fixed mnist paths + several small stuff (#48)

* fixed mnist paths + several small stuff

.github/CODEOWNERS

@@ -1,11 +1,2 @@
-# Require MaximilienLC to review all pull requests (for now).
-# Comment out when things get going.
-* @maximilienlc
-
 # Require MaximilienLC to review all changes to .github/ files.
-# Probably keep this for a while.
 .github/ @maximilienlc
-
-# Require MaximilienLC to review all changes to python test files.
-# Probably keep this for a while.
-*_test.py @maximilienlc

.github/workflows/format-lint.yaml

@@ -1,5 +1,5 @@
 # Runs the pre-commit workflow (formatting & linting)
-name: Format & Lint on PR
+name: Format & Lint
 
 on:
   pull_request:

.github/workflows/typecheck-unittest.yaml

@@ -6,11 +6,15 @@ on:
     branches: [main]
     paths:
       - cneuromax/**
+      - pyreqs/**
+      - pyproject.toml
       - .github/workflows/typecheck-unittest.yaml
   push:
     branches: [main]
     paths:
       - cneuromax/**
+      - pyreqs/**
+      - pyproject.toml
       - .github/workflows/typecheck-unittest.yaml
 
 permissions:

README.md

@@ -1,4 +1,4 @@
-# CNeuroMAX
+# CNeuroMax
 
 [![container-build-push](
     https://github.com/courtois-neuromod/cneuromax/actions/workflows/container-build-push.yaml/badge.svg)](
@@ -27,9 +27,16 @@ Full documentation available at [https://courtois-neuromod.github.io/cneuromax](
 cneuromax/
 ├─ .github/                     <-- Contains GitHub automation (tests, containers, etc) config files
 │  └─ *
-├─ cneuromax/                    <-- Contains all of the Machine Learning code and config files
+├─ cneuromax/                   <-- Contains all of the Machine Learning code and config files
 │  ├─ application/              <-- Contains the code to create applications (cozmo inference, etc)
 │  │  └─ *
+│  ├─ common/                   <-- Contains the code common to apps, deep learning and neuroevolution
+│  │  ├─ launcher               <-- Contains the base configs to launch jobs
+│  │  |  └─ *
+|  |  ├─ sweeper                <-- Contains the base configs to run hyperparameter sweeps
+│  │  |  └─ *
+│  │  ├─ fitter.yaml            <-- The config common to all fitting jobs
+|  |  └─ *
 │  ├─ deeplearning/             <-- Contains the Deep Learning code
 │  │  ├─ common/                <-- Contains the code common to more than one DL experiment
 │  │  │  ├─ datamodule/         <-- Contains common Lightning DataModules
@@ -38,17 +45,26 @@ cneuromax/
 │  │  │  ├─ litmodule/          <-- Contains common Lightning Modules
 │  │  │  │  ├─ base.py          <-- The base Lightning Module to build upon
 │  │  │  │  └─ *
-│  │  │  └─ nnmodule/           <-- Contains common PyTorch Modules
-│  │  │     └─ *
-│  │  ├─ experiment/           <-- Contains the DL experiments
-│  │  │  ├─ my_new_experiment/  <-- ! Your new DL experiment folder
+│  │  │  ├─ logger/             <-- Contains common Lightning Logger configs
+│  │  │  │  └─ *
+│  │  │  ├─ lrscheduler/        <-- Contains common PyTorch Learning Rate Scheduler configs
+│  │  │  │  └─ *
+│  │  │  ├─ nnmodule/           <-- Contains common PyTorch Modules
+│  │  │  │  └─ *
+│  │  │  ├─ optimizer/          <-- Contains common PyTorch Optimizer configs
+│  │  │  │  └─ *
+│  │  │  ├─ trainer/            <-- Contains common Lightning Trainer configs
+│  │  │  │  └─ *
+|  |  |  └─ *
+│  │  ├─ experiment/            <-- Contains the Deep Learning experiments
+│  │  │  ├─ my_new_experiment/  <-- ! Your new Deep Learning experiment folder
 │  │  │  │  ├─ datamodule.py    <-- ! Your Lightning DataModule
 │  │  │  │  ├─ litmodule.py     <-- ! Your Lightning Module
 │  │  │  │  ├─ nnmodule.py      <-- ! Your PyTorch Module
-│  │  │  │  ├─ config.yaml      <-- ! Your Hydra configuration file
+│  │  │  │  ├─ test_0.yaml      <-- ! One of your Hydra configuration file
 │  │  │  │  └─ *
 │  │  │  └─ *
-│  │  ├─ fitter.py              <-- Contains the code to fit (train/val/test) a DL model
+│  │  ├─ __main__.py            <-- Main file to launch Deep Learning experiments
 │  │  └─ *
 │  └─ neuroevolution/           <-- Contains the code for Neuroevolution models
 │     └─ *

cneuromax/common/fitter.yaml

@@ -1,7 +1,3 @@
-hydra:
-  job:
-    chdir: true
-
 data_dir: ./data
 device: cpu
 load_path: null

cneuromax/deeplearning/__main__.py

@@ -10,7 +10,7 @@
 
 @hydra.main(
     version_base=None,
-    config_name="config",
+    config_name="dlfitter",
     config_path="common",
 )
 def run(config: DictConfig) -> float:

cneuromax/deeplearning/experiment/mnist_classification/mlp.yaml

@@ -6,19 +6,21 @@ defaults:
 
 datamodule:
   _target_:
-    cneuromax.deeplearning.experiment.mnist_classification.modules.MNISTClassificationDataModule
+    cneuromax.deeplearning.experiment.mnist_classification.MNISTClassificationDataModule
   config:
     _target_:
-      cneuromax.deeplearning.experiment.mnist_classification.modules.MNISTClassificationDataModuleConfig
+      cneuromax.deeplearning.experiment.mnist_classification.MNISTClassificationDataModuleConfig
     val_percentage: 0.1
 
 litmodule:
   _target_:
-    cneuromax.deeplearning.experiment.mnist_classification.modules.MNISTClassificationLitModule
+    cneuromax.deeplearning.experiment.mnist_classification.MNISTClassificationLitModule
   nnmodule:
     config:
       dims: [784, 128, 10]
 
 logger:
   name: mlp
   project: mnist_classification
+
+save_path: lightning_ckpt