probabilities, then apply hard\n", + "# threshold of 0.5 for binary mask prediction.\n", + "mask = torch.sigmoid(mask).gt(0.5).squeeze()\n", + "mask = ToPILImage()(mask.float())\n", + "display(mask)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-k6ZLKHL1hIM" + }, + "source": [ + "# 7. Convert to ONNX/TensorRT" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "br7n55Szm4Nq" + }, + "source": [ + "Let's compile our model to ONNX." + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": { + "id": "q0AGQvEf11PT", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "76b54859-3375-4fc4-c7a7-5b86ed3d80fb" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "ONNX successfully created at: /content/model.onnx\n" + ] + } + ], + "source": [ + "from onnxsim import simplify\n", + "import onnx\n", + "\n", + "onnx_path = os.path.join(os.getcwd(), \"model.onnx\")\n", + "\n", + "input_size = [1, 3, 480, 320]\n", + "model.prep_model_for_conversion(input_size=input_size)\n", + "\n", + "torch.onnx.export(model,\n", + " torch.randn(*input_size).cuda(),\n", + " onnx_path)\n", + "\n", + "# onnx simplifier\n", + "model_sim, check = simplify(onnx_path)\n", + "assert check, \"Simplified ONNX model could not be validated\"\n", + "onnx.save_model(model_sim, onnx_path)\n", + "\n", + "print(\"ONNX successfully created at: \", onnx_path)\n", + "\n" + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" }, - "nbformat": 4, - "nbformat_minor": 0 + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 0 } diff --git a/notebooks/segmentation_connect_custom_dataset.ipynb b/notebooks/segmentation_connect_custom_dataset.ipynb index 1e2d55dad4..8cce51322c 100644 --- a/notebooks/segmentation_connect_custom_dataset.ipynb +++ b/notebooks/segmentation_connect_custom_dataset.ipynb @@ -55,43 +55,74 @@ "base_uri": "https://localhost:8080/" }, "id": "JKce1SM6voVH", - "outputId": "a6397510-a140-443f-f13c-eec1272cc1a8" + "outputId": "e27e79a3-5b89-4869-bf1b-ea54ef60331f" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ - "Requirement already satisfied: torch in /usr/local/lib/python3.10/dist-packages (2.1.0+cu118)\n", - "Requirement already satisfied: torchvision in /usr/local/lib/python3.10/dist-packages (0.16.0+cu118)\n", - "Requirement already satisfied: torchaudio in /usr/local/lib/python3.10/dist-packages (2.1.0+cu118)\n", - "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch) (3.13.1)\n", - "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.10/dist-packages (from torch) (4.5.0)\n", - "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch) (1.12)\n", - "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch) (3.2.1)\n", - "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch) (3.1.2)\n", - "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from torch) (2023.6.0)\n", - "Requirement already satisfied: triton==2.1.0 in /usr/local/lib/python3.10/dist-packages (from torch) (2.1.0)\n", - "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from torchvision) (1.23.5)\n", - "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from torchvision) (2.31.0)\n", - "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision) (9.4.0)\n", - "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch) (2.1.3)\n", - "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.3.2)\n", - "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.4)\n", - "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2.0.7)\n", - "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2023.7.22)\n", - "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch) (1.3.0)\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m12.0/12.0 MB\u001B[0m \u001B[31m36.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m135.8/135.8 kB\u001B[0m \u001B[31m21.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m684.5/684.5 kB\u001B[0m \u001B[31m39.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.9/2.9 MB\u001B[0m \u001B[31m61.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m73.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m408.6/408.6 kB\u001B[0m \u001B[31m24.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m154.5/154.5 kB\u001B[0m \u001B[31m22.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.5/79.5 kB\u001B[0m \u001B[31m11.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m4.5/4.5 MB\u001B[0m \u001B[31m89.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m13.5/13.5 MB\u001B[0m \u001B[31m96.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m68.0/68.0 kB\u001B[0m \u001B[31m10.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for pycocotools (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m17.0/17.0 MB\u001B[0m \u001B[31m88.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m3.3/3.3 MB\u001B[0m \u001B[31m82.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.2/2.2 MB\u001B[0m \u001B[31m77.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m458.9/458.9 kB\u001B[0m \u001B[31m46.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.0/46.0 kB\u001B[0m \u001B[31m6.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m11.3/11.3 MB\u001B[0m \u001B[31m74.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.8/79.8 kB\u001B[0m \u001B[31m11.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m108.3/108.3 kB\u001B[0m \u001B[31m13.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m176.0/176.0 kB\u001B[0m \u001B[31m26.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m407.7/407.7 kB\u001B[0m \u001B[31m45.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m107.7/107.7 kB\u001B[0m \u001B[31m13.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m277.4/277.4 kB\u001B[0m \u001B[31m36.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m111.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m913.9/913.9 kB\u001B[0m \u001B[31m78.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m117.0/117.0 kB\u001B[0m \u001B[31m18.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m575.5/575.5 kB\u001B[0m \u001B[31m50.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m121.1/121.1 kB\u001B[0m \u001B[31m16.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m86.8/86.8 kB\u001B[0m \u001B[31m13.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m16.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m17.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.6/120.6 kB\u001B[0m \u001B[31m16.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m13.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m13.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m84.7/84.7 kB\u001B[0m \u001B[31m13.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m14.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m15.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.8/99.8 kB\u001B[0m \u001B[31m15.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m14.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m14.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m90.6/90.6 kB\u001B[0m \u001B[31m14.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m14.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m14.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m15.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m94.0/94.0 kB\u001B[0m \u001B[31m15.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m105.0/105.0 kB\u001B[0m \u001B[31m15.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.2/46.2 kB\u001B[0m \u001B[31m6.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m106.8/106.8 kB\u001B[0m \u001B[31m17.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m194.6/194.6 kB\u001B[0m \u001B[31m29.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m58.1/58.1 kB\u001B[0m \u001B[31m9.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", + "\u001B[?25h Building wheel for pycocotools (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", " Building wheel for termcolor (setup.py) ... \u001B[?25l\u001B[?25hdone\n", " Building wheel for treelib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", " Building wheel for coverage (setup.py) ... \u001B[?25l\u001B[?25hdone\n", @@ -110,8 +141,7 @@ } ], "source": [ - "! pip install -qq super-gradients==3.4.1\n", - "! pip install -qq prettyformatter" + "! pip install -qq super-gradients==3.4.1" ] }, { @@ -160,16 +190,56 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 2, "metadata": { - "id": "_v1N3kXs3wo1" + "id": "_v1N3kXs3wo1", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "bf8d1fba-3fa0-41af-97a6-9bc23626e78e" }, - "outputs": [], + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is logged into /root/sg_logs/console.log\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 12:01:12] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n", + "[2023-11-13 12:01:13] WARNING - __init__.py - Failed to import pytorch_quantization\n", + "[2023-11-13 12:01:13] INFO - utils.py - NumExpr defaulting to 2 threads.\n", + "[2023-11-13 12:01:24] WARNING - calibrator.py - Failed to import pytorch_quantization\n", + "[2023-11-13 12:01:24] WARNING - export.py - Failed to import pytorch_quantization\n", + "[2023-11-13 12:01:24] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: boto3 required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: deprecated required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: coverage required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: sphinx-rtd-theme required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: torchmetrics required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: hydra-core required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: omegaconf required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: onnxruntime required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: onnx required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: einops required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: treelib required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: stringcase required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: rapidfuzz required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: json-tricks required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: onnx-simplifier required but not found\u001B[0m\n", + "[2023-11-13 12:01:24] WARNING - env_sanity_check.py - \u001B[31mFailed to verify installed packages: data-gradients required but not found\u001B[0m\n" + ] + } + ], "source": [ "from super_gradients.training import Trainer, MultiGPUMode\n", "\n", "\n", - "CHECKPOINT_DIR = '/home/notebook_ckpts/'\n", + "CHECKPOINT_DIR = './notebook_ckpts/'\n", "trainer = Trainer(experiment_name='transfer_learning_semantic_segementation_ppLite', ckpt_root_dir=CHECKPOINT_DIR)" ] }, @@ -188,7 +258,7 @@ "id": "_1TXuJKkKzFJ" }, "source": [ - "## 2.A Generate Proxy Dataset" + "## 2.1 Generate Proxy Dataset" ] }, { @@ -203,7 +273,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 3, "metadata": { "id": "wbdVYnIyjgv-" }, @@ -255,39 +325,15 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 7, "metadata": { - "id": "DXu4yfuZoiv0", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "dccaf4ba-159f-4a47-d13d-ba4b60eaac80" + "id": "DXu4yfuZoiv0" }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Train file `train.txt` content: \n", - "/content/example_data/images/train/0.jpg /content/example_data/labels/train/0.png\n", - "/content/example_data/images/train/1.jpg /content/example_data/labels/train/1.png\n", - "/content/example_data/images/train/2.jpg /content/example_data/labels/train/2.png\n", - "/content/example_data/images/train/3.jpg /content/example_data/labels/train/3.png\n", - "/content/example_data/images/train/4.jpg /content/example_data/labels/train/4.png\n", - "/content/example_data/images/train/5.jpg /content/example_data/labels/train/5.png\n", - "/content/example_data/images/train/6.jpg /content/example_data/labels/train/6.png\n", - "/content/example_data/images/train/7.jpg /content/example_data/labels/train/7.png\n", - "/content/example_data/images/train/8.jpg /content/example_data/labels/train/8.png\n", - "/content/example_data/images/train/9.jpg /content/example_data/labels/train/9.png\n" - ] - } - ], + "outputs": [], "source": [ "num_classes = 10\n", - "generate_proxy_dataset('/content/example_data', num_samples=10, num_classes=num_classes)\n", - "\n", - "print(\"Train file `train.txt` content: \")\n", - "! cat /content/example_data/train.txt" + "data_dir = os.path.join(os.getcwd(), 'example_data')\n", + "generate_proxy_dataset(data_dir, num_samples=10, num_classes=num_classes)" ] }, { @@ -296,12 +342,12 @@ "id": "MDksFYrIqClt" }, "source": [ - "## 2.B Create Torch Dataset" + "## 2.2 Create Torch Dataset" ] }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 8, "metadata": { "id": "AGziBKSIqaUu" }, @@ -350,14 +396,14 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 10, "metadata": { "id": "2B0hlas_1Rh-" }, "outputs": [], "source": [ - "train_dataset = CustomDataset(\"/content/example_data\", split=\"train\")\n", - "val_dataset = CustomDataset(\"/content/example_data\", split=\"val\")\n" + "train_dataset = CustomDataset(data_dir, split=\"train\")\n", + "val_dataset = CustomDataset(data_dir, split=\"val\")\n" ] }, { @@ -371,13 +417,13 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 11, "metadata": { "id": "ZsHqcq1jpN0F", "colab": { "base_uri": "https://localhost:8080/" }, - "outputId": "e9c1182a-5359-45b6-c0f3-ad430a4fc67d" + "outputId": "5976e582-e08a-4642-bd04-eaf613d04a97" }, "outputs": [ { @@ -404,16 +450,9 @@ "## 2.C Create Torch Dataloader" ] }, - { - "cell_type": "markdown", - "metadata": { - "id": "D3ThxDIopDDB" - }, - "source": [] - }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 12, "metadata": { "id": "XrWjWfjXnw_r" }, @@ -437,13 +476,13 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 13, "metadata": { "id": "O-KuZQ3XBduM", "colab": { "base_uri": "https://localhost:8080/" }, - "outputId": "489fc05e-f972-464d-c150-360e4f2dbd7e" + "outputId": "7addf002-9c95-4252-fad7-510b2a0ab333" }, "outputs": [ { @@ -454,7 +493,7 @@ ] }, "metadata": {}, - "execution_count": 10 + "execution_count": 13 } ], "source": [ @@ -494,13 +533,13 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 14, "metadata": { "id": "YDK4btf04Gbu", "colab": { "base_uri": "https://localhost:8080/" }, - "outputId": "cc7f1ab1-3a01-49c1-c9a2-6ca434dcc192" + "outputId": "11d5aad2-dc94-4231-8e61-c4b968857370" }, "outputs": [ { @@ -508,8 +547,8 @@ "name": "stderr", "text": [ "Downloading: \"https://sghub.deci.ai/models/pp_lite_t_seg75_cityscapes.pth\" to /root/.cache/torch/hub/checkpoints/pp_lite_t_seg75_cityscapes.pth\n", - "100%|██████████| 31.4M/31.4M [00:01<00:00, 32.4MB/s]\n", - "[2023-11-12 14:41:45] INFO - checkpoint_utils.py - Successfully loaded pretrained weights for architecture pp_lite_t_seg75\n" + "100%|██████████| 31.4M/31.4M [00:00<00:00, 223MB/s]\n", + "[2023-11-13 12:03:09] INFO - checkpoint_utils.py - Successfully loaded pretrained weights for architecture pp_lite_t_seg75\n" ] } ], @@ -553,7 +592,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 15, "metadata": { "id": "3eRe0hBz4G1n" }, @@ -567,7 +606,7 @@ " \"lr_mode\": \"cosine\",\n", " \"initial_lr\": 0.005,\n", " \"optimizer\": \"SGD\",\n", - " \"loss\": \"cross_entropy\",\n", + " \"loss\": \"CrossEntropyLoss\",\n", " \"average_best_models\": False,\n", " \"metric_to_watch\": \"IoU\",\n", " \"greater_metric_to_watch_is_better\": True,\n", @@ -603,38 +642,36 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 16, "metadata": { "id": "-Ojnc1bk9L3s", "colab": { "base_uri": "https://localhost:8080/" }, - "outputId": "b36b8b9b-b554-444e-d440-02bf623c3efa" + "outputId": "2c2ac68b-75f7-48c0-db0f-7b579c743013" }, "outputs": [ { "output_type": "stream", "name": "stderr", "text": [ - "[2023-11-12 14:41:51] WARNING - sg_trainer.py - Train dataset size % batch_size != 0 and drop_last=False, this might result in smaller last batch.\n", - "[2023-11-12 14:41:58] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231112_144158_892860`\n", - "[2023-11-12 14:41:58] INFO - sg_trainer.py - Checkpoints directory: /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860\n", - "/usr/local/lib/python3.10/dist-packages/super_gradients/common/registry/registry.py:72: DeprecationWarning: Object name `cross_entropy` is now deprecated. Please replace it with `CrossEntropyLoss`.\n", - " warnings.warn(f\"Object name `{name}` is now deprecated. Please replace it with `{deprecated_names[name]}`.\", DeprecationWarning)\n" + "[2023-11-13 12:03:20] WARNING - sg_trainer.py - Train dataset size % batch_size != 0 and drop_last=False, this might result in smaller last batch.\n", + "[2023-11-13 12:03:28] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231113_120328_854372`\n", + "[2023-11-13 12:03:28] INFO - sg_trainer.py - Checkpoints directory: ./notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231113_120328_854372\n" ] }, { "output_type": "stream", "name": "stdout", "text": [ - "The console stream is now moved to /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860/console_Nov12_14_41_58.txt\n" + "The console stream is now moved to ./notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231113_120328_854372/console_Nov13_12_03_28.txt\n" ] }, { "output_type": "stream", "name": "stderr", "text": [ - "[2023-11-12 14:41:59] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", + "[2023-11-13 12:03:29] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", " - Mode: Single GPU\n", " - Number of GPUs: 1 (1 available on the machine)\n", " - Full dataset size: 10 (len(train_set))\n", @@ -645,12 +682,12 @@ " - Iterations per epoch: 3 (len(train_loader))\n", " - Gradient updates per epoch: 3 (len(train_loader) / batch_accumulate)\n", "\n", - "[2023-11-12 14:41:59] INFO - sg_trainer.py - Started training for 10 epochs (0/9)\n", + "[2023-11-13 12:03:29] INFO - sg_trainer.py - Started training for 10 epochs (0/9)\n", "\n", - "Train epoch 0: 100%|██████████| 3/3 [00:08<00:00, 2.91s/it, CrossEntropyLoss=3.49, IoU=0.0319, gpu_mem=0.686]\n", - "Validating: 100%|██████████| 3/3 [00:00<00:00, 3.83it/s]\n", - "[2023-11-12 14:42:09] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231112_144158_892860/ckpt_best.pth\n", - "[2023-11-12 14:42:09] INFO - sg_trainer.py - Best checkpoint overriden: validation IoU: 0.013753225095570087\n", + "Train epoch 0: 100%|██████████| 3/3 [00:08<00:00, 2.99s/it, CrossEntropyLoss=3.04, IoU=0.0464, gpu_mem=0.686]\n", + "Validating: 100%|██████████| 3/3 [00:00<00:00, 3.41it/s]\n", + "[2023-11-13 12:03:39] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/transfer_learning_semantic_segementation_ppLite/RUN_20231113_120328_854372/ckpt_best.pth\n", + "[2023-11-13 12:03:39] INFO - sg_trainer.py - Best checkpoint overriden: validation IoU: 0.015268713235855103\n", "Train epoch 1: 0%| | 0/3 [00:00=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision) (9.4.0)\n", - "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch) (2.1.3)\n", - "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.3.2)\n", - "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.4)\n", - "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2.0.7)\n", - "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2023.7.22)\n", - "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch) (1.3.0)\n", - " Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", - " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m135.8/135.8 kB\u001B[0m \u001B[31m3.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m684.5/684.5 kB\u001B[0m \u001B[31m21.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.9/2.9 MB\u001B[0m \u001B[31m71.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m102.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m408.6/408.6 kB\u001B[0m \u001B[31m48.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m154.5/154.5 kB\u001B[0m \u001B[31m23.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.5/79.5 kB\u001B[0m \u001B[31m12.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m4.5/4.5 MB\u001B[0m \u001B[31m108.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m13.5/13.5 MB\u001B[0m \u001B[31m96.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m68.0/68.0 kB\u001B[0m \u001B[31m9.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Installing build dependencies ... \u001B[?25l\u001B[?25hdone\n", - " Getting requirements to build wheel ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m17.0/17.0 MB\u001B[0m \u001B[31m83.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m3.3/3.3 MB\u001B[0m \u001B[31m54.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.2/2.2 MB\u001B[0m \u001B[31m81.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m458.9/458.9 kB\u001B[0m \u001B[31m41.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.0/46.0 kB\u001B[0m \u001B[31m4.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m11.3/11.3 MB\u001B[0m \u001B[31m76.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m79.8/79.8 kB\u001B[0m \u001B[31m6.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m108.3/108.3 kB\u001B[0m \u001B[31m13.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m176.0/176.0 kB\u001B[0m \u001B[31m21.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m407.7/407.7 kB\u001B[0m \u001B[31m35.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m107.7/107.7 kB\u001B[0m \u001B[31m13.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m277.4/277.4 kB\u001B[0m \u001B[31m25.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.8/2.8 MB\u001B[0m \u001B[31m68.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m913.9/913.9 kB\u001B[0m \u001B[31m59.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m117.0/117.0 kB\u001B[0m \u001B[31m14.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Preparing metadata (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m575.5/575.5 kB\u001B[0m \u001B[31m52.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m121.1/121.1 kB\u001B[0m \u001B[31m18.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m86.8/86.8 kB\u001B[0m \u001B[31m12.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m13.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.0/120.0 kB\u001B[0m \u001B[31m15.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m120.6/120.6 kB\u001B[0m \u001B[31m17.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m11.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m83.5/83.5 kB\u001B[0m \u001B[31m12.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m84.7/84.7 kB\u001B[0m \u001B[31m12.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m13.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.2/99.2 kB\u001B[0m \u001B[31m13.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.8/99.8 kB\u001B[0m \u001B[31m12.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m12.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m89.4/89.4 kB\u001B[0m \u001B[31m11.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m90.6/90.6 kB\u001B[0m \u001B[31m11.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m13.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m12.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m92.6/92.6 kB\u001B[0m \u001B[31m13.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m94.0/94.0 kB\u001B[0m \u001B[31m7.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m105.0/105.0 kB\u001B[0m \u001B[31m12.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.2/46.2 kB\u001B[0m \u001B[31m6.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m106.8/106.8 kB\u001B[0m \u001B[31m15.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m194.6/194.6 kB\u001B[0m \u001B[31m24.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[2K \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m58.1/58.1 kB\u001B[0m \u001B[31m8.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n", - "\u001B[?25h Building wheel for super-gradients (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for pycocotools (pyproject.toml) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for termcolor (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for treelib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for coverage (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for xhtml2pdf (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for antlr4-python3-runtime (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for stringcase (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - " Building wheel for svglib (setup.py) ... \u001B[?25l\u001B[?25hdone\n", - "\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", - "lida 0.0.10 requires fastapi, which is not installed.\n", - "lida 0.0.10 requires kaleido, which is not installed.\n", - "lida 0.0.10 requires python-multipart, which is not installed.\n", - "lida 0.0.10 requires uvicorn, which is not installed.\n", - "tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.23.0 which is incompatible.\u001B[0m\u001B[31m\n", - "\u001B[0m" - ] - } - ], - "source": [ - "! pip install -qq super-gradients==3.4.1\n", - "! pip install -qq prettyformatter" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "892xArqDsGsQ" - }, - "source": [ - "# 1. Experiment setup\n", - "We will initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", - "\n", - "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", - "\n", - "```\n", - "ckpt_root_dir\n", - "|─── experiment_name_1\n", - "│ ckpt_best.pth # Model checkpoint on best epoch\n", - "│ ckpt_latest.pth # Model checkpoint on last epoch\n", - "│ average_model.pth # Model checkpoint averaged over epochs\n", - "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", - "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", - "└─── experiment_name_2\n", - " ...\n", - "```\n", - "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", - " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "pl0WPz1HisFz" - }, - "source": [ - "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "cell_type": "markdown", + "metadata": { + "id": "HY_HuQbxn7X0" + }, + "source": [ + "![SG - Horizontal.png](data:image/png;base64,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)" + ] }, - "id": "HAff--HysJmP", - "outputId": "4d3b9778-480b-4b72-ad8a-b3c257962771" - }, - "outputs": [ { - "output_type": "stream", - "name": "stdout", - "text": [ - "The console stream is logged into /root/sg_logs/console.log\n" - ] - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-12 13:59:13] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n", - "[2023-11-12 13:59:13] WARNING - __init__.py - Failed to import pytorch_quantization\n", - "[2023-11-12 13:59:14] INFO - utils.py - NumExpr defaulting to 2 threads.\n", - "[2023-11-12 13:59:28] WARNING - calibrator.py - Failed to import pytorch_quantization\n", - "[2023-11-12 13:59:28] WARNING - export.py - Failed to import pytorch_quantization\n", - "[2023-11-12 13:59:28] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n", - "[2023-11-12 13:59:29] INFO - env_sanity_check.py - Library check is not supported when super_gradients installed through \"git+https://github.com/...\" command\n" - ] - } - ], - "source": [ - "from super_gradients import Trainer\n", - "\n", - "CHECKPOINT_DIR = '/home/notebook_ckpts/'\n", - "trainer = Trainer(experiment_name=\"segmentation_transfer_learning\", ckpt_root_dir=CHECKPOINT_DIR)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "dwVMY4gMjQSL" - }, - "source": [ - "# 2. Dataset definition\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "fpIWhnR9j2rm" - }, - "source": [ - "\n", - "For the sake of this presentation, we'll use **Supervisely** semantic segmentation dataset." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "ZACgRb-qjzDJ" - }, - "source": [ - "SG trainer is fully compatible with PyTorch data loaders, so you can definitely use your own data for the experiment below if you prefer." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "6ulV6Hpao3IN" - }, - "source": [ - "## 2.A. Download data\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "mVwslNv-j-2C" - }, - "source": [ - "Feel free to change the download path by editing SUPERVISELY_DATASET_DOWNLOAD_PATH" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": { - "id": "dfR18Rmbo00y", - "colab": { - "base_uri": "https://localhost:8080/" + "cell_type": "markdown", + "metadata": { + "id": "oA_p5zIsoAJQ" + }, + "source": [ + "# SuperGradients Transfer Learning Semantic Segmentation\n", + "\n", + "In the following tutorial, we will demonstrate how to use one of SuperGradients pre-trained models and a custom dataset to improve the model's accuracy using transfer learning for semantic segmentation.\n", + "\n", + "Transfer Learning from a pre-trained checkpoint on your own dataset could prove to be very effective instead of deploying the pre-trained model directly on your data or training a model from scratch. For more information on Transfer Learning, please visit: https://cs231n.github.io/transfer-learning/\n", + "\n", + "In the following example, transfer learning will be used from a Cityscapes PPLiteSeg model to a subset of the Supervisely person segmentation dataset .\n", + "\n", + "For more details about the FILTERED dataset we wil use see: https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.3/contrib/PP-HumanSeg\n", + "\n", + "An example real-world use-case of the afformentioned scenario could be background removal in real time for video confrences.[link text](https://)\n", + "\n", + "The notebook is divided into 7 sections:\n", + "1. Experiment setup\n", + "2. Dataset definition\n", + "3. Architecture definition\n", + "4. Training setup\n", + "5. Training and Evaluation\n", + "6. Predict\n", + "7. Convert to ONNX\\TensorRT" + ] }, - "outputId": "72323bbd-b94f-4488-a14e-6a5dee578d62" - }, - "outputs": [ { - "output_type": "stream", - "name": "stdout", - "text": [ - "Downloading and extracting supervisely dataset to: /home/data\n", - "/home/data\n", - "--2023-11-12 13:59:29-- https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", - "Resolving deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)... 52.217.166.25, 52.217.204.241, 3.5.25.206, ...\n", - "Connecting to deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)|52.217.166.25|:443... connected.\n", - "HTTP request sent, awaiting response... 200 OK\n", - "Length: 3564001012 (3.3G) [application/zip]\n", - "Saving to: ‘supervisely-persons.zip’\n", - "\n", - "supervisely-persons 100%[===================>] 3.32G 61.8MB/s in 62s \n", - "\n", - "2023-11-12 14:00:31 (55.2 MB/s) - ‘supervisely-persons.zip’ saved [3564001012/3564001012]\n", - "\n" - ] - } - ], - "source": [ - "import os\n", - "\n", - "SUPERVISELY_DATASET_DOWNLOAD_PATH=\"/home/data\"\n", - "\n", - "supervisely_dataset_dir_path = SUPERVISELY_DATASET_DOWNLOAD_PATH + os.path.sep + 'supervisely-persons'\n", - "\n", - "if os.path.isdir(supervisely_dataset_dir_path):\n", - " print('supervisely dataset already downloaded...')\n", - "else:\n", - " print('Downloading and extracting supervisely dataset to: ' + SUPERVISELY_DATASET_DOWNLOAD_PATH)\n", - " ! mkdir $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", - " %cd $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", - " ! wget https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", - " ! unzip --qq supervisely-persons.zip" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "collapsed": false, - "id": "V9ZcklupX8Qx" - }, - "source": [ - "## 2.B. Create data loaders\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "3Mk_YixjlEhj" - }, - "source": [ - "The dataloaders are initiated with the default parameters defined in the [yaml](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/recipes/dataset_params/supervisely_persons_dataset_params.yaml)\n", - "file. Parameters as batch_size, transforms, root_dir and others can be overridden by passing as `dataset_params` and\n", - "`dataloader_params`, as implemented bellow." - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "metadata": { - "id": "S3BzMRhSX8Qx" - }, - "outputs": [], - "source": [ - "from super_gradients.training import dataloaders\n", - "\n", - "root_dir = supervisely_dataset_dir_path\n", - "batch_size = 8\n", - "\n", - "train_loader = dataloaders.supervisely_persons_train(\n", - " dataset_params={\"root_dir\": root_dir},\n", - " dataloader_params={\"batch_size\": batch_size, \"num_workers\": 2}\n", - ")\n", - "valid_loader = dataloaders.supervisely_persons_val(\n", - " dataset_params={\"root_dir\": root_dir},\n", - " dataloader_params={\"batch_size\": batch_size, \"num_workers\": 2}\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "6dHIwvs46-dk" - }, - "source": [ - "As you can see, we didn't have to pass many parameters into the dataloaders construction. That's because defaults are pre-defined for your convenience, and you might be curious to know what they are. Let's print them and see which resolution and transformations are defined." - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": { - "id": "76tzhKxi6aS-" - }, - "outputs": [], - "source": [ - "from prettyformatter import pprint\n", - "\n", - "print('Dataloader parameters:')\n", - "pprint(train_loader.dataloader_params)\n", - "print('Dataset parameters')\n", - "pprint(train_loader.dataset.dataset_params)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "I4QEOkKyy93R" - }, - "source": [ - "We can take a look at some images from the dataset." - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "metadata": { - "id": "xXPMJQCJzmb4", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 937 + "cell_type": "markdown", + "metadata": { + "id": "GqH4VGMroWec" + }, + "source": [ + "#Install SG" + ] }, - "outputId": "f7b11090-410d-4d86-8730-b27ebfe385b2" - }, - "outputs": [ { - "output_type": "stream", - "name": "stdout", - "text": [ - "Dataloader parameters:\n", - "{\"batch_size\": 8, \"num_workers\": 2, \"shuffle\": True, \"drop_last\": True}\n", - "Dataset parameters\n", - "{'root_dir': '/home/data/supervisely-persons', 'list_file': 'train.csv', 'cache_labels': False, 'cache_images': False, 'transforms': [{'SegRandomRescale': {'scales': [0.25, 1.0]}}, {'SegColorJitter': {'brightness': 0.5, 'contrast': 0.5, 'saturation': 0.5}}, {'SegRandomFlip': {'prob': 0.5}}, {'SegPadShortToCropSize': {'crop_size': [320, 480], 'fill_mask': 0}}, {'SegCropImageAndMask': {'crop_size': [320, 480], 'mode': 'random'}}]}\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "Q8uA6AWEhHN6" + }, + "source": [ + "The cell below will install **super_gradients** which will automatically get all its dependencies." + ] }, { - "output_type": "display_data", - "data": { - "text/plain": [ - "" + "cell_type": "code", + "execution_count": 1, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "-mm-E4xRoNEm", + "outputId": "dccd5fad-db84-4066-ff41-30582a4a50d2" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12.0/12.0 MB\u001b[0m \u001b[31m42.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m135.8/135.8 kB\u001b[0m \u001b[31m18.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m684.5/684.5 kB\u001b[0m \u001b[31m59.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.9/2.9 MB\u001b[0m \u001b[31m41.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.8/2.8 MB\u001b[0m \u001b[31m34.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m408.6/408.6 kB\u001b[0m \u001b[31m15.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m154.5/154.5 kB\u001b[0m \u001b[31m9.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m79.5/79.5 kB\u001b[0m \u001b[31m8.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m4.5/4.5 MB\u001b[0m \u001b[31m35.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m13.5/13.5 MB\u001b[0m \u001b[31m38.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m68.0/68.0 kB\u001b[0m \u001b[31m4.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n", + " Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m17.0/17.0 MB\u001b[0m \u001b[31m89.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m3.3/3.3 MB\u001b[0m \u001b[31m66.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.2/2.2 MB\u001b[0m \u001b[31m37.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m458.9/458.9 kB\u001b[0m \u001b[31m47.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m46.0/46.0 kB\u001b[0m \u001b[31m6.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m11.3/11.3 MB\u001b[0m \u001b[31m113.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m79.8/79.8 kB\u001b[0m \u001b[31m12.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m108.3/108.3 kB\u001b[0m \u001b[31m17.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m176.0/176.0 kB\u001b[0m \u001b[31m22.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m407.7/407.7 kB\u001b[0m \u001b[31m49.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m107.7/107.7 kB\u001b[0m \u001b[31m17.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m277.4/277.4 kB\u001b[0m \u001b[31m35.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.8/2.8 MB\u001b[0m \u001b[31m80.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m913.9/913.9 kB\u001b[0m \u001b[31m58.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m117.0/117.0 kB\u001b[0m \u001b[31m16.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m575.5/575.5 kB\u001b[0m \u001b[31m46.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m121.1/121.1 kB\u001b[0m \u001b[31m17.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m86.8/86.8 kB\u001b[0m \u001b[31m10.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m120.0/120.0 kB\u001b[0m \u001b[31m16.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m120.0/120.0 kB\u001b[0m \u001b[31m15.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m120.6/120.6 kB\u001b[0m \u001b[31m16.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m83.5/83.5 kB\u001b[0m \u001b[31m11.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m83.5/83.5 kB\u001b[0m \u001b[31m11.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m84.7/84.7 kB\u001b[0m \u001b[31m11.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m99.2/99.2 kB\u001b[0m \u001b[31m12.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m99.2/99.2 kB\u001b[0m \u001b[31m13.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m99.8/99.8 kB\u001b[0m \u001b[31m13.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m89.4/89.4 kB\u001b[0m \u001b[31m7.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m89.4/89.4 kB\u001b[0m \u001b[31m860.6 kB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m90.6/90.6 kB\u001b[0m \u001b[31m14.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m92.6/92.6 kB\u001b[0m \u001b[31m15.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m92.6/92.6 kB\u001b[0m \u001b[31m11.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m92.6/92.6 kB\u001b[0m \u001b[31m14.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m94.0/94.0 kB\u001b[0m \u001b[31m15.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m105.0/105.0 kB\u001b[0m \u001b[31m17.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m46.2/46.2 kB\u001b[0m \u001b[31m7.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m106.8/106.8 kB\u001b[0m \u001b[31m17.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m194.6/194.6 kB\u001b[0m \u001b[31m29.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m58.1/58.1 kB\u001b[0m \u001b[31m10.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Building wheel for pycocotools (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for termcolor (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for treelib (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for coverage (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for xhtml2pdf (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for antlr4-python3-runtime (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for stringcase (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Building wheel for svglib (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n", + "lida 0.0.10 requires fastapi, which is not installed.\n", + "lida 0.0.10 requires kaleido, which is not installed.\n", + "lida 0.0.10 requires python-multipart, which is not installed.\n", + "lida 0.0.10 requires uvicorn, which is not installed.\n", + "tensorflow 2.14.0 requires numpy>=1.23.5, but you have numpy 1.23.0 which is incompatible.\u001b[0m\u001b[31m\n", + "\u001b[0m" + ] + } ], - "image/png": "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\n" - }, - "metadata": {} + "source": [ + "! pip install -qq super-gradients==3.4.1" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "/usr/local/lib/python3.10/dist-packages/torchvision/transforms/functional.py:1603: UserWarning: The default value of the antialias parameter of all the resizing transforms (Resize(), RandomResizedCrop(), etc.) will change from None to True in v0.17, in order to be consistent across the PIL and Tensor backends. To suppress this warning, directly pass antialias=True (recommended, future default), antialias=None (current default, which means False for Tensors and True for PIL), or antialias=False (only works on Tensors - PIL will still use antialiasing). This also applies if you are using the inference transforms from the models weights: update the call to weights.transforms(antialias=True).\n", - " warnings.warn(\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "892xArqDsGsQ" + }, + "source": [ + "# 1. Experiment setup\n", + "We will initialize our **trainer** which will be in charge of everything, like training, evaluation, saving checkpoints, plotting etc.\n", + "\n", + "The **experiment name** argument is important as every checkpoints, logs and tensorboards to be saved in a directory with the same name. This directory will be created as a sub-directory of **ckpt_root_dir** as follow:\n", + "\n", + "```\n", + "ckpt_root_dir\n", + "|─── experiment_name_1\n", + "│ ckpt_best.pth # Model checkpoint on best epoch\n", + "│ ckpt_latest.pth # Model checkpoint on last epoch\n", + "│ average_model.pth # Model checkpoint averaged over epochs\n", + "│ events.out.tfevents.1659878383... # Tensorflow artifacts of a specific run\n", + "│ log_Aug07_11_52_48.txt # Trainer logs of a specific run\n", + "└─── experiment_name_2\n", + " ...\n", + "```\n", + "In this notebook multi-gpu training is set as `OFF`, for Distributed training multi_gpu can be set as\n", + " `MultiGPUMode.DISTRIBUTED_DATA_PARALLEL` or `MultiGPUMode.DATA_PARALLEL`." + ] }, { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "image/png": "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\n" - }, - "metadata": {} + "cell_type": "markdown", + "metadata": { + "id": "pl0WPz1HisFz" + }, + "source": [ + "Let's define **ckpt_root_dir** inside the Colab, later we can use it to start TensorBoard and monitor the run." + ] }, { - "output_type": "display_data", - "data": { - "text/plain": [ - "" + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "HAff--HysJmP", + "outputId": "23704251-ac0f-4a11-a104-d320484243de" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 12:17:11] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is logged into /root/sg_logs/console.log\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 12:17:11] WARNING - __init__.py - Failed to import pytorch_quantization\n", + "[2023-11-13 12:17:11] INFO - utils.py - NumExpr defaulting to 2 threads.\n", + "[2023-11-13 12:17:26] WARNING - calibrator.py - Failed to import pytorch_quantization\n", + "[2023-11-13 12:17:26] WARNING - export.py - Failed to import pytorch_quantization\n", + "[2023-11-13 12:17:26] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: boto3 required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: deprecated required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: coverage required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: sphinx-rtd-theme required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: torchmetrics required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: hydra-core required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: omegaconf required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: onnxruntime required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: onnx required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: einops required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: treelib required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: stringcase required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: rapidfuzz required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: json-tricks required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: onnx-simplifier required but not found\u001b[0m\n", + "[2023-11-13 12:17:26] WARNING - env_sanity_check.py - \u001b[31mFailed to verify installed packages: data-gradients required but not found\u001b[0m\n" + ] + } ], - "image/png": "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\n" - }, - "metadata": {} - } - ], - "source": [ - "from PIL import Image\n", - "from torchvision.utils import draw_segmentation_masks\n", - "from torchvision.transforms import ToTensor, ToPILImage, Resize\n", - "import numpy as np\n", - "import torch\n", - "\n", - "def plot_seg_data(img_path: str, target_path: str):\n", - " image = (ToTensor()(Image.open(img_path).convert('RGB')) * 255).type(torch.uint8)\n", - " target = torch.from_numpy(np.array(Image.open(target_path))).bool()\n", - " image = draw_segmentation_masks(image, target, colors=\"red\", alpha=0.4)\n", - " image = Resize(size=200)(image)\n", - " display(ToPILImage()(image))\n", - "\n", - "for i in range(4, 7):\n", - " img_path, target_path = train_loader.dataset.samples_targets_tuples_list[i]\n", - " plot_seg_data(img_path, target_path)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "l5GcDAg_pUGJ" - }, - "source": [ - "# 3. Architecture definition\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "fU8orO7wlwIK" - }, - "source": [ - "SG includes implementations of many different architectures for semantic segmentation tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "-oGSU3V8lqcm" - }, - "source": [ - "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", - "and extra Auxiliary heads aren't used for training.\n", - "\n", - "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "metadata": { - "id": "f6ZTsO0nrdje", - "colab": { - "base_uri": "https://localhost:8080/" + "source": [ + "from super_gradients import Trainer\n", + "\n", + "CHECKPOINT_DIR = './notebook_ckpts/'\n", + "trainer = Trainer(experiment_name=\"segmentation_transfer_learning\", ckpt_root_dir=CHECKPOINT_DIR)" + ] }, - "outputId": "dc6dc47a-402a-4bc1-b5a6-2af9728270dd" - }, - "outputs": [ { - "output_type": "stream", - "name": "stderr", - "text": [ - "Downloading: \"https://sghub.deci.ai/models/pp_lite_t_seg75_cityscapes.pth\" to /root/.cache/torch/hub/checkpoints/pp_lite_t_seg75_cityscapes.pth\n", - "100%|██████████| 31.4M/31.4M [00:01<00:00, 22.4MB/s]\n", - "[2023-11-12 14:01:13] INFO - checkpoint_utils.py - Successfully loaded pretrained weights for architecture pp_lite_t_seg75\n" - ] - } - ], - "source": [ - "from super_gradients.training import models\n", - "from super_gradients.common.object_names import Models\n", - "\n", - "model = models.get(model_name=Models.PP_LITE_T_SEG75,\n", - " arch_params={\"use_aux_heads\": False},\n", - " num_classes=1,\n", - " pretrained_weights=\"cityscapes\")" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "X-_dBewgr1dG" - }, - "source": [ - "# 4. Training setup\n", - "\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "H1Rll8Orl-Dy" - }, - "source": [ - "\n", - "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n", - "\n", - "We will be using an average of BCE and Dice loss for segmentation, with different learning rates for the replaced segmentation head layer, and the rest of the network- this is controlled by the `multiply_head_lr` parameter which is the multiplication factor of the learning rate for the newly replaced layer.\n", - "\n", - "As our `metric_to_watch`, we will be monitoring the `target_IOU` which is one of the components of `BinaryIOU` torchmetrics object (the other components are `mean_IOU` which is the mean of the background and target IOUs, and `background_IOU`)." - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "metadata": { - "id": "NShu3zLgr5qD" - }, - "outputs": [], - "source": [ - "from super_gradients.training.metrics.segmentation_metrics import BinaryIOU\n", - "from super_gradients.training.utils.callbacks import BinarySegmentationVisualizationCallback, Phase\n", - "\n", - "train_params = {\"max_epochs\": 15,\n", - " \"lr_mode\": \"cosine\",\n", - " \"initial_lr\": 0.005,\n", - " \"lr_warmup_epochs\": 5,\n", - " \"multiply_head_lr\": 10,\n", - " \"optimizer\": \"SGD\",\n", - " \"loss\": \"bce_dice_loss\",\n", - " \"ema\": True,\n", - " \"zero_weight_decay_on_bias_and_bn\": True,\n", - " \"average_best_models\": True,\n", - " \"metric_to_watch\": \"target_IOU\",\n", - " \"greater_metric_to_watch_is_better\": True,\n", - " \"train_metrics_list\": [BinaryIOU()],\n", - " \"valid_metrics_list\": [BinaryIOU()],\n", - " \"loss_logging_items_names\": [\"loss\"],\n", - " \"phase_callbacks\": [BinarySegmentationVisualizationCallback(phase=Phase.VALIDATION_BATCH_END,\n", - " freq=1,\n", - " last_img_idx_in_batch=4)],\n", - " }" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "qTECVyhcs506" - }, - "source": [ - "# 5. Training and evaluation\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "S1K5MU2kmmDb" - }, - "source": [ - "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", - "\n", - "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders.\n", - "\n", - "**Note:** While training, don't forget to refresh the tensorboard with the arrow on the top right." - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "metadata": { - "id": "u6roEj9ktFTi", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "2e9c3992-0566-4cba-e7aa-ae4853cc442e" - }, - "outputs": [ - { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "[2023-11-12 14:01:21] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231112_140121_753664`\n", - "[2023-11-12 14:01:21] INFO - sg_trainer.py - Checkpoints directory: /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664\n", - "/usr/local/lib/python3.10/dist-packages/super_gradients/common/registry/registry.py:72: DeprecationWarning: Object name `bce_dice_loss` is now deprecated. Please replace it with `BCEDiceLoss`.\n", - " warnings.warn(f\"Object name `{name}` is now deprecated. Please replace it with `{deprecated_names[name]}`.\", DeprecationWarning)\n", - "[2023-11-12 14:01:21] INFO - sg_trainer.py - Using EMA with params {}\n", - "[2023-11-12 14:01:21] WARNING - ema.py - Parameter `decay` is not specified for EMA params. Please specify `decay` parameter explicitly in your config:\n", - "ema: True\n", - "ema_params: \n", - " decay: 0.9999\n", - " decay_type: exp\n", - " beta: 15\n", - "Will default to decay: 0.9999\n", - "In the next major release of SG this warning will become an error.\n", - "[2023-11-12 14:01:21] WARNING - ema.py - Parameter decay_type is not specified for EMA model. Please specify decay_type parameter explicitly in your config:\n", - "ema: True\n", - "ema_params: \n", - " decay: 0.9999\n", - " decay_type: constant|exp|threshold\n", - "Will default to `exp` decay with beta = 15\n", - "In the next major release of SG this warning will become an error.\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "The console stream is now moved to /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/console_Nov12_14_01_21.txt\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "[2023-11-12 14:01:24] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", - " - Mode: Single GPU\n", - " - Number of GPUs: 1 (1 available on the machine)\n", - " - Full dataset size: 2477 (len(train_set))\n", - " - Batch size per GPU: 8 (batch_size)\n", - " - Batch Accumulate: 1 (batch_accumulate)\n", - " - Total batch size: 8 (num_gpus * batch_size)\n", - " - Effective Batch size: 8 (num_gpus * batch_size * batch_accumulate)\n", - " - Iterations per epoch: 309 (len(train_loader))\n", - " - Gradient updates per epoch: 309 (len(train_loader) / batch_accumulate)\n", - "\n", - "[2023-11-12 14:01:24] INFO - sg_trainer.py - Started training for 15 epochs (0/14)\n", - "\n", - "Train epoch 0: 100%|██████████| 309/309 [01:56<00:00, 2.66it/s, BCEDiceLoss=0.223, background_IOU=0.749, gpu_mem=1.14, mean_IOU=0.779, target_IOU=0.809]\n", - "Validating: 100%|██████████| 65/65 [00:15<00:00, 4.20it/s]\n", - "[2023-11-12 14:03:36] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:03:36] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.857506275177002\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 0\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.2225\n", - "│ ├── Target_iou = 0.8094\n", - "│ ├── Background_iou = 0.7486\n", - "│ └── Mean_iou = 0.779\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1821\n", - " ├── Target_iou = 0.8575\n", - " ├── Background_iou = 0.7387\n", - " └── Mean_iou = 0.7981\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "dwVMY4gMjQSL" + }, + "source": [ + "# 2. Dataset definition\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 1: 100%|██████████| 309/309 [01:42<00:00, 3.02it/s, BCEDiceLoss=0.167, background_IOU=0.815, gpu_mem=1.14, mean_IOU=0.837, target_IOU=0.86]\n", - "Validating epoch 1: 100%|██████████| 65/65 [00:16<00:00, 3.96it/s]\n", - "[2023-11-12 14:05:38] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:05:38] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.8748109340667725\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "fpIWhnR9j2rm" + }, + "source": [ + "\n", + "For the sake of this presentation, we'll use **Supervisely** semantic segmentation dataset." + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 1\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.1665\n", - "│ │ ├── Epoch N-1 = 0.2225 (\u001B[32m↘ -0.056\u001B[0m)\n", - "│ │ └── Best until now = 0.2225 (\u001B[32m↘ -0.056\u001B[0m)\n", - "│ ├── Target_iou = 0.8598\n", - "│ │ ├── Epoch N-1 = 0.8094 (\u001B[32m↗ 0.0503\u001B[0m)\n", - "│ │ └── Best until now = 0.8094 (\u001B[32m↗ 0.0503\u001B[0m)\n", - "│ ├── Background_iou = 0.815\n", - "│ │ ├── Epoch N-1 = 0.7486 (\u001B[32m↗ 0.0663\u001B[0m)\n", - "│ │ └── Best until now = 0.7486 (\u001B[32m↗ 0.0663\u001B[0m)\n", - "│ └── Mean_iou = 0.8374\n", - "│ ├── Epoch N-1 = 0.779 (\u001B[32m↗ 0.0583\u001B[0m)\n", - "│ └── Best until now = 0.779 (\u001B[32m↗ 0.0583\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1594\n", - " │ ├── Epoch N-1 = 0.1821 (\u001B[32m↘ -0.0227\u001B[0m)\n", - " │ └── Best until now = 0.1821 (\u001B[32m↘ -0.0227\u001B[0m)\n", - " ├── Target_iou = 0.8748\n", - " │ ├── Epoch N-1 = 0.8575 (\u001B[32m↗ 0.0173\u001B[0m)\n", - " │ └── Best until now = 0.8575 (\u001B[32m↗ 0.0173\u001B[0m)\n", - " ├── Background_iou = 0.7766\n", - " │ ├── Epoch N-1 = 0.7387 (\u001B[32m↗ 0.0379\u001B[0m)\n", - " │ └── Best until now = 0.7387 (\u001B[32m↗ 0.0379\u001B[0m)\n", - " └── Mean_iou = 0.8257\n", - " ├── Epoch N-1 = 0.7981 (\u001B[32m↗ 0.0276\u001B[0m)\n", - " └── Best until now = 0.7981 (\u001B[32m↗ 0.0276\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "ZACgRb-qjzDJ" + }, + "source": [ + "SG trainer is fully compatible with PyTorch data loaders, so you can definitely use your own data for the experiment below if you prefer." + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 2: 100%|██████████| 309/309 [01:40<00:00, 3.08it/s, BCEDiceLoss=0.145, background_IOU=0.844, gpu_mem=1.14, mean_IOU=0.86, target_IOU=0.877]\n", - "Validating epoch 2: 100%|██████████| 65/65 [00:19<00:00, 3.39it/s]\n", - "[2023-11-12 14:07:39] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:07:39] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.8893157839775085\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "6ulV6Hpao3IN" + }, + "source": [ + "## 2.1 Download data\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 2\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.1449\n", - "│ │ ├── Epoch N-1 = 0.1665 (\u001B[32m↘ -0.0216\u001B[0m)\n", - "│ │ └── Best until now = 0.1665 (\u001B[32m↘ -0.0216\u001B[0m)\n", - "│ ├── Target_iou = 0.8768\n", - "│ │ ├── Epoch N-1 = 0.8598 (\u001B[32m↗ 0.0171\u001B[0m)\n", - "│ │ └── Best until now = 0.8598 (\u001B[32m↗ 0.0171\u001B[0m)\n", - "│ ├── Background_iou = 0.8437\n", - "│ │ ├── Epoch N-1 = 0.815 (\u001B[32m↗ 0.0287\u001B[0m)\n", - "│ │ └── Best until now = 0.815 (\u001B[32m↗ 0.0287\u001B[0m)\n", - "│ └── Mean_iou = 0.8603\n", - "│ ├── Epoch N-1 = 0.8374 (\u001B[32m↗ 0.0229\u001B[0m)\n", - "│ └── Best until now = 0.8374 (\u001B[32m↗ 0.0229\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1405\n", - " │ ├── Epoch N-1 = 0.1594 (\u001B[32m↘ -0.0189\u001B[0m)\n", - " │ └── Best until now = 0.1594 (\u001B[32m↘ -0.0189\u001B[0m)\n", - " ├── Target_iou = 0.8893\n", - " │ ├── Epoch N-1 = 0.8748 (\u001B[32m↗ 0.0145\u001B[0m)\n", - " │ └── Best until now = 0.8748 (\u001B[32m↗ 0.0145\u001B[0m)\n", - " ├── Background_iou = 0.8025\n", - " │ ├── Epoch N-1 = 0.7766 (\u001B[32m↗ 0.0259\u001B[0m)\n", - " │ └── Best until now = 0.7766 (\u001B[32m↗ 0.0259\u001B[0m)\n", - " └── Mean_iou = 0.8459\n", - " ├── Epoch N-1 = 0.8257 (\u001B[32m↗ 0.0202\u001B[0m)\n", - " └── Best until now = 0.8257 (\u001B[32m↗ 0.0202\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "mVwslNv-j-2C" + }, + "source": [ + "Feel free to change the download path by editing SUPERVISELY_DATASET_DOWNLOAD_PATH" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 3: 100%|██████████| 309/309 [01:50<00:00, 2.79it/s, BCEDiceLoss=0.13, background_IOU=0.858, gpu_mem=1.14, mean_IOU=0.873, target_IOU=0.887]\n", - "Validating epoch 3: 100%|██████████| 65/65 [00:16<00:00, 4.00it/s]\n", - "[2023-11-12 14:09:48] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:09:48] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.8972753882408142\n" - ] - }, - { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 3\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.1296\n", - "│ │ ├── Epoch N-1 = 0.1449 (\u001B[32m↘ -0.0153\u001B[0m)\n", - "│ │ └── Best until now = 0.1449 (\u001B[32m↘ -0.0153\u001B[0m)\n", - "│ ├── Target_iou = 0.8869\n", - "│ │ ├── Epoch N-1 = 0.8768 (\u001B[32m↗ 0.0101\u001B[0m)\n", - "│ │ └── Best until now = 0.8768 (\u001B[32m↗ 0.0101\u001B[0m)\n", - "│ ├── Background_iou = 0.8581\n", - "│ │ ├── Epoch N-1 = 0.8437 (\u001B[32m↗ 0.0144\u001B[0m)\n", - "│ │ └── Best until now = 0.8437 (\u001B[32m↗ 0.0144\u001B[0m)\n", - "│ └── Mean_iou = 0.8725\n", - "│ ├── Epoch N-1 = 0.8603 (\u001B[32m↗ 0.0123\u001B[0m)\n", - "│ └── Best until now = 0.8603 (\u001B[32m↗ 0.0123\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1304\n", - " │ ├── Epoch N-1 = 0.1405 (\u001B[32m↘ -0.0101\u001B[0m)\n", - " │ └── Best until now = 0.1405 (\u001B[32m↘ -0.0101\u001B[0m)\n", - " ├── Target_iou = 0.8973\n", - " │ ├── Epoch N-1 = 0.8893 (\u001B[32m↗ 0.008\u001B[0m)\n", - " │ └── Best until now = 0.8893 (\u001B[32m↗ 0.008\u001B[0m)\n", - " ├── Background_iou = 0.8195\n", - " │ ├── Epoch N-1 = 0.8025 (\u001B[32m↗ 0.017\u001B[0m)\n", - " │ └── Best until now = 0.8025 (\u001B[32m↗ 0.017\u001B[0m)\n", - " └── Mean_iou = 0.8584\n", - " ├── Epoch N-1 = 0.8459 (\u001B[32m↗ 0.0125\u001B[0m)\n", - " └── Best until now = 0.8459 (\u001B[32m↗ 0.0125\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "code", + "execution_count": 3, + "metadata": { + "id": "dfR18Rmbo00y", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "07e535c6-2091-4179-843c-ce6e7cd591f1" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Downloading and extracting supervisely dataset to: /content/data\n", + "/content/data\n", + "--2023-11-13 12:17:26-- https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", + "Resolving deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)... 3.5.27.182, 3.5.29.153, 52.216.37.233, ...\n", + "Connecting to deci-pretrained-models.s3.amazonaws.com (deci-pretrained-models.s3.amazonaws.com)|3.5.27.182|:443... connected.\n", + "HTTP request sent, awaiting response... 200 OK\n", + "Length: 3564001012 (3.3G) [application/zip]\n", + "Saving to: ‘supervisely-persons.zip’\n", + "\n", + "supervisely-persons 100%[===================>] 3.32G 38.1MB/s in 88s \n", + "\n", + "2023-11-13 12:18:54 (38.7 MB/s) - ‘supervisely-persons.zip’ saved [3564001012/3564001012]\n", + "\n" + ] + } + ], + "source": [ + "import os\n", + "\n", + "SUPERVISELY_DATASET_DOWNLOAD_PATH=os.path.join(os.getcwd(),\"data\")\n", + "\n", + "supervisely_dataset_dir_path = os.path.join(SUPERVISELY_DATASET_DOWNLOAD_PATH, 'supervisely-persons')\n", + "\n", + "if os.path.isdir(supervisely_dataset_dir_path):\n", + " print('supervisely dataset already downloaded...')\n", + "else:\n", + " print('Downloading and extracting supervisely dataset to: ' + SUPERVISELY_DATASET_DOWNLOAD_PATH)\n", + " ! mkdir $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " %cd $SUPERVISELY_DATASET_DOWNLOAD_PATH\n", + " ! wget https://deci-pretrained-models.s3.amazonaws.com/supervisely-persons.zip\n", + " ! unzip --qq supervisely-persons.zip" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 4: 100%|██████████| 309/309 [01:46<00:00, 2.91it/s, BCEDiceLoss=0.12, background_IOU=0.867, gpu_mem=1.14, mean_IOU=0.882, target_IOU=0.897]\n", - "Validating epoch 4: 100%|██████████| 65/65 [00:14<00:00, 4.37it/s]\n", - "[2023-11-12 14:11:52] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:11:52] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9037814736366272\n" - ] + "cell_type": "markdown", + "metadata": { + "collapsed": false, + "id": "V9ZcklupX8Qx" + }, + "source": [ + "## 2.2 Create data loaders\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 4\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.1202\n", - "│ │ ├── Epoch N-1 = 0.1296 (\u001B[32m↘ -0.0094\u001B[0m)\n", - "│ │ └── Best until now = 0.1296 (\u001B[32m↘ -0.0094\u001B[0m)\n", - "│ ├── Target_iou = 0.8972\n", - "│ │ ├── Epoch N-1 = 0.8869 (\u001B[32m↗ 0.0103\u001B[0m)\n", - "│ │ └── Best until now = 0.8869 (\u001B[32m↗ 0.0103\u001B[0m)\n", - "│ ├── Background_iou = 0.8672\n", - "│ │ ├── Epoch N-1 = 0.8581 (\u001B[32m↗ 0.0091\u001B[0m)\n", - "│ │ └── Best until now = 0.8581 (\u001B[32m↗ 0.0091\u001B[0m)\n", - "│ └── Mean_iou = 0.8822\n", - "│ ├── Epoch N-1 = 0.8725 (\u001B[32m↗ 0.0097\u001B[0m)\n", - "│ └── Best until now = 0.8725 (\u001B[32m↗ 0.0097\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1215\n", - " │ ├── Epoch N-1 = 0.1304 (\u001B[32m↘ -0.0089\u001B[0m)\n", - " │ └── Best until now = 0.1304 (\u001B[32m↘ -0.0089\u001B[0m)\n", - " ├── Target_iou = 0.9038\n", - " │ ├── Epoch N-1 = 0.8973 (\u001B[32m↗ 0.0065\u001B[0m)\n", - " │ └── Best until now = 0.8973 (\u001B[32m↗ 0.0065\u001B[0m)\n", - " ├── Background_iou = 0.8312\n", - " │ ├── Epoch N-1 = 0.8195 (\u001B[32m↗ 0.0117\u001B[0m)\n", - " │ └── Best until now = 0.8195 (\u001B[32m↗ 0.0117\u001B[0m)\n", - " └── Mean_iou = 0.8675\n", - " ├── Epoch N-1 = 0.8584 (\u001B[32m↗ 0.0091\u001B[0m)\n", - " └── Best until now = 0.8584 (\u001B[32m↗ 0.0091\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "3Mk_YixjlEhj" + }, + "source": [ + "The dataloaders are initiated with the default parameters defined in the [yaml](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/recipes/dataset_params/supervisely_persons_dataset_params.yaml)\n", + "file. Parameters as batch_size, transforms, root_dir and others can be overridden by passing as `dataset_params` and\n", + "`dataloader_params`, as implemented bellow." + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 5: 100%|██████████| 309/309 [01:44<00:00, 2.97it/s, BCEDiceLoss=0.116, background_IOU=0.871, gpu_mem=1.14, mean_IOU=0.885, target_IOU=0.9]\n", - "Validating epoch 5: 100%|██████████| 65/65 [00:16<00:00, 3.94it/s]\n", - "[2023-11-12 14:13:57] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:13:57] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9058071374893188\n" - ] + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "S3BzMRhSX8Qx" + }, + "outputs": [], + "source": [ + "from super_gradients.training import dataloaders\n", + "\n", + "root_dir = supervisely_dataset_dir_path\n", + "batch_size = 8\n", + "\n", + "train_loader = dataloaders.supervisely_persons_train(\n", + " dataset_params={\"root_dir\": root_dir},\n", + " dataloader_params={\"batch_size\": batch_size, \"num_workers\": 2}\n", + ")\n", + "valid_loader = dataloaders.supervisely_persons_val(\n", + " dataset_params={\"root_dir\": root_dir},\n", + " dataloader_params={\"batch_size\": batch_size, \"num_workers\": 2}\n", + ")" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 5\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.1162\n", - "│ │ ├── Epoch N-1 = 0.1202 (\u001B[32m↘ -0.004\u001B[0m)\n", - "│ │ └── Best until now = 0.1202 (\u001B[32m↘ -0.004\u001B[0m)\n", - "│ ├── Target_iou = 0.8997\n", - "│ │ ├── Epoch N-1 = 0.8972 (\u001B[32m↗ 0.0024\u001B[0m)\n", - "│ │ └── Best until now = 0.8972 (\u001B[32m↗ 0.0024\u001B[0m)\n", - "│ ├── Background_iou = 0.8707\n", - "│ │ ├── Epoch N-1 = 0.8672 (\u001B[32m↗ 0.0035\u001B[0m)\n", - "│ │ └── Best until now = 0.8672 (\u001B[32m↗ 0.0035\u001B[0m)\n", - "│ └── Mean_iou = 0.8852\n", - "│ ├── Epoch N-1 = 0.8822 (\u001B[32m↗ 0.003\u001B[0m)\n", - "│ └── Best until now = 0.8822 (\u001B[32m↗ 0.003\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1191\n", - " │ ├── Epoch N-1 = 0.1215 (\u001B[32m↘ -0.0025\u001B[0m)\n", - " │ └── Best until now = 0.1215 (\u001B[32m↘ -0.0025\u001B[0m)\n", - " ├── Target_iou = 0.9058\n", - " │ ├── Epoch N-1 = 0.9038 (\u001B[32m↗ 0.002\u001B[0m)\n", - " │ └── Best until now = 0.9038 (\u001B[32m↗ 0.002\u001B[0m)\n", - " ├── Background_iou = 0.8351\n", - " │ ├── Epoch N-1 = 0.8312 (\u001B[32m↗ 0.0039\u001B[0m)\n", - " │ └── Best until now = 0.8312 (\u001B[32m↗ 0.0039\u001B[0m)\n", - " └── Mean_iou = 0.8705\n", - " ├── Epoch N-1 = 0.8675 (\u001B[32m↗ 0.003\u001B[0m)\n", - " └── Best until now = 0.8675 (\u001B[32m↗ 0.003\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "6dHIwvs46-dk" + }, + "source": [ + "As you can see, we didn't have to pass many parameters into the dataloaders construction. That's because defaults are pre-defined for your convenience, and you might be curious to know what they are. Let's print them and see which resolution and transformations are defined." + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 6: 100%|██████████| 309/309 [01:44<00:00, 2.96it/s, BCEDiceLoss=0.117, background_IOU=0.871, gpu_mem=1.14, mean_IOU=0.885, target_IOU=0.899]\n", - "Validating epoch 6: 100%|██████████| 65/65 [00:14<00:00, 4.57it/s]\n", - "[2023-11-12 14:15:58] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:15:58] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9092355966567993\n" - ] + "cell_type": "code", + "execution_count": 5, + "metadata": { + "id": "76tzhKxi6aS-" + }, + "outputs": [], + "source": [ + "print('Dataloader parameters:')\n", + "print(train_loader.dataloader_params)\n", + "print('Dataset parameters')\n", + "print(train_loader.dataset.dataset_params)" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 6\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.1167\n", - "│ │ ├── Epoch N-1 = 0.1162 (\u001B[31m↗ 0.0005\u001B[0m)\n", - "│ │ └── Best until now = 0.1162 (\u001B[31m↗ 0.0005\u001B[0m)\n", - "│ ├── Target_iou = 0.8993\n", - "│ │ ├── Epoch N-1 = 0.8997 (\u001B[31m↘ -0.0003\u001B[0m)\n", - "│ │ └── Best until now = 0.8997 (\u001B[31m↘ -0.0003\u001B[0m)\n", - "│ ├── Background_iou = 0.8713\n", - "│ │ ├── Epoch N-1 = 0.8707 (\u001B[32m↗ 0.0005\u001B[0m)\n", - "│ │ └── Best until now = 0.8707 (\u001B[32m↗ 0.0005\u001B[0m)\n", - "│ └── Mean_iou = 0.8853\n", - "│ ├── Epoch N-1 = 0.8852 (\u001B[32m↗ 0.0001\u001B[0m)\n", - "│ └── Best until now = 0.8852 (\u001B[32m↗ 0.0001\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.115\n", - " │ ├── Epoch N-1 = 0.1191 (\u001B[32m↘ -0.004\u001B[0m)\n", - " │ └── Best until now = 0.1191 (\u001B[32m↘ -0.004\u001B[0m)\n", - " ├── Target_iou = 0.9092\n", - " │ ├── Epoch N-1 = 0.9058 (\u001B[32m↗ 0.0034\u001B[0m)\n", - " │ └── Best until now = 0.9058 (\u001B[32m↗ 0.0034\u001B[0m)\n", - " ├── Background_iou = 0.8415\n", - " │ ├── Epoch N-1 = 0.8351 (\u001B[32m↗ 0.0064\u001B[0m)\n", - " │ └── Best until now = 0.8351 (\u001B[32m↗ 0.0064\u001B[0m)\n", - " └── Mean_iou = 0.8754\n", - " ├── Epoch N-1 = 0.8705 (\u001B[32m↗ 0.0049\u001B[0m)\n", - " └── Best until now = 0.8705 (\u001B[32m↗ 0.0049\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "I4QEOkKyy93R" + }, + "source": [ + "We can take a look at some images from the dataset." + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 7: 100%|██████████| 309/309 [01:42<00:00, 3.02it/s, BCEDiceLoss=0.108, background_IOU=0.879, gpu_mem=1.14, mean_IOU=0.892, target_IOU=0.906]\n", - "Validating epoch 7: 100%|██████████| 65/65 [00:19<00:00, 3.29it/s]\n", - "[2023-11-12 14:18:04] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:18:04] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.911444902420044\n" - ] + "cell_type": "code", + "execution_count": 6, + "metadata": { + "id": "xXPMJQCJzmb4", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 937 + }, + "outputId": "c7605343-9fe4-4bb1-b18c-12e0cad6e0a7" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Dataloader parameters:\n", + "{'batch_size': 8, 'num_workers': 2, 'shuffle': True, 'drop_last': True}\n", + "Dataset parameters\n", + "{'root_dir': '/content/data/supervisely-persons', 'list_file': 'train.csv', 'cache_labels': False, 'cache_images': False, 'transforms': [{'SegRandomRescale': {'scales': [0.25, 1.0]}}, {'SegColorJitter': {'brightness': 0.5, 'contrast': 0.5, 'saturation': 0.5}}, {'SegRandomFlip': {'prob': 0.5}}, {'SegPadShortToCropSize': {'crop_size': [320, 480], 'fill_mask': 0}}, {'SegCropImageAndMask': {'crop_size': [320, 480], 'mode': 'random'}}]}\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.10/dist-packages/torchvision/transforms/functional.py:1603: UserWarning: The default value of the antialias parameter of all the resizing transforms (Resize(), RandomResizedCrop(), etc.) will change from None to True in v0.17, in order to be consistent across the PIL and Tensor backends. To suppress this warning, directly pass antialias=True (recommended, future default), antialias=None (current default, which means False for Tensors and True for PIL), or antialias=False (only works on Tensors - PIL will still use antialiasing). This also applies if you are using the inference transforms from the models weights: update the call to weights.transforms(antialias=True).\n", + " warnings.warn(\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "from PIL import Image\n", + "from torchvision.utils import draw_segmentation_masks\n", + "from torchvision.transforms import ToTensor, ToPILImage, Resize\n", + "import numpy as np\n", + "import torch\n", + "\n", + "def plot_seg_data(img_path: str, target_path: str):\n", + " image = (ToTensor()(Image.open(img_path).convert('RGB')) * 255).type(torch.uint8)\n", + " target = torch.from_numpy(np.array(Image.open(target_path))).bool()\n", + " image = draw_segmentation_masks(image, target, colors=\"red\", alpha=0.4)\n", + " image = Resize(size=200)(image)\n", + " display(ToPILImage()(image))\n", + "\n", + "for i in range(4, 7):\n", + " img_path, target_path = train_loader.dataset.samples_targets_tuples_list[i]\n", + " plot_seg_data(img_path, target_path)" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 7\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.1079\n", - "│ │ ├── Epoch N-1 = 0.1167 (\u001B[32m↘ -0.0088\u001B[0m)\n", - "│ │ └── Best until now = 0.1162 (\u001B[32m↘ -0.0083\u001B[0m)\n", - "│ ├── Target_iou = 0.906\n", - "│ │ ├── Epoch N-1 = 0.8993 (\u001B[32m↗ 0.0067\u001B[0m)\n", - "│ │ └── Best until now = 0.8997 (\u001B[32m↗ 0.0063\u001B[0m)\n", - "│ ├── Background_iou = 0.8786\n", - "│ │ ├── Epoch N-1 = 0.8713 (\u001B[32m↗ 0.0073\u001B[0m)\n", - "│ │ └── Best until now = 0.8713 (\u001B[32m↗ 0.0073\u001B[0m)\n", - "│ └── Mean_iou = 0.8923\n", - "│ ├── Epoch N-1 = 0.8853 (\u001B[32m↗ 0.007\u001B[0m)\n", - "│ └── Best until now = 0.8853 (\u001B[32m↗ 0.007\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1122\n", - " │ ├── Epoch N-1 = 0.115 (\u001B[32m↘ -0.0028\u001B[0m)\n", - " │ └── Best until now = 0.115 (\u001B[32m↘ -0.0028\u001B[0m)\n", - " ├── Target_iou = 0.9114\n", - " │ ├── Epoch N-1 = 0.9092 (\u001B[32m↗ 0.0022\u001B[0m)\n", - " │ └── Best until now = 0.9092 (\u001B[32m↗ 0.0022\u001B[0m)\n", - " ├── Background_iou = 0.8456\n", - " │ ├── Epoch N-1 = 0.8415 (\u001B[32m↗ 0.0041\u001B[0m)\n", - " │ └── Best until now = 0.8415 (\u001B[32m↗ 0.0041\u001B[0m)\n", - " └── Mean_iou = 0.8785\n", - " ├── Epoch N-1 = 0.8754 (\u001B[32m↗ 0.0032\u001B[0m)\n", - " └── Best until now = 0.8754 (\u001B[32m↗ 0.0032\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "l5GcDAg_pUGJ" + }, + "source": [ + "# 3. Architecture definition\n", + "\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 8: 100%|██████████| 309/309 [01:44<00:00, 2.96it/s, BCEDiceLoss=0.0988, background_IOU=0.889, gpu_mem=1.14, mean_IOU=0.902, target_IOU=0.915]\n", - "Validating epoch 8: 100%|██████████| 65/65 [00:15<00:00, 4.31it/s]\n", - "[2023-11-12 14:20:06] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:20:06] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9122714996337891\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "fU8orO7wlwIK" + }, + "source": [ + "SG includes implementations of many different architectures for semantic segmentation tasks that can be found [here](https://github.com/Deci-AI/super-gradients#implemented-model-architectures)." + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 8\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.0988\n", - "│ │ ├── Epoch N-1 = 0.1079 (\u001B[32m↘ -0.0091\u001B[0m)\n", - "│ │ └── Best until now = 0.1079 (\u001B[32m↘ -0.0091\u001B[0m)\n", - "│ ├── Target_iou = 0.9146\n", - "│ │ ├── Epoch N-1 = 0.906 (\u001B[32m↗ 0.0086\u001B[0m)\n", - "│ │ └── Best until now = 0.906 (\u001B[32m↗ 0.0086\u001B[0m)\n", - "│ ├── Background_iou = 0.8893\n", - "│ │ ├── Epoch N-1 = 0.8786 (\u001B[32m↗ 0.0107\u001B[0m)\n", - "│ │ └── Best until now = 0.8786 (\u001B[32m↗ 0.0107\u001B[0m)\n", - "│ └── Mean_iou = 0.9019\n", - "│ ├── Epoch N-1 = 0.8923 (\u001B[32m↗ 0.0096\u001B[0m)\n", - "│ └── Best until now = 0.8923 (\u001B[32m↗ 0.0096\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1112\n", - " │ ├── Epoch N-1 = 0.1122 (\u001B[32m↘ -0.001\u001B[0m)\n", - " │ └── Best until now = 0.1122 (\u001B[32m↘ -0.001\u001B[0m)\n", - " ├── Target_iou = 0.9123\n", - " │ ├── Epoch N-1 = 0.9114 (\u001B[32m↗ 0.0008\u001B[0m)\n", - " │ └── Best until now = 0.9114 (\u001B[32m↗ 0.0008\u001B[0m)\n", - " ├── Background_iou = 0.8472\n", - " │ ├── Epoch N-1 = 0.8456 (\u001B[32m↗ 0.0016\u001B[0m)\n", - " │ └── Best until now = 0.8456 (\u001B[32m↗ 0.0016\u001B[0m)\n", - " └── Mean_iou = 0.8797\n", - " ├── Epoch N-1 = 0.8785 (\u001B[32m↗ 0.0012\u001B[0m)\n", - " └── Best until now = 0.8785 (\u001B[32m↗ 0.0012\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "-oGSU3V8lqcm" + }, + "source": [ + "Create a PPLiteSeg nn.Module, with 1 class segmentation head classifier. For simplicity `use_aux_head` is set as `False`\n", + "and extra Auxiliary heads aren't used for training.\n", + "\n", + "Other segmentation modules can be used for this task such as, DDRNet, STDC and RegSeg.\n" + ] }, { - "metadata": { - "tags": null - }, - "name": "stderr", - "output_type": "stream", - "text": [ - "Train epoch 9: 100%|██████████| 309/309 [01:41<00:00, 3.05it/s, BCEDiceLoss=0.0923, background_IOU=0.897, gpu_mem=1.14, mean_IOU=0.908, target_IOU=0.919]\n", - "Validating epoch 9: 100%|██████████| 65/65 [00:15<00:00, 4.11it/s]\n", - "[2023-11-12 14:22:09] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:22:09] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9127917289733887\n" - ] + "cell_type": "code", + "execution_count": 7, + "metadata": { + "id": "f6ZTsO0nrdje", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "5f5dffd8-738f-4fa7-e5c4-df8e2bd5622d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Downloading: \"https://sghub.deci.ai/models/pp_lite_t_seg75_cityscapes.pth\" to /root/.cache/torch/hub/checkpoints/pp_lite_t_seg75_cityscapes.pth\n", + "100%|██████████| 31.4M/31.4M [00:01<00:00, 26.4MB/s]\n", + "[2023-11-13 12:19:38] INFO - checkpoint_utils.py - Successfully loaded pretrained weights for architecture pp_lite_t_seg75\n" + ] + } + ], + "source": [ + "from super_gradients.training import models\n", + "from super_gradients.common.object_names import Models\n", + "\n", + "model = models.get(model_name=Models.PP_LITE_T_SEG75,\n", + " arch_params={\"use_aux_heads\": False},\n", + " num_classes=1,\n", + " pretrained_weights=\"cityscapes\")" + ] }, { - "metadata": { - "tags": null - }, - "name": "stdout", - "output_type": "stream", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 9\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.0923\n", - "│ │ ├── Epoch N-1 = 0.0988 (\u001B[32m↘ -0.0065\u001B[0m)\n", - "│ │ └── Best until now = 0.0988 (\u001B[32m↘ -0.0065\u001B[0m)\n", - "│ ├── Target_iou = 0.9194\n", - "│ │ ├── Epoch N-1 = 0.9146 (\u001B[32m↗ 0.0048\u001B[0m)\n", - "│ │ └── Best until now = 0.9146 (\u001B[32m↗ 0.0048\u001B[0m)\n", - "│ ├── Background_iou = 0.8968\n", - "│ │ ├── Epoch N-1 = 0.8893 (\u001B[32m↗ 0.0075\u001B[0m)\n", - "│ │ └── Best until now = 0.8893 (\u001B[32m↗ 0.0075\u001B[0m)\n", - "│ └── Mean_iou = 0.9081\n", - "│ ├── Epoch N-1 = 0.9019 (\u001B[32m↗ 0.0062\u001B[0m)\n", - "│ └── Best until now = 0.9019 (\u001B[32m↗ 0.0062\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1106\n", - " │ ├── Epoch N-1 = 0.1112 (\u001B[32m↘ -0.0006\u001B[0m)\n", - " │ └── Best until now = 0.1112 (\u001B[32m↘ -0.0006\u001B[0m)\n", - " ├── Target_iou = 0.9128\n", - " │ ├── Epoch N-1 = 0.9123 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " │ └── Best until now = 0.9123 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " ├── Background_iou = 0.8482\n", - " │ ├── Epoch N-1 = 0.8472 (\u001B[32m↗ 0.001\u001B[0m)\n", - " │ └── Best until now = 0.8472 (\u001B[32m↗ 0.001\u001B[0m)\n", - " └── Mean_iou = 0.8805\n", - " ├── Epoch N-1 = 0.8797 (\u001B[32m↗ 0.0007\u001B[0m)\n", - " └── Best until now = 0.8797 (\u001B[32m↗ 0.0007\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "X-_dBewgr1dG" + }, + "source": [ + "# 4. Training setup\n", + "\n", + "\n" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 10: 100%|██████████| 309/309 [01:43<00:00, 3.00it/s, BCEDiceLoss=0.0851, background_IOU=0.904, gpu_mem=1.14, mean_IOU=0.914, target_IOU=0.925]\n", - "Validating epoch 10: 100%|██████████| 65/65 [00:15<00:00, 4.22it/s]\n", - "[2023-11-12 14:24:11] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:24:11] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9131874442100525\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "H1Rll8Orl-Dy" + }, + "source": [ + "\n", + "Here we define the training recipe. The full parameters can be found here [training parameters supported](https://deci-ai.github.io/super-gradients/user_guide.html#training-parameters).\n", + "\n", + "We will be using an average of BCE and Dice loss for segmentation, with different learning rates for the replaced segmentation head layer, and the rest of the network- this is controlled by the `multiply_head_lr` parameter which is the multiplication factor of the learning rate for the newly replaced layer.\n", + "\n", + "As our `metric_to_watch`, we will be monitoring the `target_IOU` which is one of the components of `BinaryIOU` torchmetrics object (the other components are `mean_IOU` which is the mean of the background and target IOUs, and `background_IOU`)." + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 10\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.0851\n", - "│ │ ├── Epoch N-1 = 0.0923 (\u001B[32m↘ -0.0072\u001B[0m)\n", - "│ │ └── Best until now = 0.0923 (\u001B[32m↘ -0.0072\u001B[0m)\n", - "│ ├── Target_iou = 0.9251\n", - "│ │ ├── Epoch N-1 = 0.9194 (\u001B[32m↗ 0.0057\u001B[0m)\n", - "│ │ └── Best until now = 0.9194 (\u001B[32m↗ 0.0057\u001B[0m)\n", - "│ ├── Background_iou = 0.9037\n", - "│ │ ├── Epoch N-1 = 0.8968 (\u001B[32m↗ 0.0068\u001B[0m)\n", - "│ │ └── Best until now = 0.8968 (\u001B[32m↗ 0.0068\u001B[0m)\n", - "│ └── Mean_iou = 0.9144\n", - "│ ├── Epoch N-1 = 0.9081 (\u001B[32m↗ 0.0063\u001B[0m)\n", - "│ └── Best until now = 0.9081 (\u001B[32m↗ 0.0063\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1101\n", - " │ ├── Epoch N-1 = 0.1106 (\u001B[32m↘ -0.0005\u001B[0m)\n", - " │ └── Best until now = 0.1106 (\u001B[32m↘ -0.0005\u001B[0m)\n", - " ├── Target_iou = 0.9132\n", - " │ ├── Epoch N-1 = 0.9128 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " │ └── Best until now = 0.9128 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " ├── Background_iou = 0.8489\n", - " │ ├── Epoch N-1 = 0.8482 (\u001B[32m↗ 0.0008\u001B[0m)\n", - " │ └── Best until now = 0.8482 (\u001B[32m↗ 0.0008\u001B[0m)\n", - " └── Mean_iou = 0.881\n", - " ├── Epoch N-1 = 0.8805 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " └── Best until now = 0.8805 (\u001B[32m↗ 0.0006\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "code", + "execution_count": 8, + "metadata": { + "id": "NShu3zLgr5qD" + }, + "outputs": [], + "source": [ + "from super_gradients.training.metrics.segmentation_metrics import BinaryIOU\n", + "from super_gradients.training.utils.callbacks import BinarySegmentationVisualizationCallback, Phase\n", + "\n", + "train_params = {\"max_epochs\": 15,\n", + " \"lr_mode\": \"cosine\",\n", + " \"initial_lr\": 0.005,\n", + " \"lr_warmup_epochs\": 5,\n", + " \"multiply_head_lr\": 10,\n", + " \"optimizer\": \"SGD\",\n", + " \"loss\": \"BCEDiceLoss\",\n", + " \"ema\": True,\n", + " \"ema_params\":\n", + " {\n", + " \"decay\": 0.9999,\n", + " \"decay_type\": \"exp\",\n", + " \"beta\": 15,\n", + " },\n", + " \"zero_weight_decay_on_bias_and_bn\": True,\n", + " \"average_best_models\": True,\n", + " \"metric_to_watch\": \"target_IOU\",\n", + " \"greater_metric_to_watch_is_better\": True,\n", + " \"train_metrics_list\": [BinaryIOU()],\n", + " \"valid_metrics_list\": [BinaryIOU()],\n", + " \"loss_logging_items_names\": [\"loss\"],\n", + " \"phase_callbacks\": [BinarySegmentationVisualizationCallback(phase=Phase.VALIDATION_BATCH_END,\n", + " freq=1,\n", + " last_img_idx_in_batch=4)],\n", + " }" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 11: 100%|██████████| 309/309 [01:40<00:00, 3.06it/s, BCEDiceLoss=0.0809, background_IOU=0.907, gpu_mem=1.14, mean_IOU=0.918, target_IOU=0.93]\n", - "Validating epoch 11: 100%|██████████| 65/65 [00:15<00:00, 4.15it/s]\n", - "[2023-11-12 14:26:10] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:26:10] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9135512709617615\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "qTECVyhcs506" + }, + "source": [ + "# 5. Training and evaluation\n" + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 11\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.0809\n", - "│ │ ├── Epoch N-1 = 0.0851 (\u001B[32m↘ -0.0042\u001B[0m)\n", - "│ │ └── Best until now = 0.0851 (\u001B[32m↘ -0.0042\u001B[0m)\n", - "│ ├── Target_iou = 0.9295\n", - "│ │ ├── Epoch N-1 = 0.9251 (\u001B[32m↗ 0.0044\u001B[0m)\n", - "│ │ └── Best until now = 0.9251 (\u001B[32m↗ 0.0044\u001B[0m)\n", - "│ ├── Background_iou = 0.9068\n", - "│ │ ├── Epoch N-1 = 0.9037 (\u001B[32m↗ 0.0032\u001B[0m)\n", - "│ │ └── Best until now = 0.9037 (\u001B[32m↗ 0.0032\u001B[0m)\n", - "│ └── Mean_iou = 0.9182\n", - "│ ├── Epoch N-1 = 0.9144 (\u001B[32m↗ 0.0038\u001B[0m)\n", - "│ └── Best until now = 0.9144 (\u001B[32m↗ 0.0038\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1095\n", - " │ ├── Epoch N-1 = 0.1101 (\u001B[32m↘ -0.0005\u001B[0m)\n", - " │ └── Best until now = 0.1101 (\u001B[32m↘ -0.0005\u001B[0m)\n", - " ├── Target_iou = 0.9136\n", - " │ ├── Epoch N-1 = 0.9132 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " │ └── Best until now = 0.9132 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " ├── Background_iou = 0.8496\n", - " │ ├── Epoch N-1 = 0.8489 (\u001B[32m↗ 0.0007\u001B[0m)\n", - " │ └── Best until now = 0.8489 (\u001B[32m↗ 0.0007\u001B[0m)\n", - " └── Mean_iou = 0.8816\n", - " ├── Epoch N-1 = 0.881 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " └── Best until now = 0.881 (\u001B[32m↗ 0.0005\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "S1K5MU2kmmDb" + }, + "source": [ + "The logs and the checkpoint for the latest epoch will be kept in your experiment folder.\n", + "\n", + "To start training we'll call train(...) and provide it with the objects we construted above: the model, the training parameters and the data loaders.\n", + "\n", + "**Note:** While training, don't forget to refresh the tensorboard with the arrow on the top right." + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 12: 100%|██████████| 309/309 [01:39<00:00, 3.10it/s, BCEDiceLoss=0.0766, background_IOU=0.913, gpu_mem=1.14, mean_IOU=0.923, target_IOU=0.933]\n", - "Validating epoch 12: 100%|██████████| 65/65 [00:15<00:00, 4.22it/s]\n", - "[2023-11-12 14:28:08] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:28:08] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9138703346252441\n" - ] + "cell_type": "code", + "execution_count": 9, + "metadata": { + "id": "u6roEj9ktFTi", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "2526afe4-98f3-466b-bfc1-133b9ca1d047" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 12:19:47] INFO - sg_trainer.py - Starting a new run with `run_id=RUN_20231113_121947_461500`\n", + "[2023-11-13 12:19:47] INFO - sg_trainer.py - Checkpoints directory: ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500\n", + "[2023-11-13 12:19:47] INFO - sg_trainer.py - Using EMA with params {'decay': 0.9999, 'decay_type': 'exp', 'beta': 15}\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "The console stream is now moved to ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/console_Nov13_12_19_47.txt\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 12:19:49] INFO - sg_trainer_utils.py - TRAINING PARAMETERS:\n", + " - Mode: Single GPU\n", + " - Number of GPUs: 1 (1 available on the machine)\n", + " - Full dataset size: 2477 (len(train_set))\n", + " - Batch size per GPU: 8 (batch_size)\n", + " - Batch Accumulate: 1 (batch_accumulate)\n", + " - Total batch size: 8 (num_gpus * batch_size)\n", + " - Effective Batch size: 8 (num_gpus * batch_size * batch_accumulate)\n", + " - Iterations per epoch: 309 (len(train_loader))\n", + " - Gradient updates per epoch: 309 (len(train_loader) / batch_accumulate)\n", + "\n", + "[2023-11-13 12:19:49] INFO - sg_trainer.py - Started training for 15 epochs (0/14)\n", + "\n", + "Train epoch 0: 100%|██████████| 309/309 [01:55<00:00, 2.69it/s, BCEDiceLoss=0.224, background_IOU=0.746, gpu_mem=1.14, mean_IOU=0.777, target_IOU=0.807]\n", + "Validating: 100%|██████████| 65/65 [00:16<00:00, 4.02it/s]\n", + "[2023-11-13 12:22:01] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:22:01] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.852721631526947\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 0\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.2242\n", + "│ ├── Target_iou = 0.8072\n", + "│ ├── Background_iou = 0.7463\n", + "│ └── Mean_iou = 0.7768\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1872\n", + " ├── Target_iou = 0.8527\n", + " ├── Background_iou = 0.7289\n", + " └── Mean_iou = 0.7908\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 1: 100%|██████████| 309/309 [01:42<00:00, 3.00it/s, BCEDiceLoss=0.167, background_IOU=0.814, gpu_mem=1.14, mean_IOU=0.836, target_IOU=0.859]\n", + "Validating epoch 1: 100%|██████████| 65/65 [00:16<00:00, 3.91it/s]\n", + "[2023-11-13 12:24:04] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:24:04] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.8766682147979736\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 1\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.167\n", + "│ │ ├── Epoch N-1 = 0.2242 (\u001b[32m↘ -0.0572\u001b[0m)\n", + "│ │ └── Best until now = 0.2242 (\u001b[32m↘ -0.0572\u001b[0m)\n", + "│ ├── Target_iou = 0.8588\n", + "│ │ ├── Epoch N-1 = 0.8072 (\u001b[32m↗ 0.0516\u001b[0m)\n", + "│ │ └── Best until now = 0.8072 (\u001b[32m↗ 0.0516\u001b[0m)\n", + "│ ├── Background_iou = 0.8137\n", + "│ │ ├── Epoch N-1 = 0.7463 (\u001b[32m↗ 0.0673\u001b[0m)\n", + "│ │ └── Best until now = 0.7463 (\u001b[32m↗ 0.0673\u001b[0m)\n", + "│ └── Mean_iou = 0.8362\n", + "│ ├── Epoch N-1 = 0.7768 (\u001b[32m↗ 0.0594\u001b[0m)\n", + "│ └── Best until now = 0.7768 (\u001b[32m↗ 0.0594\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.155\n", + " │ ├── Epoch N-1 = 0.1872 (\u001b[32m↘ -0.0321\u001b[0m)\n", + " │ └── Best until now = 0.1872 (\u001b[32m↘ -0.0321\u001b[0m)\n", + " ├── Target_iou = 0.8767\n", + " │ ├── Epoch N-1 = 0.8527 (\u001b[32m↗ 0.0239\u001b[0m)\n", + " │ └── Best until now = 0.8527 (\u001b[32m↗ 0.0239\u001b[0m)\n", + " ├── Background_iou = 0.7812\n", + " │ ├── Epoch N-1 = 0.7289 (\u001b[32m↗ 0.0522\u001b[0m)\n", + " │ └── Best until now = 0.7289 (\u001b[32m↗ 0.0522\u001b[0m)\n", + " └── Mean_iou = 0.8289\n", + " ├── Epoch N-1 = 0.7908 (\u001b[32m↗ 0.0381\u001b[0m)\n", + " └── Best until now = 0.7908 (\u001b[32m↗ 0.0381\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 2: 100%|██████████| 309/309 [01:45<00:00, 2.94it/s, BCEDiceLoss=0.146, background_IOU=0.843, gpu_mem=1.14, mean_IOU=0.86, target_IOU=0.877]\n", + "Validating epoch 2: 100%|██████████| 65/65 [00:14<00:00, 4.37it/s]\n", + "[2023-11-13 12:26:05] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:26:05] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.8859243988990784\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 2\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1456\n", + "│ │ ├── Epoch N-1 = 0.167 (\u001b[32m↘ -0.0213\u001b[0m)\n", + "│ │ └── Best until now = 0.167 (\u001b[32m↘ -0.0213\u001b[0m)\n", + "│ ├── Target_iou = 0.8769\n", + "│ │ ├── Epoch N-1 = 0.8588 (\u001b[32m↗ 0.0181\u001b[0m)\n", + "│ │ └── Best until now = 0.8588 (\u001b[32m↗ 0.0181\u001b[0m)\n", + "│ ├── Background_iou = 0.8434\n", + "│ │ ├── Epoch N-1 = 0.8137 (\u001b[32m↗ 0.0297\u001b[0m)\n", + "│ │ └── Best until now = 0.8137 (\u001b[32m↗ 0.0297\u001b[0m)\n", + "│ └── Mean_iou = 0.8601\n", + "│ ├── Epoch N-1 = 0.8362 (\u001b[32m↗ 0.0239\u001b[0m)\n", + "│ └── Best until now = 0.8362 (\u001b[32m↗ 0.0239\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.145\n", + " │ ├── Epoch N-1 = 0.155 (\u001b[32m↘ -0.01\u001b[0m)\n", + " │ └── Best until now = 0.155 (\u001b[32m↘ -0.01\u001b[0m)\n", + " ├── Target_iou = 0.8859\n", + " │ ├── Epoch N-1 = 0.8767 (\u001b[32m↗ 0.0093\u001b[0m)\n", + " │ └── Best until now = 0.8767 (\u001b[32m↗ 0.0093\u001b[0m)\n", + " ├── Background_iou = 0.7947\n", + " │ ├── Epoch N-1 = 0.7812 (\u001b[32m↗ 0.0135\u001b[0m)\n", + " │ └── Best until now = 0.7812 (\u001b[32m↗ 0.0135\u001b[0m)\n", + " └── Mean_iou = 0.8403\n", + " ├── Epoch N-1 = 0.8289 (\u001b[32m↗ 0.0114\u001b[0m)\n", + " └── Best until now = 0.8289 (\u001b[32m↗ 0.0114\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 3: 100%|██████████| 309/309 [01:43<00:00, 2.98it/s, BCEDiceLoss=0.132, background_IOU=0.856, gpu_mem=1.14, mean_IOU=0.871, target_IOU=0.885]\n", + "Validating epoch 3: 100%|██████████| 65/65 [00:16<00:00, 3.93it/s]\n", + "[2023-11-13 12:28:07] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:28:07] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9014593362808228\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 3\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1318\n", + "│ │ ├── Epoch N-1 = 0.1456 (\u001b[32m↘ -0.0138\u001b[0m)\n", + "│ │ └── Best until now = 0.1456 (\u001b[32m↘ -0.0138\u001b[0m)\n", + "│ ├── Target_iou = 0.8852\n", + "│ │ ├── Epoch N-1 = 0.8769 (\u001b[32m↗ 0.0083\u001b[0m)\n", + "│ │ └── Best until now = 0.8769 (\u001b[32m↗ 0.0083\u001b[0m)\n", + "│ ├── Background_iou = 0.8559\n", + "│ │ ├── Epoch N-1 = 0.8434 (\u001b[32m↗ 0.0125\u001b[0m)\n", + "│ │ └── Best until now = 0.8434 (\u001b[32m↗ 0.0125\u001b[0m)\n", + "│ └── Mean_iou = 0.8706\n", + "│ ├── Epoch N-1 = 0.8601 (\u001b[32m↗ 0.0104\u001b[0m)\n", + "│ └── Best until now = 0.8601 (\u001b[32m↗ 0.0104\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1255\n", + " │ ├── Epoch N-1 = 0.145 (\u001b[32m↘ -0.0195\u001b[0m)\n", + " │ └── Best until now = 0.145 (\u001b[32m↘ -0.0195\u001b[0m)\n", + " ├── Target_iou = 0.9015\n", + " │ ├── Epoch N-1 = 0.8859 (\u001b[32m↗ 0.0155\u001b[0m)\n", + " │ └── Best until now = 0.8859 (\u001b[32m↗ 0.0155\u001b[0m)\n", + " ├── Background_iou = 0.8281\n", + " │ ├── Epoch N-1 = 0.7947 (\u001b[32m↗ 0.0334\u001b[0m)\n", + " │ └── Best until now = 0.7947 (\u001b[32m↗ 0.0334\u001b[0m)\n", + " └── Mean_iou = 0.8648\n", + " ├── Epoch N-1 = 0.8403 (\u001b[32m↗ 0.0245\u001b[0m)\n", + " └── Best until now = 0.8403 (\u001b[32m↗ 0.0245\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 4: 100%|██████████| 309/309 [01:42<00:00, 3.01it/s, BCEDiceLoss=0.122, background_IOU=0.866, gpu_mem=1.14, mean_IOU=0.881, target_IOU=0.896]\n", + "Validating epoch 4: 100%|██████████| 65/65 [00:16<00:00, 3.94it/s]\n", + "[2023-11-13 12:30:07] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:30:07] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9075297117233276\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 4\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1216\n", + "│ │ ├── Epoch N-1 = 0.1318 (\u001b[32m↘ -0.0102\u001b[0m)\n", + "│ │ └── Best until now = 0.1318 (\u001b[32m↘ -0.0102\u001b[0m)\n", + "│ ├── Target_iou = 0.8963\n", + "│ │ ├── Epoch N-1 = 0.8852 (\u001b[32m↗ 0.0111\u001b[0m)\n", + "│ │ └── Best until now = 0.8852 (\u001b[32m↗ 0.0111\u001b[0m)\n", + "│ ├── Background_iou = 0.8663\n", + "│ │ ├── Epoch N-1 = 0.8559 (\u001b[32m↗ 0.0104\u001b[0m)\n", + "│ │ └── Best until now = 0.8559 (\u001b[32m↗ 0.0104\u001b[0m)\n", + "│ └── Mean_iou = 0.8813\n", + "│ ├── Epoch N-1 = 0.8706 (\u001b[32m↗ 0.0107\u001b[0m)\n", + "│ └── Best until now = 0.8706 (\u001b[32m↗ 0.0107\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1177\n", + " │ ├── Epoch N-1 = 0.1255 (\u001b[32m↘ -0.0078\u001b[0m)\n", + " │ └── Best until now = 0.1255 (\u001b[32m↘ -0.0078\u001b[0m)\n", + " ├── Target_iou = 0.9075\n", + " │ ├── Epoch N-1 = 0.9015 (\u001b[32m↗ 0.0061\u001b[0m)\n", + " │ └── Best until now = 0.9015 (\u001b[32m↗ 0.0061\u001b[0m)\n", + " ├── Background_iou = 0.8385\n", + " │ ├── Epoch N-1 = 0.8281 (\u001b[32m↗ 0.0103\u001b[0m)\n", + " │ └── Best until now = 0.8281 (\u001b[32m↗ 0.0103\u001b[0m)\n", + " └── Mean_iou = 0.873\n", + " ├── Epoch N-1 = 0.8648 (\u001b[32m↗ 0.0082\u001b[0m)\n", + " └── Best until now = 0.8648 (\u001b[32m↗ 0.0082\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 5: 100%|██████████| 309/309 [01:44<00:00, 2.94it/s, BCEDiceLoss=0.121, background_IOU=0.865, gpu_mem=1.14, mean_IOU=0.88, target_IOU=0.895]\n", + "Validating epoch 5: 100%|██████████| 65/65 [00:14<00:00, 4.47it/s]\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 5\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1214\n", + "│ │ ├── Epoch N-1 = 0.1216 (\u001b[32m↘ -0.0002\u001b[0m)\n", + "│ │ └── Best until now = 0.1216 (\u001b[32m↘ -0.0002\u001b[0m)\n", + "│ ├── Target_iou = 0.8954\n", + "│ │ ├── Epoch N-1 = 0.8963 (\u001b[31m↘ -0.0009\u001b[0m)\n", + "│ │ └── Best until now = 0.8963 (\u001b[31m↘ -0.0009\u001b[0m)\n", + "│ ├── Background_iou = 0.8655\n", + "│ │ ├── Epoch N-1 = 0.8663 (\u001b[31m↘ -0.0008\u001b[0m)\n", + "│ │ └── Best until now = 0.8663 (\u001b[31m↘ -0.0008\u001b[0m)\n", + "│ └── Mean_iou = 0.8804\n", + "│ ├── Epoch N-1 = 0.8813 (\u001b[31m↘ -0.0008\u001b[0m)\n", + "│ └── Best until now = 0.8813 (\u001b[31m↘ -0.0008\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1194\n", + " │ ├── Epoch N-1 = 0.1177 (\u001b[31m↗ 0.0017\u001b[0m)\n", + " │ └── Best until now = 0.1177 (\u001b[31m↗ 0.0017\u001b[0m)\n", + " ├── Target_iou = 0.9054\n", + " │ ├── Epoch N-1 = 0.9075 (\u001b[31m↘ -0.0021\u001b[0m)\n", + " │ └── Best until now = 0.9075 (\u001b[31m↘ -0.0021\u001b[0m)\n", + " ├── Background_iou = 0.8344\n", + " │ ├── Epoch N-1 = 0.8385 (\u001b[31m↘ -0.004\u001b[0m)\n", + " │ └── Best until now = 0.8385 (\u001b[31m↘ -0.004\u001b[0m)\n", + " └── Mean_iou = 0.8699\n", + " ├── Epoch N-1 = 0.873 (\u001b[31m↘ -0.0031\u001b[0m)\n", + " └── Best until now = 0.873 (\u001b[31m↘ -0.0031\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 6: 100%|██████████| 309/309 [01:45<00:00, 2.93it/s, BCEDiceLoss=0.118, background_IOU=0.87, gpu_mem=1.14, mean_IOU=0.884, target_IOU=0.899]\n", + "Validating epoch 6: 100%|██████████| 65/65 [00:16<00:00, 3.90it/s]\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 6\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1177\n", + "│ │ ├── Epoch N-1 = 0.1214 (\u001b[32m↘ -0.0037\u001b[0m)\n", + "│ │ └── Best until now = 0.1214 (\u001b[32m↘ -0.0037\u001b[0m)\n", + "│ ├── Target_iou = 0.8986\n", + "│ │ ├── Epoch N-1 = 0.8954 (\u001b[32m↗ 0.0032\u001b[0m)\n", + "│ │ └── Best until now = 0.8963 (\u001b[32m↗ 0.0023\u001b[0m)\n", + "│ ├── Background_iou = 0.8702\n", + "│ │ ├── Epoch N-1 = 0.8655 (\u001b[32m↗ 0.0047\u001b[0m)\n", + "│ │ └── Best until now = 0.8663 (\u001b[32m↗ 0.0039\u001b[0m)\n", + "│ └── Mean_iou = 0.8844\n", + "│ ├── Epoch N-1 = 0.8804 (\u001b[32m↗ 0.004\u001b[0m)\n", + "│ └── Best until now = 0.8813 (\u001b[32m↗ 0.0031\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.117\n", + " │ ├── Epoch N-1 = 0.1194 (\u001b[32m↘ -0.0024\u001b[0m)\n", + " │ └── Best until now = 0.1177 (\u001b[32m↘ -0.0007\u001b[0m)\n", + " ├── Target_iou = 0.9073\n", + " │ ├── Epoch N-1 = 0.9054 (\u001b[32m↗ 0.0019\u001b[0m)\n", + " │ └── Best until now = 0.9075 (\u001b[31m↘ -0.0003\u001b[0m)\n", + " ├── Background_iou = 0.8375\n", + " │ ├── Epoch N-1 = 0.8344 (\u001b[32m↗ 0.0031\u001b[0m)\n", + " │ └── Best until now = 0.8385 (\u001b[31m↘ -0.001\u001b[0m)\n", + " └── Mean_iou = 0.8724\n", + " ├── Epoch N-1 = 0.8699 (\u001b[32m↗ 0.0025\u001b[0m)\n", + " └── Best until now = 0.873 (\u001b[31m↘ -0.0006\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 7: 100%|██████████| 309/309 [01:44<00:00, 2.97it/s, BCEDiceLoss=0.11, background_IOU=0.877, gpu_mem=1.14, mean_IOU=0.891, target_IOU=0.905]\n", + "Validating epoch 7: 100%|██████████| 65/65 [00:15<00:00, 4.10it/s]\n", + "[2023-11-13 12:36:14] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:36:14] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9087882041931152\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 7\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.1099\n", + "│ │ ├── Epoch N-1 = 0.1177 (\u001b[32m↘ -0.0078\u001b[0m)\n", + "│ │ └── Best until now = 0.1177 (\u001b[32m↘ -0.0078\u001b[0m)\n", + "│ ├── Target_iou = 0.9051\n", + "│ │ ├── Epoch N-1 = 0.8986 (\u001b[32m↗ 0.0064\u001b[0m)\n", + "│ │ └── Best until now = 0.8986 (\u001b[32m↗ 0.0064\u001b[0m)\n", + "│ ├── Background_iou = 0.8775\n", + "│ │ ├── Epoch N-1 = 0.8702 (\u001b[32m↗ 0.0073\u001b[0m)\n", + "│ │ └── Best until now = 0.8702 (\u001b[32m↗ 0.0073\u001b[0m)\n", + "│ └── Mean_iou = 0.8913\n", + "│ ├── Epoch N-1 = 0.8844 (\u001b[32m↗ 0.0069\u001b[0m)\n", + "│ └── Best until now = 0.8844 (\u001b[32m↗ 0.0069\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1151\n", + " │ ├── Epoch N-1 = 0.117 (\u001b[32m↘ -0.002\u001b[0m)\n", + " │ └── Best until now = 0.117 (\u001b[32m↘ -0.002\u001b[0m)\n", + " ├── Target_iou = 0.9088\n", + " │ ├── Epoch N-1 = 0.9073 (\u001b[32m↗ 0.0015\u001b[0m)\n", + " │ └── Best until now = 0.9075 (\u001b[32m↗ 0.0013\u001b[0m)\n", + " ├── Background_iou = 0.8402\n", + " │ ├── Epoch N-1 = 0.8375 (\u001b[32m↗ 0.0027\u001b[0m)\n", + " │ └── Best until now = 0.8385 (\u001b[32m↗ 0.0017\u001b[0m)\n", + " └── Mean_iou = 0.8745\n", + " ├── Epoch N-1 = 0.8724 (\u001b[32m↗ 0.0021\u001b[0m)\n", + " └── Best until now = 0.873 (\u001b[32m↗ 0.0015\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 8: 100%|██████████| 309/309 [01:44<00:00, 2.97it/s, BCEDiceLoss=0.0972, background_IOU=0.889, gpu_mem=1.14, mean_IOU=0.902, target_IOU=0.915]\n", + "Validating epoch 8: 100%|██████████| 65/65 [00:16<00:00, 3.94it/s]\n", + "[2023-11-13 12:38:18] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:38:18] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9094919562339783\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 8\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0972\n", + "│ │ ├── Epoch N-1 = 0.1099 (\u001b[32m↘ -0.0127\u001b[0m)\n", + "│ │ └── Best until now = 0.1099 (\u001b[32m↘ -0.0127\u001b[0m)\n", + "│ ├── Target_iou = 0.9145\n", + "│ │ ├── Epoch N-1 = 0.9051 (\u001b[32m↗ 0.0095\u001b[0m)\n", + "│ │ └── Best until now = 0.9051 (\u001b[32m↗ 0.0095\u001b[0m)\n", + "│ ├── Background_iou = 0.8892\n", + "│ │ ├── Epoch N-1 = 0.8775 (\u001b[32m↗ 0.0117\u001b[0m)\n", + "│ │ └── Best until now = 0.8775 (\u001b[32m↗ 0.0117\u001b[0m)\n", + "│ └── Mean_iou = 0.9019\n", + "│ ├── Epoch N-1 = 0.8913 (\u001b[32m↗ 0.0106\u001b[0m)\n", + "│ └── Best until now = 0.8913 (\u001b[32m↗ 0.0106\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1142\n", + " │ ├── Epoch N-1 = 0.1151 (\u001b[32m↘ -0.0008\u001b[0m)\n", + " │ └── Best until now = 0.1151 (\u001b[32m↘ -0.0008\u001b[0m)\n", + " ├── Target_iou = 0.9095\n", + " │ ├── Epoch N-1 = 0.9088 (\u001b[32m↗ 0.0007\u001b[0m)\n", + " │ └── Best until now = 0.9088 (\u001b[32m↗ 0.0007\u001b[0m)\n", + " ├── Background_iou = 0.8414\n", + " │ ├── Epoch N-1 = 0.8402 (\u001b[32m↗ 0.0013\u001b[0m)\n", + " │ └── Best until now = 0.8402 (\u001b[32m↗ 0.0013\u001b[0m)\n", + " └── Mean_iou = 0.8755\n", + " ├── Epoch N-1 = 0.8745 (\u001b[32m↗ 0.001\u001b[0m)\n", + " └── Best until now = 0.8745 (\u001b[32m↗ 0.001\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 9: 100%|██████████| 309/309 [01:44<00:00, 2.97it/s, BCEDiceLoss=0.0938, background_IOU=0.896, gpu_mem=1.14, mean_IOU=0.907, target_IOU=0.918]\n", + "Validating epoch 9: 100%|██████████| 65/65 [00:15<00:00, 4.28it/s]\n", + "[2023-11-13 12:40:20] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:40:20] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9101359248161316\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 9\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0938\n", + "│ │ ├── Epoch N-1 = 0.0972 (\u001b[32m↘ -0.0034\u001b[0m)\n", + "│ │ └── Best until now = 0.0972 (\u001b[32m↘ -0.0034\u001b[0m)\n", + "│ ├── Target_iou = 0.9184\n", + "│ │ ├── Epoch N-1 = 0.9145 (\u001b[32m↗ 0.0039\u001b[0m)\n", + "│ │ └── Best until now = 0.9145 (\u001b[32m↗ 0.0039\u001b[0m)\n", + "│ ├── Background_iou = 0.8955\n", + "│ │ ├── Epoch N-1 = 0.8892 (\u001b[32m↗ 0.0063\u001b[0m)\n", + "│ │ └── Best until now = 0.8892 (\u001b[32m↗ 0.0063\u001b[0m)\n", + "│ └── Mean_iou = 0.907\n", + "│ ├── Epoch N-1 = 0.9019 (\u001b[32m↗ 0.0051\u001b[0m)\n", + "│ └── Best until now = 0.9019 (\u001b[32m↗ 0.0051\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1134\n", + " │ ├── Epoch N-1 = 0.1142 (\u001b[32m↘ -0.0008\u001b[0m)\n", + " │ └── Best until now = 0.1142 (\u001b[32m↘ -0.0008\u001b[0m)\n", + " ├── Target_iou = 0.9101\n", + " │ ├── Epoch N-1 = 0.9095 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " │ └── Best until now = 0.9095 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " ├── Background_iou = 0.8427\n", + " │ ├── Epoch N-1 = 0.8414 (\u001b[32m↗ 0.0012\u001b[0m)\n", + " │ └── Best until now = 0.8414 (\u001b[32m↗ 0.0012\u001b[0m)\n", + " └── Mean_iou = 0.8764\n", + " ├── Epoch N-1 = 0.8755 (\u001b[32m↗ 0.0009\u001b[0m)\n", + " └── Best until now = 0.8755 (\u001b[32m↗ 0.0009\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 10: 100%|██████████| 309/309 [01:44<00:00, 2.95it/s, BCEDiceLoss=0.0853, background_IOU=0.902, gpu_mem=1.14, mean_IOU=0.913, target_IOU=0.924]\n", + "Validating epoch 10: 100%|██████████| 65/65 [00:16<00:00, 3.89it/s]\n", + "[2023-11-13 12:42:25] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:42:25] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9106564521789551\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 10\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0853\n", + "│ │ ├── Epoch N-1 = 0.0938 (\u001b[32m↘ -0.0085\u001b[0m)\n", + "│ │ └── Best until now = 0.0938 (\u001b[32m↘ -0.0085\u001b[0m)\n", + "│ ├── Target_iou = 0.9241\n", + "│ │ ├── Epoch N-1 = 0.9184 (\u001b[32m↗ 0.0057\u001b[0m)\n", + "│ │ └── Best until now = 0.9184 (\u001b[32m↗ 0.0057\u001b[0m)\n", + "│ ├── Background_iou = 0.9023\n", + "│ │ ├── Epoch N-1 = 0.8955 (\u001b[32m↗ 0.0068\u001b[0m)\n", + "│ │ └── Best until now = 0.8955 (\u001b[32m↗ 0.0068\u001b[0m)\n", + "│ └── Mean_iou = 0.9132\n", + "│ ├── Epoch N-1 = 0.907 (\u001b[32m↗ 0.0062\u001b[0m)\n", + "│ └── Best until now = 0.907 (\u001b[32m↗ 0.0062\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1127\n", + " │ ├── Epoch N-1 = 0.1134 (\u001b[32m↘ -0.0007\u001b[0m)\n", + " │ └── Best until now = 0.1134 (\u001b[32m↘ -0.0007\u001b[0m)\n", + " ├── Target_iou = 0.9107\n", + " │ ├── Epoch N-1 = 0.9101 (\u001b[32m↗ 0.0005\u001b[0m)\n", + " │ └── Best until now = 0.9101 (\u001b[32m↗ 0.0005\u001b[0m)\n", + " ├── Background_iou = 0.8437\n", + " │ ├── Epoch N-1 = 0.8427 (\u001b[32m↗ 0.001\u001b[0m)\n", + " │ └── Best until now = 0.8427 (\u001b[32m↗ 0.001\u001b[0m)\n", + " └── Mean_iou = 0.8772\n", + " ├── Epoch N-1 = 0.8764 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " └── Best until now = 0.8764 (\u001b[32m↗ 0.0008\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 11: 100%|██████████| 309/309 [01:49<00:00, 2.82it/s, BCEDiceLoss=0.0808, background_IOU=0.908, gpu_mem=1.14, mean_IOU=0.919, target_IOU=0.93]\n", + "Validating epoch 11: 100%|██████████| 65/65 [00:15<00:00, 4.13it/s]\n", + "[2023-11-13 12:44:34] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:44:34] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9110515117645264\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 11\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0808\n", + "│ │ ├── Epoch N-1 = 0.0853 (\u001b[32m↘ -0.0046\u001b[0m)\n", + "│ │ └── Best until now = 0.0853 (\u001b[32m↘ -0.0046\u001b[0m)\n", + "│ ├── Target_iou = 0.9301\n", + "│ │ ├── Epoch N-1 = 0.9241 (\u001b[32m↗ 0.006\u001b[0m)\n", + "│ │ └── Best until now = 0.9241 (\u001b[32m↗ 0.006\u001b[0m)\n", + "│ ├── Background_iou = 0.9075\n", + "│ │ ├── Epoch N-1 = 0.9023 (\u001b[32m↗ 0.0053\u001b[0m)\n", + "│ │ └── Best until now = 0.9023 (\u001b[32m↗ 0.0053\u001b[0m)\n", + "│ └── Mean_iou = 0.9188\n", + "│ ├── Epoch N-1 = 0.9132 (\u001b[32m↗ 0.0056\u001b[0m)\n", + "│ └── Best until now = 0.9132 (\u001b[32m↗ 0.0056\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1121\n", + " │ ├── Epoch N-1 = 0.1127 (\u001b[32m↘ -0.0006\u001b[0m)\n", + " │ └── Best until now = 0.1127 (\u001b[32m↘ -0.0006\u001b[0m)\n", + " ├── Target_iou = 0.9111\n", + " │ ├── Epoch N-1 = 0.9107 (\u001b[32m↗ 0.0004\u001b[0m)\n", + " │ └── Best until now = 0.9107 (\u001b[32m↗ 0.0004\u001b[0m)\n", + " ├── Background_iou = 0.8445\n", + " │ ├── Epoch N-1 = 0.8437 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " │ └── Best until now = 0.8437 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " └── Mean_iou = 0.8778\n", + " ├── Epoch N-1 = 0.8772 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " └── Best until now = 0.8772 (\u001b[32m↗ 0.0006\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 12: 100%|██████████| 309/309 [01:44<00:00, 2.97it/s, BCEDiceLoss=0.0779, background_IOU=0.911, gpu_mem=1.14, mean_IOU=0.921, target_IOU=0.932]\n", + "Validating epoch 12: 100%|██████████| 65/65 [00:16<00:00, 3.92it/s]\n", + "[2023-11-13 12:46:40] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:46:40] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9114375114440918\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 12\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0779\n", + "│ │ ├── Epoch N-1 = 0.0808 (\u001b[32m↘ -0.0029\u001b[0m)\n", + "│ │ └── Best until now = 0.0808 (\u001b[32m↘ -0.0029\u001b[0m)\n", + "│ ├── Target_iou = 0.9317\n", + "│ │ ├── Epoch N-1 = 0.9301 (\u001b[32m↗ 0.0016\u001b[0m)\n", + "│ │ └── Best until now = 0.9301 (\u001b[32m↗ 0.0016\u001b[0m)\n", + "│ ├── Background_iou = 0.9113\n", + "│ │ ├── Epoch N-1 = 0.9075 (\u001b[32m↗ 0.0038\u001b[0m)\n", + "│ │ └── Best until now = 0.9075 (\u001b[32m↗ 0.0038\u001b[0m)\n", + "│ └── Mean_iou = 0.9215\n", + "│ ├── Epoch N-1 = 0.9188 (\u001b[32m↗ 0.0027\u001b[0m)\n", + "│ └── Best until now = 0.9188 (\u001b[32m↗ 0.0027\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1115\n", + " │ ├── Epoch N-1 = 0.1121 (\u001b[32m↘ -0.0006\u001b[0m)\n", + " │ └── Best until now = 0.1121 (\u001b[32m↘ -0.0006\u001b[0m)\n", + " ├── Target_iou = 0.9114\n", + " │ ├── Epoch N-1 = 0.9111 (\u001b[32m↗ 0.0004\u001b[0m)\n", + " │ └── Best until now = 0.9111 (\u001b[32m↗ 0.0004\u001b[0m)\n", + " ├── Background_iou = 0.8453\n", + " │ ├── Epoch N-1 = 0.8445 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " │ └── Best until now = 0.8445 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " └── Mean_iou = 0.8784\n", + " ├── Epoch N-1 = 0.8778 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " └── Best until now = 0.8778 (\u001b[32m↗ 0.0006\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 13: 100%|██████████| 309/309 [01:50<00:00, 2.79it/s, BCEDiceLoss=0.0748, background_IOU=0.916, gpu_mem=1.14, mean_IOU=0.926, target_IOU=0.935]\n", + "Validating epoch 13: 100%|██████████| 65/65 [00:16<00:00, 3.97it/s]\n", + "[2023-11-13 12:48:53] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:48:53] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9118204712867737\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 13\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0748\n", + "│ │ ├── Epoch N-1 = 0.0779 (\u001b[32m↘ -0.0031\u001b[0m)\n", + "│ │ └── Best until now = 0.0779 (\u001b[32m↘ -0.0031\u001b[0m)\n", + "│ ├── Target_iou = 0.9349\n", + "│ │ ├── Epoch N-1 = 0.9317 (\u001b[32m↗ 0.0032\u001b[0m)\n", + "│ │ └── Best until now = 0.9317 (\u001b[32m↗ 0.0032\u001b[0m)\n", + "│ ├── Background_iou = 0.9165\n", + "│ │ ├── Epoch N-1 = 0.9113 (\u001b[32m↗ 0.0052\u001b[0m)\n", + "│ │ └── Best until now = 0.9113 (\u001b[32m↗ 0.0052\u001b[0m)\n", + "│ └── Mean_iou = 0.9257\n", + "│ ├── Epoch N-1 = 0.9215 (\u001b[32m↗ 0.0042\u001b[0m)\n", + "│ └── Best until now = 0.9215 (\u001b[32m↗ 0.0042\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.111\n", + " │ ├── Epoch N-1 = 0.1115 (\u001b[32m↘ -0.0006\u001b[0m)\n", + " │ └── Best until now = 0.1115 (\u001b[32m↘ -0.0006\u001b[0m)\n", + " ├── Target_iou = 0.9118\n", + " │ ├── Epoch N-1 = 0.9114 (\u001b[32m↗ 0.0004\u001b[0m)\n", + " │ └── Best until now = 0.9114 (\u001b[32m↗ 0.0004\u001b[0m)\n", + " ├── Background_iou = 0.8461\n", + " │ ├── Epoch N-1 = 0.8453 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " │ └── Best until now = 0.8453 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " └── Mean_iou = 0.879\n", + " ├── Epoch N-1 = 0.8784 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " └── Best until now = 0.8784 (\u001b[32m↗ 0.0006\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "Train epoch 14: 100%|██████████| 309/309 [01:45<00:00, 2.94it/s, BCEDiceLoss=0.0742, background_IOU=0.915, gpu_mem=1.14, mean_IOU=0.925, target_IOU=0.934]\n", + "Validating epoch 14: 100%|██████████| 65/65 [00:16<00:00, 3.89it/s]\n", + "[2023-11-13 12:51:00] INFO - base_sg_logger.py - Checkpoint saved in ./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500/ckpt_best.pth\n", + "[2023-11-13 12:51:00] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9122186303138733\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "===========================================================\n", + "SUMMARY OF EPOCH 14\n", + "├── Train\n", + "│ ├── Bcediceloss = 0.0742\n", + "│ │ ├── Epoch N-1 = 0.0748 (\u001b[32m↘ -0.0006\u001b[0m)\n", + "│ │ └── Best until now = 0.0748 (\u001b[32m↘ -0.0006\u001b[0m)\n", + "│ ├── Target_iou = 0.9343\n", + "│ │ ├── Epoch N-1 = 0.9349 (\u001b[31m↘ -0.0006\u001b[0m)\n", + "│ │ └── Best until now = 0.9349 (\u001b[31m↘ -0.0006\u001b[0m)\n", + "│ ├── Background_iou = 0.9147\n", + "│ │ ├── Epoch N-1 = 0.9165 (\u001b[31m↘ -0.0017\u001b[0m)\n", + "│ │ └── Best until now = 0.9165 (\u001b[31m↘ -0.0017\u001b[0m)\n", + "│ └── Mean_iou = 0.9245\n", + "│ ├── Epoch N-1 = 0.9257 (\u001b[31m↘ -0.0012\u001b[0m)\n", + "│ └── Best until now = 0.9257 (\u001b[31m↘ -0.0012\u001b[0m)\n", + "└── Validation\n", + " ├── Bcediceloss = 0.1104\n", + " │ ├── Epoch N-1 = 0.111 (\u001b[32m↘ -0.0005\u001b[0m)\n", + " │ └── Best until now = 0.111 (\u001b[32m↘ -0.0005\u001b[0m)\n", + " ├── Target_iou = 0.9122\n", + " │ ├── Epoch N-1 = 0.9118 (\u001b[32m↗ 0.0004\u001b[0m)\n", + " │ └── Best until now = 0.9118 (\u001b[32m↗ 0.0004\u001b[0m)\n", + " ├── Background_iou = 0.8469\n", + " │ ├── Epoch N-1 = 0.8461 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " │ └── Best until now = 0.8461 (\u001b[32m↗ 0.0008\u001b[0m)\n", + " └── Mean_iou = 0.8796\n", + " ├── Epoch N-1 = 0.879 (\u001b[32m↗ 0.0006\u001b[0m)\n", + " └── Best until now = 0.879 (\u001b[32m↗ 0.0006\u001b[0m)\n", + "\n", + "===========================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "[2023-11-13 12:51:04] INFO - sg_trainer.py - RUNNING ADDITIONAL TEST ON THE AVERAGED MODEL...\n", + "Validating epoch 15: 97%|█████████▋| 63/65 [00:17<00:00, 5.89it/s]" + ] + } + ], + "source": [ + "trainer.train(model=model, training_params=train_params, train_loader=train_loader, valid_loader=valid_loader)" + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 12\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.0766\n", - "│ │ ├── Epoch N-1 = 0.0809 (\u001B[32m↘ -0.0042\u001B[0m)\n", - "│ │ └── Best until now = 0.0809 (\u001B[32m↘ -0.0042\u001B[0m)\n", - "│ ├── Target_iou = 0.9327\n", - "│ │ ├── Epoch N-1 = 0.9295 (\u001B[32m↗ 0.0032\u001B[0m)\n", - "│ │ └── Best until now = 0.9295 (\u001B[32m↗ 0.0032\u001B[0m)\n", - "│ ├── Background_iou = 0.9127\n", - "│ │ ├── Epoch N-1 = 0.9068 (\u001B[32m↗ 0.0059\u001B[0m)\n", - "│ │ └── Best until now = 0.9068 (\u001B[32m↗ 0.0059\u001B[0m)\n", - "│ └── Mean_iou = 0.9227\n", - "│ ├── Epoch N-1 = 0.9182 (\u001B[32m↗ 0.0045\u001B[0m)\n", - "│ └── Best until now = 0.9182 (\u001B[32m↗ 0.0045\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1091\n", - " │ ├── Epoch N-1 = 0.1095 (\u001B[32m↘ -0.0005\u001B[0m)\n", - " │ └── Best until now = 0.1095 (\u001B[32m↘ -0.0005\u001B[0m)\n", - " ├── Target_iou = 0.9139\n", - " │ ├── Epoch N-1 = 0.9136 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " │ └── Best until now = 0.9136 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " ├── Background_iou = 0.8503\n", - " │ ├── Epoch N-1 = 0.8496 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " │ └── Best until now = 0.8496 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " └── Mean_iou = 0.8821\n", - " ├── Epoch N-1 = 0.8816 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " └── Best until now = 0.8816 (\u001B[32m↗ 0.0005\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "code", + "execution_count": 10, + "metadata": { + "id": "X8BJq1crcbjl" + }, + "outputs": [], + "source": [ + "print(\"Best Checkpoint mIoU is: \"+ str(trainer.best_metric))" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 13: 100%|██████████| 309/309 [01:42<00:00, 3.02it/s, BCEDiceLoss=0.0756, background_IOU=0.915, gpu_mem=1.14, mean_IOU=0.924, target_IOU=0.934]\n", - "Validating epoch 13: 100%|██████████| 65/65 [00:15<00:00, 4.18it/s]\n", - "[2023-11-12 14:30:11] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:30:11] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9141221642494202\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "3Nybj15cchxd" + }, + "source": [ + "Now you can download your trained weights from this directory" + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 13\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.0756\n", - "│ │ ├── Epoch N-1 = 0.0766 (\u001B[32m↘ -0.0011\u001B[0m)\n", - "│ │ └── Best until now = 0.0766 (\u001B[32m↘ -0.0011\u001B[0m)\n", - "│ ├── Target_iou = 0.9336\n", - "│ │ ├── Epoch N-1 = 0.9327 (\u001B[32m↗ 0.0009\u001B[0m)\n", - "│ │ └── Best until now = 0.9327 (\u001B[32m↗ 0.0009\u001B[0m)\n", - "│ ├── Background_iou = 0.9148\n", - "│ │ ├── Epoch N-1 = 0.9127 (\u001B[32m↗ 0.0021\u001B[0m)\n", - "│ │ └── Best until now = 0.9127 (\u001B[32m↗ 0.0021\u001B[0m)\n", - "│ └── Mean_iou = 0.9242\n", - "│ ├── Epoch N-1 = 0.9227 (\u001B[32m↗ 0.0015\u001B[0m)\n", - "│ └── Best until now = 0.9227 (\u001B[32m↗ 0.0015\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1086\n", - " │ ├── Epoch N-1 = 0.1091 (\u001B[32m↘ -0.0004\u001B[0m)\n", - " │ └── Best until now = 0.1091 (\u001B[32m↘ -0.0004\u001B[0m)\n", - " ├── Target_iou = 0.9141\n", - " │ ├── Epoch N-1 = 0.9139 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " │ └── Best until now = 0.9139 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " ├── Background_iou = 0.8508\n", - " │ ├── Epoch N-1 = 0.8503 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " │ └── Best until now = 0.8503 (\u001B[32m↗ 0.0005\u001B[0m)\n", - " └── Mean_iou = 0.8825\n", - " ├── Epoch N-1 = 0.8821 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " └── Best until now = 0.8821 (\u001B[32m↗ 0.0004\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "code", + "execution_count": 11, + "metadata": { + "id": "_iHsFgPSciQh" + }, + "outputs": [], + "source": [ + "print(trainer.checkpoints_dir_path)" + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "Train epoch 14: 100%|██████████| 309/309 [01:39<00:00, 3.12it/s, BCEDiceLoss=0.0731, background_IOU=0.916, gpu_mem=1.14, mean_IOU=0.925, target_IOU=0.935]\n", - "Validating epoch 14: 100%|██████████| 65/65 [00:14<00:00, 4.53it/s]\n", - "[2023-11-12 14:32:07] INFO - base_sg_logger.py - Checkpoint saved in /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth\n", - "[2023-11-12 14:32:07] INFO - sg_trainer.py - Best checkpoint overriden: validation target_IOU: 0.9143829941749573\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "yuhYeXLA18q5" + }, + "source": [ + "# 6. Predict\n" + ] }, { - "output_type": "stream", - "name": "stdout", - "text": [ - "===========================================================\n", - "SUMMARY OF EPOCH 14\n", - "├── Train\n", - "│ ├── Bcediceloss = 0.0731\n", - "│ │ ├── Epoch N-1 = 0.0756 (\u001B[32m↘ -0.0025\u001B[0m)\n", - "│ │ └── Best until now = 0.0756 (\u001B[32m↘ -0.0025\u001B[0m)\n", - "│ ├── Target_iou = 0.9352\n", - "│ │ ├── Epoch N-1 = 0.9336 (\u001B[32m↗ 0.0016\u001B[0m)\n", - "│ │ └── Best until now = 0.9336 (\u001B[32m↗ 0.0016\u001B[0m)\n", - "│ ├── Background_iou = 0.9158\n", - "│ │ ├── Epoch N-1 = 0.9148 (\u001B[32m↗ 0.001\u001B[0m)\n", - "│ │ └── Best until now = 0.9148 (\u001B[32m↗ 0.001\u001B[0m)\n", - "│ └── Mean_iou = 0.9255\n", - "│ ├── Epoch N-1 = 0.9242 (\u001B[32m↗ 0.0013\u001B[0m)\n", - "│ └── Best until now = 0.9242 (\u001B[32m↗ 0.0013\u001B[0m)\n", - "└── Validation\n", - " ├── Bcediceloss = 0.1082\n", - " │ ├── Epoch N-1 = 0.1086 (\u001B[32m↘ -0.0004\u001B[0m)\n", - " │ └── Best until now = 0.1086 (\u001B[32m↘ -0.0004\u001B[0m)\n", - " ├── Target_iou = 0.9144\n", - " │ ├── Epoch N-1 = 0.9141 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " │ └── Best until now = 0.9141 (\u001B[32m↗ 0.0003\u001B[0m)\n", - " ├── Background_iou = 0.8514\n", - " │ ├── Epoch N-1 = 0.8508 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " │ └── Best until now = 0.8508 (\u001B[32m↗ 0.0006\u001B[0m)\n", - " └── Mean_iou = 0.8829\n", - " ├── Epoch N-1 = 0.8825 (\u001B[32m↗ 0.0004\u001B[0m)\n", - " └── Best until now = 0.8825 (\u001B[32m↗ 0.0004\u001B[0m)\n", - "\n", - "===========================================================\n" - ] + "cell_type": "markdown", + "metadata": { + "id": "VjRA1tu1mvXQ" + }, + "source": [ + "When the training is complete you can use the trained model to get predictions on the validation set, your data or some other image. Let's load some image and\n", + "run a model inference to create a binary segmentation mask." + ] }, { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-12 14:32:09] INFO - sg_trainer.py - RUNNING ADDITIONAL TEST ON THE AVERAGED MODEL...\n", - "Validating epoch 15: 98%|█████████▊| 64/65 [00:13<00:00, 3.64it/s]" - ] - } - ], - "source": [ - "trainer.train(model=model, training_params=train_params, train_loader=train_loader, valid_loader=valid_loader)" - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "metadata": { - "id": "X8BJq1crcbjl", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "3baac32f-83fe-4e0a-d3d8-d75ac230caa5" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Best Checkpoint mIoU is: 0.9143829941749573\n" - ] - } - ], - "source": [ - "print(\"Best Checkpoint mIoU is: \"+ str(trainer.best_metric))" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "3Nybj15cchxd" - }, - "source": [ - "Now you can download your trained weights from this directory" - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "metadata": { - "id": "_iHsFgPSciQh", - "colab": { - "base_uri": "https://localhost:8080/" + "cell_type": "code", + "execution_count": 12, + "metadata": { + "id": "Ads7RyGN2JwQ", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "outputId": "a9fc3231-0de4-49bb-863c-6c5765381cae" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "\rValidating epoch 15: 100%|██████████| 65/65 [00:18<00:00, 6.54it/s]\rValidating epoch 15: 100%|██████████| 65/65 [00:18<00:00, 3.56it/s]\n", + "[2023-11-13 12:51:23] INFO - base_sg_logger.py - [CLEANUP] - Successfully stopped system monitoring process\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Best Checkpoint mIoU is: 0.9122186303138733\n", + "./notebook_ckpts/segmentation_transfer_learning/RUN_20231113_121947_461500\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "from torchvision.transforms import Compose, ToTensor, Resize, Normalize, ToPILImage\n", + "\n", + "# Initiate a model with best checkpoint.\n", + "model = models.get(model_name=Models.PP_LITE_T_SEG75,\n", + " arch_params={\"use_aux_heads\": False},\n", + " num_classes=1,\n", + " checkpoint_path=os.path.join(trainer.checkpoints_dir_path, \"ckpt_best.pth\")).cuda().eval()\n", + "\n", + "pre_proccess = Compose([\n", + " ToTensor(),\n", + " Normalize([.485, .456, .406], [.229, .224, .225])\n", + "])\n", + "\n", + "demo_img_path = os.path.join(root_dir, \"images\", \"ache-adult-depression-expression-41253.png\")\n", + "\n", + "img = Image.open(demo_img_path)\n", + "# Resize the image and display\n", + "img = Resize(size=(480, 320))(img)\n", + "display(img)\n", + "\n", + "# Run pre-proccess - transforms to tensor and apply normalizations.\n", + "img_inp = pre_proccess(img).unsqueeze(0).cuda()\n", + "\n", + "# Run inference\n", + "mask = model(img_inp)\n", + "\n", + "# Run post-proccess - apply sigmoid to output probabilities, then apply hard\n", + "# threshold of 0.5 for binary mask prediction.\n", + "mask = torch.sigmoid(mask).gt(0.5).squeeze()\n", + "mask = ToPILImage()(mask.float())\n", + "display(mask)\n" + ] }, - "outputId": "919b2142-4b98-4724-dcac-c5e6744d80d2" - }, - "outputs": [ { - "output_type": "stream", - "name": "stdout", - "text": [ - "/home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664\n" - ] - } - ], - "source": [ - "print(trainer.checkpoints_dir_path)" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "yuhYeXLA18q5" - }, - "source": [ - "# 6. Predict\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "VjRA1tu1mvXQ" - }, - "source": [ - "When the training is complete you can use the trained model to get predictions on the validation set, your data or some other image. Let's load some image and\n", - "run a model inference to create a binary segmentation mask." - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": { - "id": "Ads7RyGN2JwQ", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 1000 + "cell_type": "markdown", + "metadata": { + "id": "-k6ZLKHL1hIM" + }, + "source": [ + "# 7. Convert to ONNX/TensorRT" + ] }, - "outputId": "a40cb318-010e-4a52-dba8-2c931fd773fb" - }, - "outputs": [ { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "image/png": "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\n" - }, - "metadata": {} + "cell_type": "markdown", + "metadata": { + "id": "br7n55Szm4Nq" + }, + "source": [ + "Let's compile our model to ONNX." + ] }, { - "output_type": "display_data", - "data": { - "text/plain": [ - "" + "cell_type": "code", + "execution_count": 14, + "metadata": { + "id": "q0AGQvEf11PT", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "249acd62-694c-460b-adbf-dbdd3d86057e" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "ONNX successfully created at: /content/data/model.onnx\n" + ] + } ], - "image/png": "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\n" - }, - "metadata": {} - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "[2023-11-12 14:33:33] INFO - checkpoint_utils.py - Successfully loaded model weights from /home/notebook_ckpts/segmentation_transfer_learning/RUN_20231112_140121_753664/ckpt_best.pth EMA checkpoint.\n" - ] + "source": [ + "from onnxsim import simplify\n", + "import onnx\n", + "\n", + "onnx_path = os.path.join(os.getcwd(), \"model.onnx\")\n", + "\n", + "input_size = [1, 3, 480, 320]\n", + "model.prep_model_for_conversion(input_size=input_size)\n", + "\n", + "torch.onnx.export(model,\n", + " torch.randn(*input_size).cuda(),\n", + " onnx_path)\n", + "\n", + "# onnx simplifier\n", + "model_sim, check = simplify(onnx_path)\n", + "assert check, \"Simplified ONNX model could not be validated\"\n", + "onnx.save_model(model_sim, onnx_path)\n", + "\n", + "print(\"ONNX successfully created at: \", onnx_path)\n" + ] } - ], - "source": [ - "from torchvision.transforms import Compose, ToTensor, Resize, Normalize, ToPILImage\n", - "\n", - "# Initiate a model with best checkpoint.\n", - "model = models.get(model_name=Models.PP_LITE_T_SEG75,\n", - " arch_params={\"use_aux_heads\": False},\n", - " num_classes=1,\n", - " checkpoint_path=os.path.join(trainer.checkpoints_dir_path, \"ckpt_best.pth\")).cuda().eval()\n", - "\n", - "pre_proccess = Compose([\n", - " ToTensor(),\n", - " Normalize([.485, .456, .406], [.229, .224, .225])\n", - "])\n", - "\n", - "demo_img_path = \"/home/data/supervisely-persons/images/ache-adult-depression-expression-41253.png\"\n", - "img = Image.open(demo_img_path)\n", - "# Resize the image and display\n", - "img = Resize(size=(480, 320))(img)\n", - "display(img)\n", - "\n", - "# Run pre-proccess - transforms to tensor and apply normalizations.\n", - "img_inp = pre_proccess(img).unsqueeze(0).cuda()\n", - "\n", - "# Run inference\n", - "mask = model(img_inp)\n", - "\n", - "# Run post-proccess - apply sigmoid to output probabilities, then apply hard\n", - "# threshold of 0.5 for binary mask prediction.\n", - "mask = torch.sigmoid(mask).gt(0.5).squeeze()\n", - "mask = ToPILImage()(mask.float())\n", - "display(mask)\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "-k6ZLKHL1hIM" - }, - "source": [ - "# 7. Convert to ONNX/TensorRT" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "br7n55Szm4Nq" - }, - "source": [ - "Let's compile our model to ONNX." - ] - }, - { - "cell_type": "code", - "execution_count": 14, - "metadata": { - "id": "q0AGQvEf11PT", + ], + "metadata": { + "accelerator": "GPU", "colab": { - "base_uri": "https://localhost:8080/" + "provenance": [] }, - "outputId": "8916ce5d-c1a4-447f-d7b7-472e6e18fe71" - }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "ONNX successfully created at: /home/data/model.onnx\n" - ] + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" } - ], - "source": [ - "from onnxsim import simplify\n", - "import onnx\n", - "\n", - "input_size = [1, 3, 480, 320]\n", - "onnx_path = \"/home/data/model.onnx\"\n", - "\n", - "model.prep_model_for_conversion(input_size=input_size)\n", - "\n", - "torch.onnx.export(model,\n", - " torch.randn(*input_size).cuda(),\n", - " onnx_path,\n", - " opset_version=11)\n", - "\n", - "# onnx simplifier\n", - "model_sim, check = simplify(onnx_path)\n", - "assert check, \"Simplified ONNX model could not be validated\"\n", - "onnx.save_model(model_sim, onnx_path)\n", - "\n", - "print(\"ONNX successfully created at: \", onnx_path)\n" - ] - } - ], - "metadata": { - "accelerator": "GPU", - "colab": { - "provenance": [] - }, - "kernelspec": { - "display_name": "Python 3", - "name": "python3" }, - "language_info": { - "name": "python" - } - }, - "nbformat": 4, - "nbformat_minor": 0 + "nbformat": 4, + "nbformat_minor": 0 }