From a67ffd630bd58da4f8b9b13c99baca481cbc92d2 Mon Sep 17 00:00:00 2001
From: Sanjana Garg <garg.sanjana.95@gmail.com>
Date: Mon, 10 Apr 2023 20:34:22 -0500
Subject: [PATCH] Added huggingface and torchvision notebooks with outputs

---
 huggingface_dataset.ipynb | 898 ++++++++++++++++++++++++++++++++++++++
 torchvision_dataset.ipynb | 869 ++++++++++++++++++++++++++++++++++++
 2 files changed, 1767 insertions(+)
 create mode 100644 huggingface_dataset.ipynb
 create mode 100644 torchvision_dataset.ipynb

diff --git a/huggingface_dataset.ipynb b/huggingface_dataset.ipynb
new file mode 100644
index 0000000..df4ce2e
--- /dev/null
+++ b/huggingface_dataset.ipynb
@@ -0,0 +1,898 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Run CleanVision on a Hugging Face dataset"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/cleanlab/cleanvision/blob/main/examples/huggingface_dataset.ipynb) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!pip install -U pip\n",
+    "!pip install cleanvision[huggingface]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "**After you install these packages, you may need to restart your notebook runtime before running the rest of this notebook.**"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from datasets import load_dataset, concatenate_datasets\n",
+    "from cleanvision.imagelab import Imagelab"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Download dataset and concatenate all splits\n",
+    "\n",
+    "Since we're interested in generally understanding what issues plague our data, we merge the training and test sets into one larger dataset before running CleanVision. You could alternatively just run the package on these two sets of data separately to obtain two different reports.\n",
+    "\n",
+    "[CIFAR10](https://www.cs.toronto.edu/~kriz/cifar.html) is classification dataset, but CleanVision can be used to audit images from any type of dataset (including supervised or unsupervised learning).\n",
+    "\n",
+    "Load all splits of the CIFAR10 dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Found cached dataset cifar10 (/Users/sanjana/.cache/huggingface/datasets/cifar10/plain_text/1.0.0/447d6ec4733dddd1ce3bb577c7166b986eaa4c538dcd9e805ba61f35674a9de4)\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "83d6c575e3724d5484d6db3605b89b23",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "dataset_dict = load_dataset(\"cifar10\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "See more information on the dataset like features, number of examples in each split"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "DatasetDict({\n",
+       "    train: Dataset({\n",
+       "        features: ['img', 'label'],\n",
+       "        num_rows: 50000\n",
+       "    })\n",
+       "    test: Dataset({\n",
+       "        features: ['img', 'label'],\n",
+       "        num_rows: 10000\n",
+       "    })\n",
+       "})"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset_dict"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Concatenate train and test splits"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dataset = concatenate_datasets([d for d in dataset_dict.values()])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Dataset after concatenating"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Dataset({\n",
+       "    features: ['img', 'label'],\n",
+       "    num_rows: 60000\n",
+       "})"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "tags": []
+   },
+   "source": [
+    "`dataset.features` is a `dict[column_name, column_type]` that contains information about the different columns in the dataset and the type of each column. Use `dataset.features` to find the key that contains the Image feature."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'img': Image(decode=True, id=None),\n",
+       " 'label': ClassLabel(names=['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck'], id=None)}"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset.features"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Let's look at the first image in this dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<PIL.PngImagePlugin.PngImageFile image mode=RGB size=32x32>"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset[0][\"img\"]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Run CleanVision"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Checking for dark, light, odd_aspect_ratio, low_information, exact_duplicates, near_duplicates, blurry, grayscale images ...\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 60000/60000 [00:46<00:00, 1283.96it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 60000/60000 [00:41<00:00, 1444.52it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Issue checks completed. To see a detailed report of issues found, use imagelab.report().\n"
+     ]
+    }
+   ],
+   "source": [
+    "imagelab = Imagelab(hf_dataset=dataset, image_key=\"img\")\n",
+    "imagelab.find_issues()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Get a report of all the issues found"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Issues found in order of severity in the dataset\n",
+      "\n",
+      "|    | issue_type      |   num_images |\n",
+      "|---:|:----------------|-------------:|\n",
+      "|  0 | near_duplicates |           40 |\n",
+      "|  1 | dark            |           29 |\n",
+      "|  2 | light           |            3 |\n",
+      "|  3 | low_information |            1 | \n",
+      "\n",
+      "\n",
+      "Top 4 sets of images with near_duplicates issue\n",
+      "Set: 0\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 400x200 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Set: 1\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUoAAACxCAYAAACiC/RrAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAshElEQVR4nO2daaydZdnvr2da89pzd4e9O9EytIVSReGVg+/LgRxQOJExxaBGEgiJ0fBFEyVWPxkjCWriBxFi/EDMCWjU6InMBD2ggjJJaS20hd3uqXtee6/5me7zobu76xo2z168HUCuX9Kk91r3M9/PvZ/1v67nf1nGGAOKoijKsthnewcURVE+6OhEqSiKkoBOlIqiKAnoRKkoipKATpSKoigJ6ESpKIqSgE6UiqIoCehEqSiKkoBOlIqiKAl8ZCbKsbEx+MIXviB+t2nTJqhUKm2vc3h4GK688krYvn077Ny5E37961+zPrfeeit84hOfWGq/9tprcNlll8GFF14It99+OwRBsPTd/fffD+effz5s374dfvzjH7e9P8qHj9MxLgEAbrrpJuju7oZbb72VfRfHMVx22WXou9tvvx3OP/98uPDCC+Hee+9d+vxHP/oR7Ny5E3bt2gXXXHMNTExMLH33y1/+ErZt2wY7duyAr3/96+9rPz80GMVs3LjRlMvltpcbGxszr732mjHGmPHxcbNu3TpTqVSWvn/qqafM7t27zSWXXLL02cc//nHz4osvGmOM+d73vmcefPBBY4wxTz/9tLnuuuuM7/vGGGMmJibe7+Eo/ya833FpjDHPPfec+cMf/mBuueUW9t1DDz1kdu/ejb57/PHHTRzHxvd985//+Z/m2WefNcYYMz8/v9TnJz/5ibnnnnuMMcYcOHDAfPKTnzQLCwvGmH//8fqReaIcGhpaerKr1Wpwyy23wPbt2+GOO+4As/i6+6OPPgo33ngjAADs27cPduzYAbVabdl1rl27Fnbt2gUAAGvWrIG+vj6YnZ0FAIAgCOD73/8+7NmzBy1z9OhRuOyyywAA4KqrroLf/va3AADw4IMPwr333gue5wEAQH9//6k5cOUDzekYlwAAV155JRSLRfb57OwsPPLII3D33Xejzz/zmc+AZVngeR7s2rULRkdHAQCgo6NjqU+tVgPLsgAA4Oc//zncc889S9v4dx+vH5mJspWf/vSnMDAwAPv374fdu3fD0aNHAQDgtttuA9d14eGHH4a77roLHnjgAcjlcvDyyy/DXXfd9Z7rfOWVVyCKIli/fj0AHP/J8uUvf5kN1i1btsCTTz4JAAC/+93vlgbkwYMH4ZlnnoFLL70UrrnmGnj77bdP9WErH3BOx7ikfPvb34bvfOc74DiO+H25XIY//vGPcOWVVy599oMf/AA2btwIDz/88NIf/oMHD8LevXvhU5/6FHz605+Gl1566f0d9IeFs/1Ie6Z49913l34C33DDDeYvf/nL0nfd3d1LP3EmJiZMX1+f+epXv7ridc/MzJjt27cvrXNkZMRcddVVJo5jtF1jjNm3b5+5+uqrzSWXXGL27NljLr74YmOMMTt27DDf+MY3jDHGPPHEE+byyy//bx2v8uHgdI7L5557Dv28fvXVV81tt90mfmeMMXEcm5tvvtncf//94vp++MMfmu9+97vGGGOuv/56c9ttt5kwDM0bb7xhzjnnHBPH8Yr37cPGR/KJEgCWfkJQjh49Cp7nwfj4+IrW02w24cYbb4RvfetbcPnllwMAwOuvvw779++HzZs3wxVXXAF79+6F6667DgAAtm/fDs888wy8/PLLcO2118K5554LAAADAwNw8803AwDAtddeq0+UH1FO1biUePHFF+H555+HTZs2wec//3l4/PHH0U/wb37zm9Dd3b1sYOaLX/wi/OY3vwGA4+P1hhtuAMdx4KKLLoJMJgPT09Pve98+6HwkJ8orrrgCHn30UQAAeOKJJ2Bubg4AAHzfh7vvvhsee+wx8H0ffvWrX73neowxcMcdd8BVV10FX/rSl5Y+v/7662F8fByGhobghRdegIsuuggee+wxAACYmpoCAIAwDOG+++5bGqif+9zn4E9/+hMAAPz973+HjRs3ntJjVj74nKpxuRxf+cpXYHR0FIaGhuCRRx6Bz372s/DQQw8BAMDPfvYzeO211+CBBx5Ayxw8eHDp/7///e/hggsuAAA8Xo8cOQKVSgV6e3vf1359KDjbj7RnitafONVq1dx0001m27Zt5o477jAbNmww5XLZ7Nmzx+zZs8cYY8zo6Kg577zzzNTUlPnHP/5h7rzzTrbO559/3liWZS6++OKlf2+88cay2zXGmPvvv9+cd9555txzzzX33Xff0ueNRsPs3r3b7Nixw1x66aXm1VdfPR2nQfmAcTrGpTHGXH311aavr89ks1kzMDBg/vrXv6Lv6U9vx3HM1q1bl8bxL37xC2OMMXfeeafZsWOH2blzp7n++uvN8PCwMeb4z/Svfe1rZvv27WbXrl3mqaeeOuXn5oOEZYw6nCuKorwXH8mf3oqiKO2gE6WiKEoCOlEqiqIkoBOloihKAjpRKoqiJKATpaIoSgLuSjrFcQxjY2NQLBaXfXNAUVoxxkC5XIZ169aBbZ/+v8c6RpV2aWeMrmiiHBsbWzJ7UJR2GB4ehsHBwdO+HR2jyvtlJWN0RRPlCQecp3//fyCfzwEAQCzkqZsV/CHPuHjm9pw0aschn9mDKEBt1+MbOvjWIdT+24uvonb/4Aa2TKWOraq2njPA+mw7By83MTaK2pMTY2yZuRn8zmtpbpb1qS7Mo3ZKuBIeeTJqEBPXaq3Olpkp4/XWY+E6eXhjIbmWkq9Mp5dB7UqTb3s+aJ5cZxTDS68fE62+TgcntvPSn/8vFAr54x9G/Ngtcj4c4enT9vCYtF0PtWPhnEZxhJcRzuLbBw6j9ut730TtdFeBLVNdKKP2f1z8MdZn/QC2OBs/NozaE8f4GK3Mz+HtlBdYn0YNbzuV4ucqCkO8TKOJ2pNTk2yZ0gLeVuDwez5y8LZCsp28ha8JAEDeS6H2dIOP0UbLmAjCCJ75874VjdEVTZQnfsrk8zko5I8Pwvc9UXr4pKTIRBlJE2VIJkrhguVyWdROp/BJy2TwjQ4AEJLBnctmWZ8Tx3uCCtlOVlhvPU22neIXNSQzozRRpshNTCe4wOM3o0f+EIXSROni5SwTo7YD/PymyDJexK+TK3isnKmfwSe2UyjkoVhYnHCimPdjEyXfZzv1PibKiEyUFr82+VwOtemYlMZo5PvvuQ4AOHm8iyzkcZ9cVlhvkzygBGnWByI86aWF+y6kkxwZS2mPD+wUGbeWMFGGZKK0LHzOU8L5TZFteRHvE1nCH88VjFEN5iiKoiSwoifKE6RtF9LO4iLSG+JkYraFmTqddkkfPOtHwtOMDfRnTcT6DKxbgz9w8Xb+/Kfn2DJbtmxG7bEUfwKZGjqI2pND+Cd+ucR/VsdEKogCn/Wp16p4GeEvXZ484Voh3r/ISE9M+DPX4Zc4IMvRpyzpD2yTPNXTNgCAafm5aYTreCZIWw6kF8eULZjTuuSJWxqj9InSkCeeSDh2P6Lnna937dpu1J56Gv9EHtl3jC2zeRMeo2/863XWZ/gI3t+jh/ejdmWhxJYxIR6Tkd9kfWpkjKY8/svII7/c2HaafOyn6POZ9FM0JE/+pI8jTECt9acAAIKQ3x9Ry6+eaCU/gRfRJ0pFUZQEdKJUFEVJQCdKRVGUBHSiVBRFSaCtYI6JIzCLaRC2MMemHJrywlcfk0BCFGMB1raEdAKiyYcBz49yAOdZfe76a1F7ahwL5wAAzz/7FGqv6ulmfWzTQO2+It6/tCekxJC0I9cWgloZvNzcglDongQI6Pm1pPWSXLJASONySUpMGOJjdIU0lSbJY/NJGwAA7Nb9O0t+0HF8/B8AWEJAySNj0rWFFBJyziISJLANDyZ6Nv4sDBqsD8nagRv+9/9E7Qce+jlb5tW/v4DaHV18jOYsPE46yL7QlDEAAJem2wjJszH5rFzmYzSTxmOFjkkp/YqOYzGmwgJmJAgnvElTJwEpKY2rNXhntxFw1CdKRVGUBHSiVBRFSUAnSkVRlATa0iiDKIJgUaN0gSdzAtF7bEF8oK8LURWhSZJcAQCOvouTvifG3mV9hsfw+6yOh1893LKGv0e7/xWsuYwcmWd91m/AieyBTY+bnweqT9mCXpfJYC1RerWq5uP1WGn6iiU/v4YmVAvvO9P3cdnrYMKrf40AL2Ok69+aWC+s40wQxzHEixqlEcZfYBG9McWfFTyqBZNjKQkJ3MMjeExOTfHk8ckJ7BNQJK/HXryFm3r84fEDqE19BAAAtq5fi9p2ltzWEddUqX+CEcafbeP1+OJLBpgUSUB3XWEOIC8CmJjvXxziz2iyu7QvDXJMNEkdAKD19jXCeVkOfaJUFEVJQCdKRVGUBHSiVBRFSUAnSkVRlATaCubUwxCcRRGVOYAAQMrFQq5oMESCOYcPYjee//cMd/n55ysvoXbG48nO/b0deF+JyW1XRydb5hMXYFPev/1ziPVpNIgXZhfOGraE5GOayGoJSd9RgAMEtuBZWatgY2Enxn0kz0PqE1r3uegNJBE3m8XuM2ViaAwAEJCrmUrzoZNuGROSc8uZoB5H4MbHxweLuwGAIQnRNPEaAMAj53CIjNEnn/ojW+aNvf9A7VyBu+qs6sLBm5AYL3cWetkyOzZi5+1/HDjE+gQk6Ac5fC9ILlOGBKiEvHAAcm6ojykAQI0EUCxiqBsLTl/07q0IBrsp6vxEEsyrIU/o922aeM/HqNcScJZujeXQJ0pFUZQEdKJUFEVJQCdKRVGUBNpzOM84kM4cX8QWEoprdVw0aHy0xPrsP7APtV9//XXUPvIOTyavkQTfdavyrA91bC6mSWJ7AxdKAgDIOTiBO58RNJgySYDvJu7XhgsdLtEo45DrYIaYSjhC/ZuY5Mv6rB4Jr3NSo8YArAdAmiTv0gJRnuAM7tFtC4Yc+ZZE7cA6Oxql69ngLhqVUOd2AIB6E1/Pyclx1meIFAH756svo/Y7Q2+xZaanJlB7ndvHdy7G442qmKbKTSf6C1gTX1XkdZ1MgK8frZMkJXS7xBwkiLjuH5OE7HSGj7eghvVsWqhOGgXNCN+rlpCU7pAxGJM1SfeLQ447IxhneC1u9dKLIMuhT5SKoigJ6ESpKIqSgE6UiqIoCbSlUVZmRsDUj2sk40eG2PcjR0dQe+jICOuz701scEGlkXROKGzegQ0tsjlesLxUwdocLUifibmWmHLxZ91FrsFUyP6FBn9Aa2ID8DxKSaehOX5CAUjIEVPUiOhK8w2eSxYSPZTWOgbgBgg+qRLpCnWW81T/kRLvzDL/P4PUpsfArh/X9abGR9n3h4kG/s4Roc+BI6gdEkMQJ8UPLpvFWmJHmuftzpbI9SJ1s4sRz191LDwm13dz7TMgFR8N0QDpvQAAEFPzZiHfkZpKpIWKlZGLx6gx+P6t1PkYBWIsnBJyWU1ETVgwWUF/zBJTHqn2t9+yoqCNMapPlIqiKAnoRKkoipKATpSKoigJ6ESpKIqSQFvBnL88+yRkFs0bZid4oi41cajUeQAlQ4TnOVLZrUyCMgAAa/qxWYBU1XCugl+stzzcJ5/hJgXUNTnHiw9C5BPzCrLtSDB/MNTFXdhf6pPhCcJzliTAL5DE4kYTi/YA3JhEqjN3wgH8BDYJ3rhC1T6IiHO6sF7TsjWpAuKZ4MXnnoZs+vg5KE1zl/G5efxSxHyNJ1rbMR6D9Qpexo95gGJVNw7eWCEPjsyX8Xoy5Pp2CFVLHRK0yHg82GmR5H6LhA8lN3paoZCO2RNrQvsiBPnSJFi40MDnM5CS3T28bSnx25AgKU2Qt4WXIoCOa8HhvHWMGq3CqCiKcurQiVJRFCUBnSgVRVESaEujnBodhfTiy+ieYFhLfCggFow+s2t6UDuXw5ra6LEptgytllaam2V9CsR8Nt+Jk9Rzgosrra4HgvbkEDHRIXKK4MkLNjnuMBa2TTQhW9CegOg7QYA133Sa664O+dtHDToAeBU8qvmamOt2IdWaBP2nVdOStNszweTRIcikjg/rlGDMkSePBpEnJNev6kLtHKnUeGyaV0Kka6lWuQlLRx6P0Y4OnKSeM3xf6KWIqnyMGqrxEb1b0oupxicZZ1Dt2hI0VLpYSN4g8QRDatvBC9nCdbJJQrxNkslDoYKiTW7GSLjvTMt6pGNeDn2iVBRFSUAnSkVRlAR0olQURUlAJ0pFUZQE2grmxGEE0aIu7EjV1UiwIRLsOdwMnpt7+3DFuFSOOziPHBlD7azNXX42bRzAHxD9OmxwZxabiNPNOk/gpkm29C+L5woJwDSIJYjGNNnVj/i5YsboJADkCInsNhGwaXAHAMAhQrkhy9BKjgAAbpocpy8kai/z/zNJFPoQLQYHYiEn2W/gZPJQGqMOHl+rertR28tyh/3xUexClO/hgbZ1a7DzT0Rcm2JfSLwm46/W4C7oLglkWsQ73RacdmjAUXLhpw77TSHIF9EXJciYlLZNE+Jp4AYAwCEvpvg+rfYojH2yLUtwbfdaXPgNPcD3QJ8oFUVREtCJUlEUJQGdKBVFURJoz+HcDyC1qGflhCpo1QbWEfyQawCZFNbrLOI83gy4nje3gJNsOwtcIyoWcYI5ECfyps01mCox4LCEl/M7itgpwyL6nZRQGxhqOsFPc0j6UGdyAICAGheQbQsSJdCCdrag5aRTWNPyiYN3ICSLW0Dco4Xj9lrcrW1r5cm8p5JKEEBwQgNzuE4418DmKTXu2wI9xOk+ItehWufHNlXC6+3p7WF9urqwcUZYxxUhm9wPBmbnsJFGKFT9LGSw4z8dJ6kUPw8RGevUdAIAICCaZCSYa1Btnb5ckQY+T9CnM1px8fhnRG+0iN64Ao0yKxjhNFv0UCNU6VwOfaJUFEVJQCdKRVGUBHSiVBRFSaAtjbLkh+At/q4PhFy7hSbWbuoNruWs68ZGAE2SH1Wq8WXKTbyt4UluOPAxopVYAdY1A0H7LBOj4X6S0wkAkM1ijbJUxrqSLRhTuESTbBrBZIIYctDcPQBeTc+iRgFsCQCSfgaeIGR61GCAGg0L6/Wp2YGQ+xaGYcv/z44pxnwYQ2rxeGjuHQDAHBkHNZ/vZwfRzIIm7lNpCsbRZNyOzQpjlN4zZFzEhmt1dR8Llz19XaxPJoPHToPcd65QiROIRk6NZwAEA2o6uIBr7TQ/kxpdAwA4dnIfmidJ29R4GAAgoudT0PDjFh0zFo55OfSJUlEUJQGdKBVFURLQiVJRFCUBnSgVRVESaCuY06hHELrHRVRXEPPtNBaVy6TqHABAkyahW3gX/AAn7gIAOCSLeq7EhfKRMeyMvr4fJ6DPznMzgXIDB3xW93ezPkASZgPi7F2e5cfoFXECcFP4exSFONDg2jzw0CCmCUASao3giB0SkTsQzDYCi2yLVrwTri1NUBZzdVvE9NicnWBOrdmAYNENIwLBsIQYWlQEJ/JaQF+KwMcSxdw8hfqgTE5Nsj7j47gqZE8B79/k3Dxbpk4MVXp6hTFKznVQx2O9VuHH6JCXDqgzOQBASKoYCF2AFAYF18HBz1hwwo9CPL5ooQEAgBS551kXqcIiGZSh8NJDaxAoFiplLoc+USqKoiSgE6WiKEoCOlEqiqIk0JZGmQID3qJWFfhcS/SyOJk8n8uxPq5NNAySUJunrg4AcP45a1A7k+bmvk3i0npoHOs0gZBYnCvi9YSGm/s2faxjjM5gTbIsaJ9BCn/mW1wrSwHRKIUijAGrIkcSzqlpKgCERF9rCAJQMYc3tqYDJ8135nnyOzUDMYJxc+thCp4FZ4SUdfwfAEAYcJcJL4tfKsikuMEKNQBxHKxJ5j0+9rdtwaa8uSwf+7NlvJ5SjbwUIejJ6Ry+FnHMtew6Wc/4JNbr52tcU42JYQg1/gAQTKsFI+SIxBwsg6cUSUf3ycsIkfC81lnAn/V14jFbTOF5BADAJnppLDwGOi0vgzhSh2XQJ0pFUZQEdKJUFEVJQCdKRVGUBHSiVBRFSaCtYE6h0wbPPT63BkL1urBBRG/BEcchSZ61Cg58zM/NsGX6Vq9G7a7eTtZndAon1b47jAXtDX1c/B1cjddzwSAX9sfGcOLwvyo4mGMHgrjewKryFNf+wbNItcSUkEBLEr+DILl6HY3/RELid3WeOMfUceDh8hv/F1um0cCuSf96803WJ505eY4j6+wknOfyHZBadMtp+lX2fdTEgY+OFI9QuDbe90YNX8ByeY4t09fXj9r9q1axPocm8NiZmC2h9nqhcuPmtXhMbl3Nx+hbB3GV0rdncRDLCK5ENJBUbgqu9h4eX15KqBJJLHrovGAJz2I8wMPXWyaO8Z35QdTe9R+72DKH33obtasVnsCfaUm0j6lr+nugT5SKoigJ6ESpKIqSgE6UiqIoCbSlUVYDH7xFfaEzX2TfOzFOCl2Y5cnYsU9ciEOsRfg+T44dHp1A7X2Hxlifso/nfJckpE4Irug7rt2J2tf9jwtZn717se7xz31YQ63Zgg7WwPtSarAuYFtYl8lSVwXgVeWaJNnYFqrX0aqLkkbpEQ0OiNa5ftNmtszMLL4GkulFq39DfHaKMEK5EUNqMcm+u9jFvk+R814plVif0McvHlBzknqDJ7IfHSY64cGjrM9MhPXFiBjCWPPcSOOL116G2v91ybmsT959DbXf3IvNN4zHX9AIfKJRNvj1tMn0UHCFuESI7+cUmVHoGAbgOjo1vQEAMGSs5/PY5GbTVj5Gh44eRu16iccP7JaN8xc6lkefKBVFURLQiVJRFCUBnSgVRVESaEuj3P/2zFLFtM3rua5wzmAvam85v5f1ofJYs9GF2lOCyYSTwdqO8bhWUpvA2uHmHmx+sKGH559t3YRzs/o3b2N98hMk39H5M2r3dvD1lkOsYbkW31+bVGrsSXO9MUPy2HyiEUVCFTmLVLRz0jx/1CcCYmSwlvPnF/7KlqnVcU5aKJkdRCe3HURnxxXj4NCxJTOHDat72PdbB3G+44b1g6yPH2PdrdgglULnuC6dcskYjbnBikOMeTcN4H1Z38nvl4F1A6id3cR19Oy+cdTOZfH1Tef4tZit4tiApBPmSY7p2iwfo1Tm84mea3HLXTBEWzcpnm9NvGhgenoWtf/0wl/YMmMz06gtyeStedzUnOO90CdKRVGUBHSiVBRFSUAnSkVRlAR0olQURUmgrWDObNksOVfXDk6w72t1HMTYeeEA67OqCwdZLGIGkSvy4IiTwsntthDMyZCAyQWD2PDi6svP58t4WO49OsyPaXSGuFDbOKm+vwsnwgIA9AAW8kdqvFIjDbpIAZ88CfB05PG5mZ/n6+3qwueqs5sbiJSIycMoEcoPHxnl+5LDhg2hEKyJW+TzQCqtdwaYLtWWkpyrZZ4YXinjY9914TmsT18fPmdelgR3OrvYMibC56ejizuc5zL4dtt1Hg7mXHnZRWyZDHG+nxg5xvoMTRGTjgzdF55w3tPAQb5jQsWCFLnP7Ii/DNJRwMdp23jbUpXLvtW4kmSU4vb+cwt4f0pVfE+9exQHsAAAYnoPCa7tVkswJ9AqjIqiKKcOnSgVRVES0IlSURQlgbY0ysg5WV2vFvDk07eP4GTxueow63PhebiiYlcBJ5v2rcG6DQDA5ARO1J2b4dpcPot1mMkSTfjlmllQK6H2yMEDrM/rrxxE7Rqp3FYNuQ6yqh8nDmdnuP4zX8GfzcZc86v6ONG52IH1oKZQhdEm5gHd3Vwra0T43MSAt11v8mMqFPF6QovrO62aT3iWNMrYTgOc0CgDfk4PjWA9b6G5n/XZcS4eo90d+Jx29fDE8LER/MLDwjx3QskQnXdqHo/rlGA6ETfwekcOTbM+e9/Ex1A1WKfuCPht3t2F7zN3kmuflTq+xhNNbnRbAawDZ4gOG9p8224B36uFDB8rlRr+LCDVRcNYGPtprHU2fSE20GLiEQgvbCyHPlEqiqIkoBOloihKAjpRKoqiJKATpaIoSgJtBXNsiME6IfxbPEk0IIGOsQme8Lsw/y5qbxjAyb0bB9ezZbq68bbmprnD0NRMCbW9VBdqj0xxYXfbuRvxBzF32jk0hJ2qh8m2Z+a4EO2kcJ9SnTstV3y8XE2oqGg1iaN5DYv/OaGCoDOHAzXdXfyYqsThZY5U5FvdI7hSe3ioGI9f/zg4KZTHhgcmzgRxbOBEVT8L+PkJAAdUxoQSmeXyO6i9YR0OfAyu5Y5D3d246uL87Lusz9QMvn6dHX2oPSwEKTduxvdDNM8TuMdGccXR4Rk8tuZL3O2oSYJt8w1+verkpYKaxYNjNnGDd4g9WJZmzANAlgRaB/t4n2oN96kRV3knxStWsilJeIkjNpH4/yT0iVJRFCUBnSgVRVES0IlSURQlgbY0SsvAUmqysXjyaUQSl6WU44Um/vTgUWzIUPf5Lm1YizWirj6e8JtpYl2wuxtrc/1r1rFlNm37OGoPHeXVHT3y0v9sGb+MX89xw4EK0ZHKDW4mYGgiriDpUZ3Ppn34JQA/xNdg+BjXveqkgmJEnNSBG05DlVTLjIDrrq7TsoNnSaM0VgjmhN4raFSGJfZzfaxEEq2bwyXUrvlc+1y/Go/JYjcfF9kQn2dqWJLp5S9bdG3Zjtr1d3l1RzeLx+j0/AhqB3luNDNfx7r0fMDPVUTGqB8JJhNkLNEKn26G388LVTx2xg2PZTTIOLaJJhlKVUBpBVd2wwDYjt3y9crHqD5RKoqiJKATpaIoSgI6USqKoiSgE6WiKEoCbQVzYjcG60TSqZB8SufdmNayBICICMIRcd85NMwdh2pNnHy6ecNa1qe3Cx+K1Syh9tjoEbZMQMrKbt7IAz4XXLAJtZ/8G3YYctPYsR0AoODhzxrhDOsTRljQjgUXIloRNqJth1++aogF7WzIgwo+dSc3RDgXkt8NKeEaBFyAb00CNoLYfiawUgasRRFf3AeTPEYNccauEGfvw2M4WAIAUG3ioNnA6j7Wp+jhAGM5wAno7xzDie4AAJXwE6jdM8jX208+M3vxetIZXrY3Z+PzUDH8JY5GhK9xLEQcbRIwi8g5N8I00SCuPQ2h9HGTLBjZOIAWAF9xRK6b9MKBcVrGaKwJ54qiKKcMnSgVRVES0IlSURQlgbY0StdNL1UPDCX5h+gKls01AiC6QUwqoRkhkXl8Gms5vuCevXkQmxLkyBvybx7mutLvH3sKtdcN8qqRVZLIvnkTMT+o8axvl1SV27qZa6qeg4+7XOKO2LPEBd3N40zwzm7sZA0AkHXxJe3I8SqRx47hbXkksdgVnNOtGF+XlMv/xtrOyeO2z5JGaTvppaRiQX4EKktZjmAAQiVcslAkaHVjxCzFj/g53LgeJ6WnyHl+6x0+Rh9/8lnU7il2sT5p4uy9dQsex4061/MKZJysHuhmfUILn8C5eW4g0qjj+EE+h5PfCznusE/9VGyPX6jaHDbycNMkaV7Q0enc4nhcn49bxmgsVBVYDn2iVBRFSUAnSkVRlAR0olQURUmgPY3SWGCf0CEFAYh6EBhB6wKS/0T7uA5fJiY5h5OT3OghJrrlmn6szeU7uAbz97dwbqU5cIj1aRLz3PWDOCeteniCLVNrYg1w4wA3O+gpkpM1yHPdSjWcx2bncB5eoZPncM5N4yqD8xO8al9ENLe1a3DVQVcyEQ7xecgKpsFxywAwgiHFmcCN46W//lQzP/6ZTdq8j0fGZEi0LFfSNckzx8wMN8sFNkax7tbXhU0yAADeegsbtUQNng9MNdQ1a/B6Dh+S8njxvqzu5mMpWyAVFddxM5p6A4/1LDHokOIUk1O44uPc/Bzrk83jsd7RibcdhzyWYRMTDNcV8ihb5i2aL/te6BOloihKAjpRKoqiJKATpaIoSgI6USqKoiTQVjBn4+rMkpBthJfSq8TJu9EUXjo3OEHbb+K2JQSAImLi4Df5/F4v44TfCYNF5kJGSKK2cQJ3LsOPKZUiL/0TAdi2JeMPfEzZFBeN3Rgn0dtCVcvVvVgYbxp8fsM6rr4HAJCO8XF3d/Hj9tJYKPfJ/tYrPIm+txPvny3Yq9f9kwGMWHoj4Qywqd8DdzEZ3rL48PbJmPR9fhw0wNNohOR7vt2QnEM6BgAAGlVS/XIWJ3D3dfL9TQEO+GQcXn3QJY706QZxsBf2hQaAcinex4tw0DRl8wTuQgHvT0CWCXjMBXJpvG2njweSAuIi75MXP2xHSKIv4nEd+Ny4JWgxnwlDNcVQFEU5ZehEqSiKkoBOlIqiKAm0pVHuuqATUt6idiBoiZU61gSMoBF5LtYWFhaw9hAEXHsg+dvQbHKRKAiw3pAmUk4XfzcfevP478Sqngzr45HDLNWxVjc0iisuAgAExEBidT9PJC7YtDIjP26f1LH0qKZiBN3VwYKVoQ4EADBXw+t950gJtednuYnrul5s7CElnOeyJ69LcJY0yo9dUIT04r5JpgdBgPfLFsax7RCD3QoegEHI9bxqDeuNUi5ztUF0QWJyUshzQa+rE+9vbwfXCane6JJrk86U2DJpUn1yYA1/IcMhL5XEIX+uCg3edkDOeSyY53Z34W3Vheu0sIDvj4MHhlA7m+alQjv6sJ4be/x+jlqeDf1AKGO6DPpEqSiKkoBOlIqiKAnoRKkoipKATpSKoigJtOke1AR30X3FFYIEqSL+LJ3hiaTFLO5TzRHVWxDXY5YAzEVvi7h028Q5pJDj6814WIhOsQALQEgq8HVksWBc8Lio3EzjoNaq1XzbThNHl+ImD35QKTpLIkuxEDGIPfyZUOAOUiThvFrB+7IwzQNUWeKi3dPFV9zqPO8HK0/mPZVkLRsyJ9yPBJcf28NRvnSKX79CATvH1xvYxTuI+DgJArzeyPDj9w3u42XwtruKPKjhAhmjwF2JQhI46yYBn5SQTJ7O4D6rVvN7NSTu5WFVeLmCfBSSbHw/4OcKHNwnIwQli6vwNRg7QqKxQkCtu4Cvty0k54ctpUybwssGy6FPlIqiKAnoRKkoipKATpSKoigJtKVRZiwb0ov6jyVU2avUcKLy3Dh3IjersJN3muhugaBpZDNYayh0cn20QXSaKtGVshlejbCQxev1G1yzCH1SfTCN2wWP6yBli1RYLHOX8YJLDC/qvMJdGOJzQas71gXHAepenha05DTJWe7twX3G8/yYPFIpLwK+v+XqycRsmth9pnBsD5xFV23H4cO7VsPjYmqSG4usWoUrbWZIcrOJuP6YTeFzZgkO780Yj/VmiBPZMy5/KyLt4Gvj1wR9nuiClsEaeSbN98Uh+vzUDHdBz7r4GtbqvFIoTb53iKN5UzCmcMiLKJagJToZ3KerswvvywLX0R0L70uzwV+cqNZOHnc7Oro+USqKoiSgE6WiKEoCOlEqiqIk0JZGmfVcyKSOz620ShoAQExeMvddrqfMlrBely/ivMR0SjBxqGCdwxLmd2r4S993Hz/G9VIbsHaTT3ONKJfDn4XERFiQAKEjj0VAv847TRGnj5RQrc6y8Tk25HJZwsZtolfFtqC7hlinc1y8Xjcl5MtZWC+t+jXWJ2zRrUM4OxqlnQrATh/fdibHz0+TGEenBZmqVMNVAXMG5/S5gi5dreKxLhQpBZvkdYYRvlYTx7imZpF8zEwqz/pk0njs1Jv4fslnea6o5eBxXeG3ByyQfNG0YERhEeOJ2ML3hwX8XNlEd7UtwRCGnMAUyYtuCPpzSPKtqV5/fNst/29jiOoTpaIoSgI6USqKoiSgE6WiKEoCOlEqiqIk0FYwpx7AkkRvZbhQblwcxMh3CQEf6tpNEnXrTR58iCKbtPm+OdQhnCT3GpuLykAFYp//3ZidmyWrxfufyeAEegCAosFC+ewUD3xMT2P1vKPIhfI1a/pQmybEL1S5+N/Ric0Nwho/WWlSkdKhgrvgOO0HeJlshjttt/qFuEEMALOsz+nGDwDsxd33UlLQD5/nfEEwTiCBBIsEu+qCKQvL/RfcSDwSbIgjfE4jIUDhucRERngpYnxqEn9Akt07C/xaBWTbszN8jJbK+Pp1dPH19Pbi8V8nL04EIT9X+Sw+zqDJjT6yeRy0suntLVQEaJJAq5vi1zbXUvrA8TXhXFEU5ZShE6WiKEoCOlEqiqIk0JZGOb3QgJR7fG6dqfLf946H5103LVS4I+0oJC/sx4LZKtE1aUItAECzgl+Sz2exnhIJL957RFNtCuuNIrxcbNHj5stk0+S0OlzPLXbi44wjbjgwPTOP2rS6XkjPHQBMTuFl6oKJQi6DtZwgwPpU2sMJ1gAAUYw1o9I8N/rIthgZCH4dZ4TSVH3JaKU2zzOKqeG04/AxCi6+xoZollKCNBujDW4aUvGxLp1OYQ0/CLimnyHJ4r5kNkKqRpoIj0lHeOnAtsl95nL9scPuRe2mz7PSFxawvkjPb6MuaJ8lrKn6wrnK5/G25kt4f1MpXtl0oYaP2xXuzWLhZPxAqpS5HPpEqSiKkoBOlIqiKAnoRKkoipKATpSKoigJtBXMKWbSkF4M2PghF5VpECOd4/OwA1icPjZWQu1ajQc1cjniHh1xcdoB/FmtikXm0jxPavXJ34m8IOwPrsZu18YhLsoxF6up43QQCxUrMySJvsGVZZfq7cR9xhOq17nkmGLDtx0b7AoThlj03rRxK1tmeBIL8LHNhXKrJck6EMbHmaAzm4fMYtArEtT6PKlAmc3yIIYP+PyMT4zj7wUXfuoy5QkBFHBwkKhJghjlKh+jsVVC7ZTgiNTf34WXIS9OuEJ10djgaNuCUJHQJoEZO+BBLMciY5JUQ80KVS5dB39WFyqZmhifm84OHGB0Mt1smaPHDqH26i4elAyCk/OLOpwriqKcQnSiVBRFSWBFP73N4k/J1hwu6aeVTR9lfaGoEXlHk/6El9ZL+0S06joABCSnkOa+SU/Z9MdGYKR3nGlOXUS+5/tC/UID4BsPLXJMwnFbpE9s02Piy9AzTg2MAXhhuJBs2xcWCskL9rHgTNt67U6s05g2ktX+G5zYTrPlesXCT2/HI8cm/EQOyMhokmssFU6j7w3TcyotFwbvfS8AALAhGfBnmyb52RyTfbEEcwT6rrQv/Ky2yPvq0lhn9wc5d9LPWzZNCH0s+r49uakcRzKkptdJOO6W9Zz4fiVj1DIr6DUyMgLr169PXJmiUIaHh2FwcPC0b0fHqPJ+WckYXdFEGccxjI2NQbFYBEt4K0FRKMYYKJfLsG7dOv4WyGlAx6jSLu2M0RVNlIqiKB9lNJijKIqSgE6UiqIoCehEqSiKkoBOlIqiKAnoRKkoipKATpSKoigJ6ESpKIqSwP8HnwKzJKZVcTEAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 400x200 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Set: 2\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 400x200 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Set: 3\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 400x200 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "Top 4 examples with dark issue in the dataset.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 800x200 with 4 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Found 3 examples with light issue in the dataset.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 600x200 with 3 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Found 1 example with low_information issue in the dataset.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAK4AAAC+CAYAAACs2jR1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAJZUlEQVR4nO3dW0iU6x7H8f+Y5arlIbdlNY6HpN3JOaHSMqO8a12MwqYuIqQrESLKoPKyCLqrLiK6CLqoyw47yPJCDYpNh50YaIorirTMHE95yOlApe9vX7gd1viMNp7Iv/0+sGDN4+v4rta3x+fhnZnXBgBCpEzUzz4BoulguKQSwyWVGC6pxHBJJYZLKjFcUonhkkoMl1RiuHMMgOzfv1/WrVsnubm50tLSYhxz9+5dyc7OFpfLJfn5+dLU1CQiIoODg5Kbmyter1ecTqdcunRJREQCgYB4vd7gPwkJCXLu3LmQ5zx27JisWLFizv/7fhrQpIaHh2f0/Xfu3MHu3buNf/+7+vp6dHZ2AgCqq6uxffv24M/+9OkTAODjx4/IyMjA+/fvQ77XsiykpaWhtbU1ONbc3Izi4mIkJSXN6NznswUXbiAQwJ9//gmn0wmn04mqqioAQGVlJbxeL9xuN/bu3QsAaGlpQUFBAVwuF4qKitDX1wcAKCgowOHDh5GTk4PLly+jqqoKeXl58Hq9KC4uxtevXyM+n9LSUty6dQvAaGRr1qyBZVkTHt/f34/MzExjvK+vD+np6ejt7Q0Zf/ToEbZs2RIy5vP50NbWtqDDXXBLherqaklKSpKmpiZpbGyUrVu3Sk9Pjxw6dEgqKyvl2bNncuHCBRERKSsrkwMHDkhjY6Ns27ZNTp48GXyexYsXy9OnT6WwsFDOnDkj9+7dk/r6esnMzAz+yv67EydOyO3bt41xv98vKSkpIiJis9kkMTFR+vr6Jjz/K1euyM6dO4OPBwcHxePxiMPhkPLycuPX//Xr12XPnj3Bx9euXZPc3FxJS0uL7A9Mq5/9N2e2vXjxAg6HA+Xl5Xj8+DEAoKKiAiUlJcaxycnJwdmvo6MDXq8XwOiMW1tbC2D01/vKlSvh8Xjg8XiwceNGHD16NOLz8fl8qKurCz7evHmzMWuOefLkCTZs2GAsBwCgq6sL+fn56OrqCo5ZlgWHw4H29nYAo8uJ/Px8fP78GQA442qyfv16aWhokKysLDly5Ehwdg3HZrNN+LVly5aJiIhlWeLz+aShoUEaGhrk+fPncvbs2YjPx263S0dHh4iMbtQGBgYkKSnJOO7169eyb98+uXnzZtivr1q1Sjwejzx48CA49vDhQ0lPTxeHwyEiIq2trfLq1SvZtGmTZGRkyMDAgLjd7ojPVZWf/TdntnV0dODLly8AgKtXr6KkpATd3d1Yu3Yt3r17BwDBtazP58ONGzcAAKdPn0ZZWRmA0Rm3qakJANDT04P09HS8efMGAPDhw4eQjdCPVFRUhGzOdu3aZRzT398Pl8uFmpqakPGuri4MDQ0BAAYHB5GVlYXGxsbg1w8ePIjz589P+LMX8oy74MKtqqqC0+mEx+NBXl4empubAYxG4/F44Ha7UVxcDGB0c7Zjxw64XC4UFhaGbM7GwgWAmpoa5OTkwOVywePx4P79+8bPPX78OCoqKozxkZERlJaWIjMzE9nZ2Xj58iUAoK6uLrh8OXXqFGJjY4PLkbHNVm1tbfCcXS4XLl68GPK8KSkp8Pv9E/5ZLORwbQDfAUH6LLg1Lv0aGC6pxHBJJYZLKjFcUonhkkrRkRxkWZb4/X6Ji4ub9GoT0UwAkEAgIHa7XaKiJp9TIwrX7/dLamrqrJwc0Y+0t7cHL2NPJKJw4+Ligk8YHx8/8zMjCmNoaEhSU1ODvU0monDHlgfx8fEMl+ZcJMtRbs5IJYZLKjFcUonhkkoMl1RiuKQSwyWVGC6pxHBJJYZLKjFcUonhkkoMl1RiuKQSwyWVGC6pxHBJJYZLKjFcUonhkkoMl1RiuKQSwyWVGC6pxHBJJYZLKkX0EUz0awp3X5v58mmdnHFJJYZLKjFcUonhkkrcnCkW7p6gke2dwnxj2Oea243Y+M3fVG5yyhmXVGK4pBLDJZW4xlUMGDbG3vf2GmPfh0PXjkuXLjWO+e23GGOsu7vbGIuOXmyMjYyMGGOrV68OebxkyRLjmJngjEsqMVxSieGSSgyXVOLmTLHhb1+MsXdvW42xT99C56eEWPMmi/998h9jrK3trTFWVPQvY2xgYMAYG393yJgYc/M3lQsO43HGJZUYLqnEcEklhksqcXOmRLhtTPQS8wrYPzc4jbFnTX+FPH7zts045k7lv42xosJdxtgff+SFORPLGBm/8bIs85iZvPqMMy6pxHBJJYZLKjFcUombMyXCbmOizP99sC0yxkaGQ1/+2NvTaRyTmLjcGEtIMK+wWZb5EsaoKHP+Gz8WyWc0TGWzxhmXVGK4pBLDJZUYLqnEzZlitjDX06IXmZuz8Tq73hljq5Idxpjdbo4tiuD5w5ntz2jgjEsqMVxSieGSSgyXVOLmTLFwb9kK98EbK1cmhTy2hbkOFx+XYIwtX/6P6Z/cHOOMSyoxXFKJ4ZJKXOMuMOFeqbVx46aQx+s3rDeOaX31yhj7/ffYiH7mz7g7D2dcUonhkkoMl1RiuKQSN2e/gPEbpTWr7cYxXZ3mK8bi48237oR//umd10xwxiWVGC6pxHBJJYZLKnFzpth0r05lrs00xr6F+XTz5csTIz2TaZ3HTHDGJZUYLqnEcEklhksqcXP2C4AV+rLD5OTVxjHhxhZFz988OOOSSgyXVGK4pBLDJZXm7+qbZs+4C1u2MO9Lm+v3iM02zrikEsMllRguqcQ1Lv2fecvS8PPa/FgLc8YllRguqcRwSSWGSypxc/YLmMmtR+crzrikEsMllRguqcRwSSWGSyoxXFKJ4ZJKDJdUYrikEsMllRguqcRwSSWGSyoxXFKJ4ZJKDJdUYrikEsMllRguqcRwSSWGSyoxXFKJ4ZJKDJdUYrikEsMllRguqcRwSSWGSyoxXFKJ4ZJKDJdUYrikEsMllRguqcRwSSWGSyoxXFJpSreLAiAAJj1mIdyKiOY/zrikEsMllRguqcRwSSWGSyoxXFKJ4ZJKDJdUmtIFCJvNxgsMNC9wxiWVGC6pxHBJJYZLKk1pc9bf3y/fv38PPo6NjTWOiYmJmflZEf0AZ1xSieGSSgyXVIpojTv2rodAIBAyblmWcSzXuDRdQ0NDIiI/fJeNSIThjgXrdrtncFpEkQkEApKQkDDpMTZEkLdlWeL3+yUuLo6XfGnOAJBAICB2u12ioiZfxUYULtF8w80ZqcRwSSWGSyoxXFKJ4ZJKDJdUYrik0v8ASW3kw7xCzEoAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 200x200 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "imagelab.report()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "View more information about each image, such as what types of issues it exhibits and its quality score with respect to each type of issue."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>odd_aspect_ratio_score</th>\n",
+       "      <th>is_odd_aspect_ratio_issue</th>\n",
+       "      <th>low_information_score</th>\n",
+       "      <th>is_low_information_issue</th>\n",
+       "      <th>light_score</th>\n",
+       "      <th>is_light_issue</th>\n",
+       "      <th>grayscale_score</th>\n",
+       "      <th>is_grayscale_issue</th>\n",
+       "      <th>dark_score</th>\n",
+       "      <th>is_dark_issue</th>\n",
+       "      <th>blurry_score</th>\n",
+       "      <th>is_blurry_issue</th>\n",
+       "      <th>is_exact_duplicates_issue</th>\n",
+       "      <th>is_near_duplicates_issue</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.813863</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.670485</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.761960</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.447264</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.889314</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.928179</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.870204</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.497561</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.868758</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.799635</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.752100</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.507733</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.883888</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.992232</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.872505</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.530581</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.902695</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.911035</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.897581</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.530771</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59995</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.860407</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.794629</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.996078</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.523458</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59996</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.888932</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.939203</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.843293</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.498186</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59997</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.818150</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.960275</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.865067</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.444907</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59998</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.900018</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.892104</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.952069</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.528622</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59999</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.858985</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.809504</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.932046</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.501550</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>60000 rows × 14 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       odd_aspect_ratio_score  is_odd_aspect_ratio_issue  \\\n",
+       "0                         1.0                      False   \n",
+       "1                         1.0                      False   \n",
+       "2                         1.0                      False   \n",
+       "3                         1.0                      False   \n",
+       "4                         1.0                      False   \n",
+       "...                       ...                        ...   \n",
+       "59995                     1.0                      False   \n",
+       "59996                     1.0                      False   \n",
+       "59997                     1.0                      False   \n",
+       "59998                     1.0                      False   \n",
+       "59999                     1.0                      False   \n",
+       "\n",
+       "       low_information_score  is_low_information_issue  light_score  \\\n",
+       "0                   0.813863                     False     0.670485   \n",
+       "1                   0.889314                     False     0.928179   \n",
+       "2                   0.868758                     False     0.799635   \n",
+       "3                   0.883888                     False     0.992232   \n",
+       "4                   0.902695                     False     0.911035   \n",
+       "...                      ...                       ...          ...   \n",
+       "59995               0.860407                     False     0.794629   \n",
+       "59996               0.888932                     False     0.939203   \n",
+       "59997               0.818150                     False     0.960275   \n",
+       "59998               0.900018                     False     0.892104   \n",
+       "59999               0.858985                     False     0.809504   \n",
+       "\n",
+       "       is_light_issue  grayscale_score  is_grayscale_issue  dark_score  \\\n",
+       "0               False                1               False    0.761960   \n",
+       "1               False                1               False    0.870204   \n",
+       "2               False                1               False    0.752100   \n",
+       "3               False                1               False    0.872505   \n",
+       "4               False                1               False    0.897581   \n",
+       "...               ...              ...                 ...         ...   \n",
+       "59995           False                1               False    0.996078   \n",
+       "59996           False                1               False    0.843293   \n",
+       "59997           False                1               False    0.865067   \n",
+       "59998           False                1               False    0.952069   \n",
+       "59999           False                1               False    0.932046   \n",
+       "\n",
+       "       is_dark_issue  blurry_score  is_blurry_issue  \\\n",
+       "0              False      0.447264            False   \n",
+       "1              False      0.497561            False   \n",
+       "2              False      0.507733            False   \n",
+       "3              False      0.530581            False   \n",
+       "4              False      0.530771            False   \n",
+       "...              ...           ...              ...   \n",
+       "59995          False      0.523458            False   \n",
+       "59996          False      0.498186            False   \n",
+       "59997          False      0.444907            False   \n",
+       "59998          False      0.528622            False   \n",
+       "59999          False      0.501550            False   \n",
+       "\n",
+       "       is_exact_duplicates_issue  is_near_duplicates_issue  \n",
+       "0                          False                     False  \n",
+       "1                          False                     False  \n",
+       "2                          False                     False  \n",
+       "3                          False                     False  \n",
+       "4                          False                     False  \n",
+       "...                          ...                       ...  \n",
+       "59995                      False                     False  \n",
+       "59996                      False                     False  \n",
+       "59997                      False                     False  \n",
+       "59998                      False                     False  \n",
+       "59999                      False                     False  \n",
+       "\n",
+       "[60000 rows x 14 columns]"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "imagelab.issues"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Get indices of all **dark** images in the dataset sorted by their dark score."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "indices = imagelab.issues.query('is_dark_issue').sort_values(by='dark_score').index.tolist()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "View the 5th darkest image in the dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<PIL.PngImagePlugin.PngImageFile image mode=RGB size=32x32>"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset[indices[5]]['img']"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "View global information about each issue, such as how many images in the dataset suffer from this issue."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>issue_type</th>\n",
+       "      <th>num_images</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>near_duplicates</td>\n",
+       "      <td>40</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>dark</td>\n",
+       "      <td>29</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>light</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>low_information</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>blurry</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>grayscale</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>odd_aspect_ratio</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>exact_duplicates</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         issue_type num_images\n",
+       "0   near_duplicates         40\n",
+       "1              dark         29\n",
+       "2             light          3\n",
+       "3   low_information          1\n",
+       "4            blurry          0\n",
+       "5         grayscale          0\n",
+       "6  odd_aspect_ratio          0\n",
+       "7  exact_duplicates          0"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "imagelab.issue_summary"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "**For more detailed guide on how to use CleanVision, check the [tutorial notebook](https://github.com/cleanlab/cleanvision/blob/main/examples/tutorial.ipynb).**"
+   ]
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/torchvision_dataset.ipynb b/torchvision_dataset.ipynb
new file mode 100644
index 0000000..f5f5e6b
--- /dev/null
+++ b/torchvision_dataset.ipynb
@@ -0,0 +1,869 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "8fa4a083-1a56-419f-9e3f-fa93acd45405",
+   "metadata": {},
+   "source": [
+    "# Run CleanVision on Torchvision dataset"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "6c52419e-f8d7-4f73-8a5d-c6d5cf90fa77",
+   "metadata": {},
+   "source": [
+    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/cleanlab/cleanvision/blob/main/examples/torchvision_dataset.ipynb) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3381332c-2220-48ea-9b8a-8f974c4d5460",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!pip install -U pip\n",
+    "!pip install cleanvision[pytorch]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "dfcd1327-924c-47ae-834d-3b9d922800f4",
+   "metadata": {
+    "tags": []
+   },
+   "source": [
+    "**After you install these packages, you may need to restart your notebook runtime before running the rest of this notebook.**"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "50a8cd5a-899e-4a58-84f7-fb1ddf4bfcf2",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from torchvision.datasets import CIFAR10\n",
+    "from torch.utils.data import ConcatDataset\n",
+    "from cleanvision.imagelab import Imagelab"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7dcf23f9-c031-4bdf-b63c-5eb02003321b",
+   "metadata": {},
+   "source": [
+    "### Download dataset and concatenate all splits\n",
+    "\n",
+    "Since we're interested in generally understanding what issues plague our data, we merge the training and test sets into one larger dataset before running CleanVision. You could alternatively just run the package on these two sets of data separately to obtain two different reports.\n",
+    "\n",
+    "[CIFAR10](https://www.cs.toronto.edu/~kriz/cifar.html) is classification dataset, but CleanVision can be used to audit images from any type of dataset (including supervised or unsupervised learning).\n",
+    "\n",
+    "Load all splits of the CIFAR10 dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "79ae476e-b1dd-40d3-925e-d570c5237110",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Files already downloaded and verified\n",
+      "Files already downloaded and verified\n"
+     ]
+    }
+   ],
+   "source": [
+    "train_set = CIFAR10(root=\"./\", download=True)\n",
+    "test_set = CIFAR10(root=\"./\", train=False, download=True)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "42add096-2ad2-4e0d-8f41-2ed1f8558fbf",
+   "metadata": {
+    "tags": []
+   },
+   "source": [
+    "Concatenate train and test splits"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "2801693b-48a3-4797-8d41-1956b9259076",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dataset = ConcatDataset([train_set, test_set])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "287a2e90-9f9b-4e6a-adde-a454430889d3",
+   "metadata": {
+    "tags": []
+   },
+   "source": [
+    "A sample from the dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "e159ebe1-b5ca-46ca-abf3-9a5e77db0830",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(<PIL.Image.Image image mode=RGB size=32x32>, 6)"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset[0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "3a86c8ad-884b-425c-a538-569fa5b8e540",
+   "metadata": {},
+   "source": [
+    "Let's look at the first image in this dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "3a8ee57f-9542-4458-8eb4-7ebc96ed8cc0",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<PIL.Image.Image image mode=RGB size=32x32>"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset[0][0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f5f79ba0-3ef0-4ba9-8d85-726ee534bfc8",
+   "metadata": {},
+   "source": [
+    "### Run CleanVision"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "92c32767-ddd3-4881-853b-c0bee34cf94e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "imagelab = Imagelab(torchvision_dataset=dataset)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e0fe56ff-35d7-49d2-a99f-f867d1cfe376",
+   "metadata": {},
+   "source": [
+    "We set `n_jobs = 1` as CleanVision parallelization may interact with torch dataloaders in unexpected ways."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "71df74ca-63ee-48f2-bdd0-c335f17c96f8",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Checking for dark, light, odd_aspect_ratio, low_information, exact_duplicates, near_duplicates, blurry, grayscale images ...\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 60000/60000 [00:33<00:00, 1813.45it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 60000/60000 [00:15<00:00, 3862.80it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Issue checks completed. To see a detailed report of issues found, use imagelab.report().\n"
+     ]
+    }
+   ],
+   "source": [
+    "imagelab.find_issues(n_jobs=1)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "3e2c1bc7-3042-4e8a-91f9-2a7840e03255",
+   "metadata": {
+    "tags": []
+   },
+   "source": [
+    "Get a report of all the issues found"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "779adbac-7dde-463a-b6d2-4afc872f31b2",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Issues found in order of severity in the dataset\n",
+      "\n",
+      "|    | issue_type      |   num_images |\n",
+      "|---:|:----------------|-------------:|\n",
+      "|  0 | near_duplicates |           40 |\n",
+      "|  1 | dark            |           29 |\n",
+      "|  2 | light           |            3 |\n",
+      "|  3 | low_information |            1 | \n",
+      "\n",
+      "\n",
+      "Top 4 sets of images with near_duplicates issue\n",
+      "Set: 0\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUoAAACxCAYAAACiC/RrAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAjoklEQVR4nO2da4ycV5nnn7pX36vddrvd7Utjx5nEndhOQlZRYK04CHYQ9jjEyNEGyFhKwqLlC1osBaQIaZBQ8gEBYhF8ICMYLoIAGmUDIixaZpLJwobJzclgE+Ir3e62++K+d92rzn7o7nI9F/d5q1Jtx/j/kxA55fOe9/bU6bf+z/P+T8g55wgAAMBlCV/tAwAAgHc7mCgBAMADJkoAAPCAiRIAADxgogQAAA+YKAEAwAMmSgAA8ICJEgAAPGCiBAAAD9flRDkyMkIf//jHzX/r7++n+fn5msccGhqie+65h3bs2EE7d+6kn/3sZ5V/O3z4MG3dupV2795Nu3fvplOnThERUTabpfvvv5+2b99Oe/fupYmJifpOCPxVsBpxuUw6naYtW7bQkSNHKp/dc889dNNNN1XiMpPJEBHRxMQE7d27l7Zv3073338/ZbNZIiL63ve+R93d3ZX+Tz/9dN3Hc61xXU6Uvb299KMf/aihY0ajUfr6179Ox48fp9/85jf02c9+lhYWFir//o1vfIOOHj1KR48epW3bthER0VNPPUVbt26lEydO0MGDB+nJJ59s6DGBa4vViMtlvvzlL9Ndd92lPv/5z39eicumpiYiInryySfp4MGDdOLECdq6dSs99dRTlf4PPfRQpf8DDzywKsf6buS6nCjPnj1L733ve4lo8S/twYMHaceOHXT48GFafvX96aefpvvuu4+IiI4dO0YDAwOUTqcvO+aGDRto9+7dRETU09NDa9eupcnJyRWP49lnn6VPfvKTRET0iU98gn7xi1+8wzMD1zKrEZdERCdOnKC33nqLPvzhDwc6DsSl5rqcKKv51re+RX19fXT8+HE6dOgQDQ4OEhHRAw88QNFolL7//e/TI488Qt/+9repubmZXnnlFXrkkUdWHPPVV1+lUqlEmzZtqnx25MgR2rVrF33hC1+gUqlERIs/tfr6+oiIKJVK0fT09OqcJLjmaGRcHjlyhJ544gnz3x588EG67bbb6Ktf/Wrls5mZGero6CAior6+PhoeHq78249//GPauXMnPfjggzQ6Otqo0333465Dzpw54+644w7nnHMHDhxwv/vd7yr/1tnZ6ebm5pxzzo2Ojrq1a9e6z3zmM4HHvnjxotuxYwcbc2RkxJXLZZfJZNyhQ4fcN7/5TeeccwMDA258fLzSr7u7+x2dF7i2WY24fOaZZ9yRI0ecc85997vfdZ/73Ocq/3bu3DnnnHPT09Nuz5497pe//KVzjsfh2NiYGxgYcM45NzEx4bLZrHPOua997WvuYx/7WN3neq1x3T9REhGFQiHz88HBQYrFYnT+/PlA4+RyObrvvvvo85//PN19992Vzzds2EChUIiSySQ99NBD9PLLLxPRoia1/Nd6enqaUqnUOzsR8FdFI+LypZdeop/85CfU399PR44coe985zv0pS99iYio8mumo6ODDh06VInLjo4OmpmZISKi4eFh6u3tJSKirq4uSiQSRET06KOPVvpfD1z3E+X73//+Svbu17/+NU1NTRERUT6fp0996lP0q1/9ivL5PP30pz9dcRznHB0+fJjuvffeir6zzHJAl8tlevbZZ2lgYICIiPbt20c/+MEPiIjohz/8Ie3bt6+h5wauXRoVl0888QQNDQ3R2bNn6Stf+Qo9+uij9MUvfpGKxWKlyiKfz9Nzzz132bjcv38/ERFduHChMu4zzzxT6X9dcLUfaa8G1T9xFhYW3Ec/+lF38803u8OHD7vNmze7ubk59/jjj7vHH3/cOefc8PCwu/HGG934+Lh7+eWX3cMPP6zGfPHFF10oFHK7du2q/O/NN990zjm3d+9ed+utt7qBgQH38MMPV36+pNNpd+DAAbdt2za3Z88eNzY2doWuAHg3shpxWU31T+/5+Xl3++23u1tvvdXt2LHDPfbYY65cLjvnFn9u79mzx23bts0dOHDApdNp55xzjz32mBsYGHA7d+50H/zgB93p06dX61K86wg5B4dzAABYiev+pzcAAPjARAkAAB4wUQIAgAdMlAAA4AETJQAAeMBECQAAHqJBOpXLZRoZGaG2trbLvi0AQDXOOZqbm6Pe3l4Kh1f/7zFiFNRKLTEaaKIcGRlhBg8ABGVoaIg2bty46vtBjIJ6CRKjgSbKtrY2IiL6H//wD5RIJomIKBGLqX6xeJy1I2E9fCjEZ+7ykpPOMvl8Xm1TKPDPSoWC7iO2k+MUxX6IFp9CqnGiTUQUEn9p5NOKHMPEqOmX25WM4yuI8yzJ4w0wrry+1r7UdTDGDfJeQiQSqfx3oZCnX/706UrsrDbL+3no8SMUTy6+jxxfitVq4kkeo9GIFaP8HjvHr08hX1Tb5LM50c6qPoUcj8mC2MaK61JJ3M8A8SafjsrO2Mat2DS3s2K0KK5FscDb5bIVS2Jc4/hKRTmOiNGS3iZIjIar7m2xUKBX/vf/CRSjgSbK5cBJJJOUXJ4oxaRIFGyiDIuJUl78cNWXrTKO+Kxk9JHBISe4SFEHt5pUjCBU49YxUQaZ0KwglOcg+9Q7bvgKTJTLXKmfwcv7iScTlQky0WRNlAnWrmeiDEf0hBbkPGWfkLikRePnn+9eWTRsohT7ksdCRBSSaQ75IGQcr7yeIesBxfM9c+F3PlFebl/mdt4eAABwnRPoiXKZmakpyiYW/0rHonrTmPg5Hjb+WodD/KlDPd4bT35F+Rhe1H/RZR/fz0si/RcoyF8W+dRkicBBxpV9rL90UXGN9ZOO/4nS2rcaRz5Z1/n6f/W1sGSMK8Hk6ATFEou/bJb/v5q4+Cxixqh4mhHXQ8Ya0eLPONbO6xgtyxhVP1PrfEryJCLM+JMPgtZuxGbmkYixI1H+/QiV9b7LTnxWCvA03rAnyqoY9fau2q6GvgAAcF2CiRIAADxgogQAAA+YKAEAwENNyZz56VnKJxZrv6IxI5kT5ckcSyiXdZQhksK5UVNVkqK3LlOQQm6QMgpVrhEg8VFPGYglMgd6W8WzL+scgxyvKh8JUB4UBJbMuUp+0LMXJykaX4zD5f+vRpWwBSg1k5h1f7J0pmScvyzdqqPUp563nMyYDYsYNbqExDlYe5bfX3lOYeMcZW1lOWyVxq08bt11lNUxWkM6B0+UAADgARMlAAB4wEQJAAAeatIo52ZmKxqPqVGKgvMg73orzcWQDRwJfcKqjpXanPhnS9tphKtNPZolkT6+IPpokH+vp4g+iKZaq/4TDvv3uxpMT01RdCkOo4YfgSw4l0X9RPo1W1V4HeQFAuPY5FZyHPPlBTlGkFdCdXDpLvL+WFq2bIe1nlsOiRccArzoIYvHraJ05bEgX9+ts+C8+vrV8lIEnigBAMADJkoAAPCAiRIAADxgogQAAA81JXMW0gsUWzLRtXz8VDLHcmYJr+y+Y4rVKjOju8hxVCFxnV51zpfwsbR1lViyzkm4rhjHV08ypx5/zCDF+oE8+6qTOTIhcoVIz2cpEl18QSEW0y4/hRz/LBK1kny1x2g9hf7CSMv2YhXDWAmfkEzgqQ5qEwqLL5FsL+2MNa2vUCiii8XZvzfKj1T5Z9YX+9XHY33fLweeKAEAwAMmSgAA8ICJEgAAPNSkUeazOSovvYxejvoX61rWiqqROoyvuJcogP5IhpYoBRVjG+8YpItSVdsowpXFvLLIfmkg3raKbp0cRw+jhnW1G1zUWqgbZJxaDAcaST6bo8hSbFoLq8nPwoZGGZEaZUToecb9lAXSlpYYkeOI+xvEQdxc80UdixQ2rW2EE7m1b5Jae4AXPep8AUOidPQGxZO7zH/7wBMlAAB4wEQJAAAeMFECAICHmjTKQrFQ+V1vGyfwdskw2FUaZQBtx2c0S2TrlrVSa63g4sEY9XLSZCLIC/zW0sueekd7FUZ/TaTPqDfI6o4+Soax6pWgVCxeOn5LexWfRcqG0UNErmctNUq/iYMVx2URoxGxFGK0zlrfCMl4E/uOWNqiuOfhAFq2FUseEwwrbtSqqkYf7yqq5QbEaHHlGtBq8EQJAAAeMFECAIAHTJQAAOABEyUAAHioLZlTKNIlndzvgh1xhimGLEoXAncQMwGz4LwOo4cgbuBy31JkNtM/8lgCOGIbC/tVivsvt2+18h/pFSqtFSt9gnu9yZzq61cq6pcNrgT5fK6SMDQdrMW5WQlHGV8RaWYRIOFoupWr6yyd5fX3RUa69WRTkoXhssC8YBlIiA/MRKY4Pus7VF45Jq0YDRbHnqRQA5I5xWIhcF88UQIAgAdMlAAA4AETJQAAeKhJoywWChX9KohGWTb6hMsrF48H0QmDFFEHKXxVK+cF0J7UNsa44ZKyKdD7VnqV6qJkGHkOsnB3sc871yhr1XosCoXg+k8jKeQKFF4uGLdOQ90/Q+8WBeflIIYXwijDRYxnEDmu+i74X+KwFiB15ZV117BhYEPSzMU0zlh5P4ufCR1dtgNolFKLt8ZVMRngZYvLLOlaoVhDjOKJEgAAPGCiBAAAD5goAQDAAyZKAADwUGPBeaGSoAmSHAniHiSLe4MkVCyBOBLl40TL/NSCOJRYyZF4PM7aiUSC79c4liAr8inZ2dCdi6WVkzdBCs6l4/lin9VJ5lTf/8LSap1XGlcqVY6jZCQfZIbMSsaVwiIZIpM55mqJ/pciylFpryViv6iPt5Dn97wQ09c10Zxkbe2wbxSTy5ctTId90dajUKkoYlQkSPJGHMjkjStZiRkRx+K+BUkm+5z7a3kpAk+UAADgARMlAAB4wEQJAAAeais4LxUr2lq9BdzSUECuTGdpO3IVQ7lKHhFRJMpPRepMyiiAiFrbmll7TWdK9WlKcv1HalzW8UblsRhaYj7PtZu8UfxaKPI+pRxvW0XdhZJwj1Y9AjinWxqldG336D9Xq+C8VCxV3Lot1275EkTIKDgP+1bRNIrJQ+KzqKVjSk0yJPV5HaOdazpYu7d3g+rT2tYmjpePY31X5XezmNdaYnp+gbWz6Yzqk8vy7TK5LB/XyFOolRCsgnNfjHqKya1tJNAoAQCggWCiBAAAD5goAQDAQ20aZbFU0ReCmOda9VvysyCmvCExnyvNKMA4La0tapuOFNd/Uql21ae7ay0fp4nrmgsLXMchIhofH2ftidELqs/5C/wzaxyrrrOa1tZW9VlTC/8sbxmIeLQbU6OskVoMBxpJuaqO0jQgljFq1JmqGJV1sWZFoaghDrBSYyjGv37tRvwlk7yONyc0QCKirlQna2/dto0fmXGOfzp+nLVP/unPqs/46Bhrz8/NqT6lghhbnHeiSWj8RNTVvY5/YHznpYavDDqC1FGqHhxolAAA0EAwUQIAgAdMlAAA4AETJQAAeKgpmVMul+0X7JfwFXgSaaFcrdxoCLuJODeiSMoicNJmFfrfY+qzkZER1j7x9tuqT2cbF9hv330ba28TwjkR0enTp1n77RMnVJ9MhhfvSvGaiCiXy7G2PEdL2F8jEl3xpia973RafCLviVEAvOIWGl8iarVwzlUqms1Ca6u/GoQ3i+J6RIVRChFRu0gMdnZ2qj5NzfxezGf4fQgbBfKTk5OsffJtHUtHS6+xdn9/P2vv37dPbbNp02bWPvYfx1QfaWwSi+rvUGZ+lrXlix95EcNERBMTE6zd3dOj+siVL9MiZoMkcyyqp5+y8SLI5cATJQAAeMBECQAAHjBRAgCAh5o0ymqCmLtafaRG2dHBtR2prxARbd26lbU7DfOK0dFR1j52jGsua7rWqG2KJWEyYeiEp4Te+JczZ1n70KFDapv9+/fzfa/R+z569ChrzxnFvBdEUbpeEVJtQjPT06y9Lq6126YE13gvTlxkbUsn1jTOcKCRhELhygsJpl9tgDGSTfyabdjUx9qbb+DxSES0oY+bVbQZLwOcH+aa+Osv/Ttrd63rUtt0dPXycdv1uCN/GWLtP77xBmuXi1qL+/Sn/xtrb/z0o6rP8//6r6x96i2t4ctvuHxxIl8yVgpNc219dOS86rOhj593Oi3GTWt9Xr4YULJWlqwKAGiUAADQQDBRAgCAB0yUAADgARMlAAB4qCmZE4vFVhT65b+tX79e9dm+fTtry4JtyxFHCrnzCzrxMXRukLUvTnIHn3XdWiiXx9LX26f6vBl+nbWlIP/888+rbe68807W/shHPqL6bNmyhbVff/111UeuADk8PMzasmidiCgW44kIVx5TfW64QbjLiJX0xsf1NpZjk+506T9Lhoh/JYjF4xRecsC3iuJDUX4eW7b1qz533HkHa7/n5hv5PpK68PrCuXOsPTE2rvoMnj3F2vkMj2tXTqltejbypMYNN29Xfd6M8+MpCdfxY//xR7XNmZO8cP1vP6qL0tdv5N/f1/79FdXnpd+/xNpvH3uLtcdHeaKQiCgS4ccbjunrGY/wqWn7jfwe/PlPfD9ERNl5XpReDuvpLVoV6rW4ZOGJEgAAPGCiBAAAD5goAQDAQ00aZTKRqLz0bhWTS32xyyi0vuWWW1i7r4/rgmFjhTupd83OTKs+TcJJedeunawdN0wz1q3jTsuWK/ekcCtfmJ9n7XaxAh4R0fr13azdZvTZufNW1u7u7lZ9+t/Tz9pnRbH7wrx2RZeGIZGINnCQ17hTOGQXDSODGVEQ73evvzp/g2OxKIWXtHJrpb7mFu5074xi7JDYrKWFu9pHk/qa7rz9dtbOGfrxLSIm82JVQ0v/lzq1tbrjwiQ3pihm+fcll9X3c/fu3ay9PqU1/LzY7r13/SfVR+r6527nuYLxC1rvLuX48eVJ34Os+C6WRDj19W9S2wyeOMPahZK+/5GqELWd6m3wRAkAAB4wUQIAgAdMlAAA4KEmjTIejVA0ulSjFtKbxsRL6KdOnlR9nNC27v3Avazd1KJXS+wR9ZhrhbZIRHTzDq59RqJ+YwdZjzk5oWu+7rrrLtb+z+97H2t3tHL9iogoHOb6baGkNaJYjF+HrpSuH93z/rtZe/PGjaw9ePKs2mZ2mutVMzmtY4bi/NrEhQbcGdfGyJkoNyEoG/pftUYZxER1NYjGL2mUlitGVKx8ODU5pfq88PwLrJ3o4DHZsVbreet7eIx2rdF9NmzkulqiieuPiYj+TmXn+P0bvaBX9BzYtYu1P/TBD/HjTeoYjYnrMHRuSPWRq0a6vK6N3biBm+7GRe3ifFHHfnaO1zum52ZVn0iCx2iozOMpOa8NbDpDvB5zMqJzDq5Kl3RSjF4BPFECAIAHTJQAAOABEyUAAHjARAkAAB5qSuY0JRMUXSo4t8T6VrHKnCxkJiIaEcYOZ87wItGU4USezfJEwrouLZSnUinWbmvlRd7NRtKlVObidMlwPJamHdu38naqXY87fIEbZ4yNagF+Zownjk4Zq+BlclywLknzAKcTVjJZESnqv4Ux4eC9MMeL6K0ibOkQf1GsDkjEHdilG/uVIp5MVhJ5ZoyKlyLWr9WJwdNnuav9SZGU3FTQSY20TFBs0AXnqS7+fehcl2LtNR28TUQUEfcqndNJjHbxYoc00ti1TRtpTAsn/DdOaffykSGe4Dn+wkuqz/Q5HuuujSe+wj36+0zCmKRs5F3lKowkYjJsmK5s3sDPe258WPUpVH/Ha8g34okSAAA8YKIEAAAPmCgBAMBDbQXn8TjFlnQyS89rEfpPyXgpva2F95GaZcww8cwscP1n2tDH5GqOst25RuulUaGVLCzMqz6v/uEPrP3Ps7xI/TP//dP6WDq5LnP25F9Un9//y4us/cI/P6v6dInC+t337mHtNX3aaDgsTBO6Yto8uSQKbbPNXM9tT+qi/1MjXIuamdVFwu+GgvNEInrJuKWkjVvkPU+n06rPGlEsPnGeGzv09nAtjIjo4gLXoWcu6pcXmoV+19LFY3StYSIjdctJI/b/+NqbrP0/X+YGu4f//iG1zUf2cTPpi4bRzNGXX2PtPzzznOrTEuVF83fv/1vWXtevV6xMipdKYgltMlIUMTo2ya/nSCu/dkR6NcrolH4ODFXFZbmGGMUTJQAAeMBECQAAHjBRAgCAB0yUAADgoeZVGJeTLZbDtXQ9Tya0C01BOJDMTE6zdvst2g28IIpLFxa0I478bGyMC/Ctbdqdp7mZH9/g2bOqz29/+1vWzouC37ghRO//u/2sfeL0adVncnqGtXcJh2wiogMfu5+1t+zmrui69JiUY046qwufRy6cZ+3Bwlkxrk6CpDM86bH84sHluFrJnHgsdimZE9HnIaO2ucVwfxLO3qOjo6zdajhcRZt5HMzN6GTXvHCJn5rnfSYMN/BUO09aTI3pJNEr//Z71s7O8KTkP/7jP6ltEmtSrD02rVeNPDfIV5a86bbdqs/eD3yAte/Zx5M56YJ2DyrkeeSmc1nVZ3CcJ8cGz/Nk4lRGJ15HZ4QTlGFgXv1CRikMh3MAAGgYmCgBAMADJkoAAPBQoylGkmJLq8Ll81ohk7pVJKTn4XKR643jooB2YlxrJTfe9DesLYuqiYiKogBeGmnksloHuSiKgoeGtMuzNHdobuaa1quv8qJcIqLNwjjjgnFOC0Ie+a9//6Dqk2jj+3rpjVf5GHNaf8zluavz9Oy06jMjDBFCYlXGkXFt4lEUpgRydUBJuAb9p5Ekm5IUrbwUoY0TomEeO854cSItXiqYvchj9I3XXlfb3H43d8Lf1P8e1Scivh+FPI9Ja1XN2RmuZVsO53mxYmE8yY005P0mInr+//IXHtoNDX9CbHfnh/6L6tO5fTNr/9PPfszaF6d0gbzMZWSs76bYTl676Sl+XYj0CqkxI0bDVXNSqajj43LgiRIAADxgogQAAA+YKAEAwAMmSgAA8FBTMqe1rbUi4hcKeilI+dmssQzl9NQ0a48JcfqFf3lebXP2zFnW7hFLZBIRdQnX8+ZmXhRcdoZoL8TfXEaLytIFOZnkRerhsE4svfYKT7oUjQJuF+HJjqHpCdXnwgle6DwoCoDbklqA7xUuz5mMdseRibiMKPidNpyBIiKBFo9rl6fqlxAKBf9ywatBW0cbxZaOzYpR+cLA1LRernZaFC7Pij6vvfj/1DYjZ3gisKd3g+rT1c3doNo721k7IZIwRETpDE/YZYyXLSLC1T7VysctGkmtk2/+ibXXdmp3rVCRB//Q+fOqz8gY//4ODXKnLJlkJSL6mx03s3be6e+HTKLm0zxGZ8U8QqRfcmhqalJ9qCrJiGQOAAA0EEyUAADgARMlAAB4qEmjbGtrpURiUUcpGCvR5fL8BfhCTvdZXiGvcgBC+7ognLSJiCbGuFYXi+lC0tZWrklKl+ouY3XHNuE4TcY5retay9q9G7j2JFeRJCIaPce1RLmCIZHW/F5/RRcxzws9SmowHYZGmRN6oyv7zSnSotA5argJNAkDEXn8RNyPI1q4OgXn7am2SqGxqVFKUxbSemwszeOrQxhIzM1qQ4ahM1ybGzyl4yIhCqBb2nn8da7VcZISzvzOuDfNQotb18Nd7RdEAT0R0bAwmYgb36my0N/PvH1S9ckIcwr5QklrZ0ptMzXFNd+S8b2TGmWQuG5u4i9oxJLGSxFVl69o7Pdy4IkSAAA8YKIEAAAPmCgBAMBDjRplW6WOsGjUIMmVGVPtujarbwNfOfAmYXiRyWijB2leIU15iYgmhXHByLDUOnWtltY1tUYkda600A2XNdtq5oRBa1NSGxirbQyj1z+/9WfW7uvjNZL9m7ghARHRvKgNHTeu1Wmhq8qV/fr7+9U2LcLgVmpI8rN8XtdZXgk6Uu0UX7ondozyOOg07nlp80bWLoq602xWm9HOCCPmSWMVxovj/LMpUQs4eoFr8UTcxIGIqKVZmwY3iXszP8Pjz4rrrKqdnVZ9SOqAeW1yfGGI6/H5HP++vG/PJrVNborX9p4/r/MSY+f5tZBzQHtXSm3Tt5nvK2qYYoSq6peLhoZ9OfBECQAAHjBRAgCAB0yUAADgARMlAAB4qCmZ097WRsml4lZLKJfOxVTWxbGhTv5ZSWxjCawbN/IEkLUK44xwgpYJCikGL/aRSSItpt9ww3bWlsmdiQltZiGvjbXvXI4nBJZXt2RjX+RjS5OREcOkYFYkkqZnplUf6Yi9fj0vULaORX5mOZxXO9zHPKs0rhbtHe2UWDHhyOPN8GNQcSyTlGXD6GFdD38xIZPtVX0W5njczkzxxNv0hHYDnxFJSmkqQ0TULJKSnR0p1h4eHlbbSLOKkdODqk92gSddEnGduJRmFQVxzZ/7X7/S+y7zfS8Yxi1U5DEaFSu6tnfpRLFcYUGuRkBEFIldejaUq0GuBJ4oAQDAAyZKAADwgIkSAAA81LYKY3NzxQwziEbpSoZhrTB2UNsk9CE1lbk20tyiC7g7Um2s3dPTzdqywJZImxtYWqLUJGV72ljhThrjWn2kRimvC5E2npDF+LKwnYgoKgx129rbVZ+Ojg7WlhplWxu/lkREra3cgEMaGBNxjUiaGFwpmpqaKdm0rFFqvbtUlPqjvu4yJlWMGvdK6pbWiwgtzfwapoTZRk8fN/YlIsoIw5KpSb364KyI44Us3yab0Zp+Op0TbX2/SmJFzwWntcRqs2arXTReIInGeVy3r9Hx1ixWhWxPpVi7W3y/iRZNm6tpMeI4Gru0b2nivBJ4ogQAAA+YKAEAwAMmSgAA8ICJEgAAPNSUzIlGo6qosxolgod0wseJPnIVQ1MoVzkhy12bJzEiEf43IJ7QRdTJBHeGbjcSHzIRI11zrKRWOs1Fb5k8sZAiOJEuoJUJAqugtrmFFx+3tmtBW56nHNdK1MjPrKL06nNYKU5Wk0gkUkmCqRcgiIgi8jr7HeDVvTFiVJegBxhHPKZY1yye4IX9iRbtHpQSiRhZpB6N6uehTIYnMvJ5nfgKidO0POvDYsUCebzWSojNrU2ibcSxSObIFSqtFxrkviNGjFYnSPOGC9TlwBMlAAB4wEQJAAAeMFECAICHmoSkUqlUMQgIUnBeLgUoSjf0Hr2NKFI3NnFizg9J/Sesdc1wmGsYsYQ2epDu0Wu7eVGwNFkgInJOGi/oA5afhA2NUq1YKXSZaETfvpDQoyKGlhMW+m1IqE+We3koLLfRMI2yHES1azyFQqGim5UsU4wiPy6r4FyZYEit09TR/XEtP4sIrV1eYyKicJzfv1jUWIG0ievHqQ6u75W3cMf2xZ2J3IDuQWGpUVo3XXytIiImo1EjnyDjz1jRMyJ2Jp3eLU1fa8DGAVf1CZf0fi8HnigBAMADJkoAAPCAiRIAADzUpFFm0umKziJ1HOszq45N6jRymyCapXNaezBr5rzj+PclhRmp+UVjhrYoNBdLT9G7qb2PtY26NBH9t1D5KYvr4AzFygm92acRFa6SRrkwn6bikg5pxajW0XXcyO2kubSsBbYIolEGKOG07oR3XNkjZOiEUqeWGiCR1iTtePPEpKETqhi1NF9fTAbRKK0QrOpSwiqMAADQODBRAgCAB0yUAADgARMlAAB4qCmZs5BOV4TtIImaQC7oUrQ1hF0p0lrJnECJmTrwJYksgds6b4lV1O0bu55kjisGEL0D/HutyaZ8LrjhQCNJp9OVFQate6eTOf6kZJAYDYJvu3rHrSdGqejvoz6z+si2fDEhwLhBYst7bHWMW0uM4okSAAA8YKIEAAAPmCgBAMBDTRplLnvJINQq5pVYferRe1Sd7urIkSY+/SeI1mihdFerj2cbU4MRBb7OMgaoY9xadSS5yuSVIp3O1KRRWsXjso/WyAMUfQeK68YEsm+cel94CLKdKm6XZhZ1aPFB+jRGo8QqjAAA0DAwUQIAgAdMlAAA4AETJQAAeKgpmZPN5SribZBkTj2Ctr0Ko+zjHTYQ9RS6SoKcY6PE9EDHK5I3yikowL7rTVBVk8sHF8obSTaboVL58smcIMFTT2F4kDiuJ+HTiBitl0DxV8fLC0H61BP7tW5TqCFG8UQJAAAeMFECAICHQD+9l38iVNfGXcmf3nJxrr/Gn95BjEivpZ/ey+/RrtY7+JLl/VT/nLq6P7392117P73NTjWNEbTPlfzpHeg+uAC9zp07R5s2bfIOBoBkaGiINm40VgFsMIhRUC9BYjTQRFkul2lkZITa2tqu6l84cO3gnKO5uTnq7e1tyBOqD8QoqJVaYjTQRAkAANczSOYAAIAHTJQAAOABEyUAAHjARAkAAB4wUQIAgAdMlAAA4AETJQAAePj/7HejMpPK3VcAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 400x200 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Set: 1\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 400x200 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Set: 2\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 400x200 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Set: 3\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 400x200 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "Top 4 examples with dark issue in the dataset.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAoAAAACrCAYAAAD2OS3/AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA7oklEQVR4nO2da6wkx1n3n+rLzLns8d6crL3HNsnKieXYu+vYm9hrE3uJAFmyuSZ8SWKCFEUKQfCaoHATEAQIPiA+vAKEEQiQIr6AUIQdA+aSRIT4jfEi34iQ/PqW1+wax97ruc5Md9f7YffMmef/PGd6Zvdcdrf/vyjy1kx3VXXV0zV1uv/1rxBjjEIIIYQQQhpDstUVIIQQQgghmwsngIQQQgghDYMTQEIIIYSQhsEJICGEEEJIw+AEkBBCCCGkYXACSAghhBDSMDgBJIQQQghpGJwAEkIIIYQ0DE4ACSGEEEIaBieAF0GMUT7zmc/IjTfeKIcOHZKXX37ZHPPaa6/JPffcIxMTE/IHf/AH6rvPf/7zcsstt8jNN98sv/M7vyMiIq+//rocOXJE3ve+98mBAwfkr//6r/vH/9Zv/ZbccMMNcvXVV2/shZHLhouNwc997nNy6623yoEDB+Sf//mfRWR4DH7sYx+Tm266SW699Vb5pV/6pY29OHLZMEoc/tM//ZPcfvvtsn//frn77rvlhRdeEJELG/P+9E//VN7znvdICEHm5+c39uLIZcPFxOHp06fl0KFDctttt8mtt94qf/Inf9I/56d+6qdkz549cujQIZXXyy+/LIcOHZIbb7xRPvOZz8hlt7FabDBFUVzU+Y899lj8yEc+Yv49yIkTJ+I3v/nN+Mu//Mvx93//9/ufHz16NN5zzz2xLMu4uLgY3/Wud8Xjx4/H48ePx2eeeSbGGOMbb7wR9+7dG+fn52OMMf77v/97PH78eNy9e/dF1ZtcOmxlDD766KPxgQceiGVZxhMnTsQ77rgjFkUxNAb//u//PlZVFbvdbrz33nvjv/zLv1xU/cmlwWbE4TPPPBPfeOONGGOMTzzxRPzQhz4UY4wXNOY9//zz8ZVXXonf9V3fFefm5i6q7uTSYSvjsCiKuLCwEGOMcX5+Pr7rXe+Kb7/9dowxxn/7t3+LR48ejXfccYfK6yMf+Uh87LHHzL8vFy6rJ4Dz8/Ny//33y/79+2X//v3yxBNPiIjI448/Lu9///vl4MGD8rGPfUxERF555RU5cuSIHDhwQH7wB39QTp48KSIiR44ckYcfflgOHTokX/ziF+WJJ56Qw4cPy/vf/375xCc+Id1ud+T6PProo/LQQw+JiMgDDzwgTz75pPkLYNeuXXLnnXdKnufq8xCCLC8vS7fbleXlZZmYmJBt27bJtddeK7fddpuIiFxzzTVy9dVX9+v+gQ98QK699trxG46sG1dSDP7Xf/2X3HfffZIkiezatUt2794tTz/99NAYvP/++yWEIHmey2233SbHjh0bvxHJRXM5xuFtt90m11xzjYicG8tWYudCxrz9+/fLu9/97pHrRzaGKykO0zSVqakpERHpdDoSY+yfe88998ju3btVPjFGefLJJ+WBBx4QEZFPfOIT8thjj43eeJcAl9UE8IknnpDdu3fLCy+8IM8//7wcPnxYvvOd78hP//RPy5e//GV57rnn+q+4fuZnfkY++9nPyvPPPy/33HOP/Pqv/3o/nzzP5ejRo/Lggw/K7/7u78pXvvIVeeaZZ2Tfvn3qse8Kv/ZrvyaPPvqo+fz48eMyOzsrIucmdDt37pQTJ06MdC233367fM/3fI/s3btXbrjhBnn44YdlZmZGHfMf//EfUpalXH/99aM2EdlgrqQYPHDggPzd3/2ddDodef311+Xo0aNmQrdWDM7Nzcnjjz8uR44cGakssr5c7nH4F3/xF/L93//95nOOeZcXV1ocnj59Wg4ePCjXXXedfP7znx8qtzpx4oTs2rVLQggiIjI7O3vZ/UGcbXUFxmH//v3y8MMPy8///M/Lj/zIj8jhw4fla1/7mnz4wx/ud/quXbtEROTpp5/uz8Yfeuih/ixdROTHfuzHRETkm9/8Zj9oRc7N+gePW+E3fuM31v1aXnrpJXnppZfk2LFjsrS0JEeOHJHv+77vk3379omIyMmTJ+XHf/zH3eAnW8eVFIP333+/PPXUU3LnnXfK7OysfPd3f7ekadr/fq0YjDHKT/zET8hP/uRP8od6i7ic4/Cpp56SP/7jP5ZvfOMb6nOOeZcfV1oc7tixQ5577jl588035Ud/9Eflox/9qOzZs+eiy7pUuawmgO9973vl2WeflS9/+cvyuc99Tj7+8Y/LDTfc4B67Miv3WHnMW1WVPPDAA/Lnf/7nF1SfvXv3yrFjx+TQoUMSY5RTp06Zx8Rr8aUvfUnuvvtumZyclMnJSfnQhz4kR48elX379kmn05Ef/uEfll/8xV+Uu++++4LqRjaGKykGRUS+8IUvyBe+8AUREfne7/1eec973iMiMjQGf+EXfkF27twpP/dzP3dBdSYXz+Uah6+++qo89NBD8qUvfUl9zzHv8uRKi8MV9uzZIwcPHpSvf/3r8tGPftQta/fu3XLy5EmJMUoIQY4dOyZ79+69oHpvFZfVK+Djx4/L9PS0fPKTn5SHH35Ynn32WbnrrrvkK1/5Sv/R64qu4NChQ/I3f/M3IiLyl3/5l3Lvvfea/A4fPixf/epX5dvf/raIiJw9e1ZeffXVkevz4IMPyhe/+EUROad5OHz48NAgH+T666+Xr33ta1KWpSwvL8uTTz4pN910U//pyoc//OG+loFcOlxJMVgUhZw6dUpERL7xjW9Ip9ORW265ZWgMPvLII/LMM8/IH/3RH41cR7L+XI5xeOrUKfmhH/oh+cM//EO55ZZb+p9zzLt8uZLi8M0335S5uTkRETlz5oz867/+q9x0001rlhVCkLvuuksef/zx/jX9wA/8wMh1vSTY/HUnF84//MM/xFtvvTUePHgw3nXXXfFb3/pWjPHcap+DBw/GAwcOxI9//OMxxhhffvnleO+998b9+/fHBx98MJ44cSLGGON9990XX3jhhX6e//iP/xjvuOOOuH///njw4MH41a9+1ZT7q7/6q/Fv//ZvzedlWcZPf/rTcd++ffH222+PL774Yowxxqeffjp+6lOfijHGePLkyTg7OxtnZmbijh074uzsbIzx3IqjT33qU/Hmm2+ON998c/zt3/7tGGOMX//612MIIR48eLD//+effz7GGOOv/MqvxNnZ2ZgkSZydnY2/93u/tx7NSsbgSorB+fn5fvzdd9998eWXX44xDo/BNE3jjTfe2P/8z/7sz9ajWcmYXI5x+Ju/+Ztx27Zt/dj54Ac/GGO8sDHvkUceibOzszFN07h37974sz/7s+vRrGRMrqQ4fOqpp/p13r9/f3zkkUf6+X7yk5+M11xzTczzPM7Ozsa/+qu/ijHG+OKLL8bbb7897tu3L37605+OZVlebJNuKiHGy824hhBCCCGEXAyX1StgQgghhBBy8XACSAghhBDSMDgBJIQQQghpGJwAEkIIIYQ0DE4ACSGEEEIaBieAhBBCCCENY6SdQKqqkuPHj8vMzMzIJrOkGcQYZW5uTvbu3StJsrF/TzAOyVpsVhwyBslacCwklwLjxOFIE8Djx49zz08ylNdff12uu+66DS2DcUjq2Og4ZAySOjgWkkuBUeJwpAngzMyMiIiE8L8lhEkREYlSwlEte2Jsq2SQFA7oQXrJ5hEiJDEPTQVlROFfR+vNYIvGuCQin+3HyEayUsbMDddIOP+XTYJ//V7AX8Pohb4ef1GXUukyZBS/9QApTONfc/ZewLqbNBzv+cDjR0miz8IzYqWvtXKuNUB7hDi8ffDavXKlWv0kVpXMvf4/Gx6HK/lvn2r127aV6WE0Te1f3WkGbdTTV1OVOt11xrketFlV6TE4hAS+x+N1WkQETpEA420r0dfWynS98mDzbKc6j+lcF7J9QvftdMTfEpEcsoVqmdsc48e72zCusTkitG/h5FHCPVgM5FlUUf7Pm91NHQvb7dU4xP73WiFvwU8+NEqS6obNstzk0e10oZSaexeSWWanHb1eD9K6L9JUx93sO6dU+vQcziVEzi50VHpm14RK5/B0bMJRxOGMpFfoqCiWdHpiCu6Xlm2/HgRWCe2TwLWa4BeRqoLPBvqxqip5+/W3R4rDkSaAqwE22Z8AipkAtsUy7gTQKxyDa3iVbRmcAK43dhKxPpOm2nJX4jBJ+hNAU+6F1GMDJoAjTF9qz7KTNxigoh2wQjLeBND8KoqYqmKetlBIOnliDnZMq58AWpxyNjgOV8fC0P83/hGCE2b3M+y6Cr+2eZg/COom+zXpc5/VpeHaatIiIvjWKYVrzyCdR5sH/mzWTwCHp0WcCaD5vuZeEScunYI2dSwciMNRyq07xvS3E8u1edRMAEfJsy6Nf2SNkie+DjVpZwKIn+A5ODbWlXHuM53G8DfnOBNAE3jemDtCPHARCCGEEEJIwxjpCWD/4Nb/SAjnHqPGSj8Glu4+ewK8xqjCvEqHTKfbkzotItLr6TyKzg4oQz/WjTCn5Svg9ScM/vXh/PW+0SRDngCO8ldP3fbXqftXG8QV5FHC+6RQmWdetfUyZZrHHKP8RTf8CYV5Fe09wTBpzBOfmOKftM6rRsgDX3OMtCU5HBMH2mezw7CMoX9NJdQrda6/KvUbExynAjzRyL2OgY/wVRI+0woJ1KvuSa6IpInulwz6NoU8Q2Jf31ZwbR3o+xLCpfCeQ8Dr7RxjDsd580DEeW0G9cDfhgjX6j2Fxjfeg+WMEsLrTQjJ6pNoGKOy3Hs6CzKBUl9QBq/4y9K+CMfLTCBmygJiHZ+8OnnWvbmZbOsypnI9dTlRLptz8hbcU3BtGcTQJL4eF5EC7ttuMfyReZZrKVzwfkvwJSU+NUwxtu09hnIF3cajD4Z8AkgIIYQQ0jA4ASSEEEIIaRicABJCCCGENIyxNIBl+ZaEcG5lb6y+rb7Lg9XvZe1pfX5yQqWT9hmV3n2N1vOJiJx+a6dKzy3j0mZcfYzWCCZLsq6E0Ra4bhK+rQnqhzQga/IVFJgHnJMG1BOhHVF9nhisWIa9NOdaUYt4ATYwZiUb5IF6R6cSNk8oOIWaoHXMKP2oTtlkAVYZq1UNIOg9S2d1dgJ66NJoePS45a0eDCB0xCOMdrNmxaKIHR+NThAFe8b9wYI6VLSsKFATiLYXYiSAUhZgC4K6sogx6mgAUUNas6oadVYiIgXk0QUbmM0nyErP48pYbxwrQfMXsW969dZBaFlkbKAgT9QIu7eqsZ5C6yD9O39mTq9BmF+yukLUAGLB6ZT+vpd42mXU0YImdlLXq9XSsZyUNs/lZe1+ksK14ViAdk/nDoLerdb4dw18AkgIIYQQ0jA4ASSEEEIIaRicABJCCCGENAxOAAkhhBBCGsZYi0CkLEXCOUFirF5WX1XJK+bwKtUb+sR0QaWLVC8C6VTvNHn0yjt0nvG9Kh0C7ptXL1ImVw4odEaht0j9IhCz044j3LXmpyC4RsPckbaoG76tVlmhoSqK2J0FHGNuF+ZhrsXsXYT72OI+v/Xbg+F2caHCPD3z0yFmwJsswB8sDY2ge6WtCxqDY4gV8IGzBsTmCX1tTZv18dHrexC+g7+uuTZcFxET+xOCazpacIOBT7AUuApLnJ3yoE17sPCgxIUnI2xxiK7OdSbq5z6CLcmSwe82nxBWhxZcfOGt1cKFLbhQCBd9lE4sYzm4DZs1zZeh34uIVFAvNCTvgOv5/LJOu90N49DkBBhB57DJhDPum8VFhS5oZkrPcSbhptvWtte6DAtWih7cEHAKLoASEUnA1Fr9doyx8pVPAAkhhBBCGgYngIQQQgghDYMTQEIIIYSQhjGWBjAP030j6FK0EWNR/l9zfG8ZtAABiutpTeBpx4ixWEYj2xyOWBpWZRGxJqPk8ibG2Bd9YMS4BsIyRDsmYsQyuBn3+ZN0nmioKmjsi9pUb1Nw0HCBeCqFcyq8P1xHVdzwXh+T4U7kjl4EzWFRX5eB7rYCDZDnQ5qiJLJGi+Rp+lAnlGerY4FnWLuRVCH09ZRFhW1u9YuokzJXB33tGRljbOepHk8TyAObFM2nRUR6IMhrgaYvgzzMtTohiOa5EzDsd8Gzd96JmKvgHpyYAJ1Zoc10TRy7twYaFA/XBnugvHWw9XqbG4Iick7TuxKHeA94el/UxUXQGbcn4Pc12IsqO6i/M+LKYVX2GeLxLiJyamEZyhxuVi9ijbFzGITyDEpx9I4LC3B/wG/HtpYO7l5HHz+PvwMikoHQturVaNQ9QTCOJ4P3thG1rw2fABJCCCGENAxOAAkhhBBCGgYngIQQQgghDWMsDWAc2Hh6ArQCV+95hzn+2Jtvq/Tk1LRK73jHLpX+nzc9/yxdDnpfSQJaEPStii2Tp9EFGv3AVjg6XT7ENf69WYTz/xMRydIRQhi6NzM6Oa3b8DSA6F1l/LLAq8lsDO/oMjIoJwX9FWp4qqqAtOdbBZ9BEu8f9PBz8zX6KtCsJPV6nKrUdccWDgnqIZ1+xY8G9Eyj+BuuJ1lS9cs0vn+uf9zwNsL4Kh0fxDqM1hU1gV6/gB4P9Z8JePRhWBvvPBETHznE+XKhx995Rzy3LQf9VktXFO9hjFGjY7XVEhwYjGemI3DEmg7e1q4kd4MZ1AB6+mcL/j7Ct9BurpwM29r85mIcjn9voi9oieJLrJhXT8hjaRHG6BJ8Ap3HYRlo+Can9CA0t6TXQnSWQTNYTZg8UVaZQOVBZiv5NK57EOnBPVMsr+aBPo3D4BNAQgghhJCGwQkgIYQQQkjD4ASQEEIIIaRhjKUBLEPseycF0Ojs2rnTHH/i9CmVnoZ32Ttntqn0G8f08SIiErX/TxpAtBImdR2jficfBY4XETF5gCYwghbL6FyoEdxK8qTV1+mhd56ng0EtFMpHKtT3Obq4FOI9R4M00VrUNmja0szeaiaqQPeFV2L28XTiEK8Ny8Bal46OsChRwwMHgHdYAv6E6Hl37hTYDxa+N7oYJ49B3z8RkeWlzmr+m+wDuC0LfU3dyUJfTel4f5m+g3jA7729bDOjtYQyjIEY9JM3buH+01AuhjmOhdERiWE9KjwH+mq+Y8fo5RboBvFau7hPNu4/6wkxcRxAXRkcbnOwbTyAs2XrhqOus0aba44X258d0672gq2mr+b3EJsZN4sWkRQ/M/UcrvlzrfLgoG4X4g40gacLG4dZrsecJRjHliFPjO0p57cEN9zOZiAuU71uoXJ8RTtLiyqtdOtjSC75BJAQQgghpGFwAkgIIYQQ0jA4ASSEEEIIaRicABJCCCGENIyxFoGEMCkhnDM2XFrWp77wrf9rjq8CmCR2Tqv0W995Sx8vUyaPPJvTxxRnVbos2lCmXlgi6ZLJU5J5nQbRdii3D/1eAphPk00lT1JJkvN9gosHgv2bxvtMfZ/p/u31bP9GcOdM8uGLQlBsjUa/IiKdTmfoMSiMLkBg7C94qTPI1XmkibNgAc2iYdFHq63vuRzqudyx95xdBDPcTNYzj+31YFHDYJ6bvAgkDVl/EUgGZslFYfulBEF6kmJfGsduAy6MQfNsXBWCYe9pwzGEetCOOZjmmzh2F4EMN+2d0uEjXXS+FekvNlwBvW1RW2+19u7VQrWGLywoPQPjIWbSW7EIpKqqNY2gvVHPmssba2tIe4tpMJZ1jKQ5LnCqWVQm1vi8zjw6Cpo626vFMQcvtbOs5ydFz17rVSluJKGvtYq63mkO35vFgiJpG9qrrY/pLOh6mQUyIpLCgqZkwDi9cu6nteATQEIIIYSQhsEJICGEEEJIw+AEkBBCCCGkYYylAYxxm4icN16Oe9R3Zdm2JyTaxLkqtLYqgsFhyBZsHuEllczautzY02UEeYf+PtjNmKW6Sp8D78wD6nFAb0Ab6K0lSRJXU7fyXR2onUrQCHwD8PR6mdGoVEPTqNerTJzWl1vCOVlih4B2qs1PSxCTpSjxwvZ0TJzRjLsE0RbqXLBtPIoB3VBVWrPUjWSxt6q9Qndgo+cTq7UrC6hvQONbC/alTcPxFd4Lnp5L17UHQrZuUlMvLwTRpBeubbJdf79NQve3IX6m4KcLTX89I23TXrUjeX09B/PooVBxk8GxzxsL62JoFIzfOEpR4V7udbWmrfLaCeuR1ui2IR4mMGBEpD2j9XtFF+YfUCYMe+c+y3Ds0+kcxqmJNpSJ97lY7Wns6GM6C3pOk2dOHA6RetfIJxV8AkgIIYQQ0jA4ASSEEEIIaRicABJCCCGENIzxNIDVjEg4pwFMwqz+TrbZ4+WMSoeIuhedrkrrv9apXlTpFvrqJMfgjGuhzH0mz1BeDxXV+sUAPoHoOSTi6B3JphFj7Os3UCeHflKjkIIOzvNdQl8q9ArEeqCGbZQ8EcwTlTOefqdO05OAtsbzSAyg0YmQRs2f0dI41+rqfoZg/cmGe5h5x28k80udfn2SEUQ3RotZDm/T1NNAQTkBhEB4hokfp42wV9BDrAuehlmi8whOsxdQz7KAmINfnUmn+SZAaJpU+n7L0NMPxmjj8SfWS9D4FeI5wRtLUEe4mgfqEDeDkCS148jF4F2R+QzHYND8RdDBGf9KEQk1mj8EdXG7t1sP4fleodKLizqdwB0zMW1FgJPb9GcQhkZ7jHrH6Iyv1SLOc+C+hnMyp20y8A7MJ1ZvKtRXD4NPAAkhhBBCGgYngIQQQgghDYMTQEIIIYSQhjGmD+C0yMp+vck18O20c7xOB/D9M3ZZ3vvyoPcC7vT+Ew74b52n7NZZhm+bPJPsVqio1gRWFegJAmj+aAS4pQz6AKK2ytOC1e3LO4qGZmlJ72+LWkPU/OH3Xr3q9r+09cIbqr7emEeWaU1L4shFFufQj1OXO7FtUqVL8I7DfW9Fxu8Dr73sOf6/N4OijP0ysWxPs2O0Y+iVhzogx4MONUuoAUxxD2cYYJdAEyUi0kMtIlxLr0LNqP4+c/YC7pb6oA6kp6HMtuNXh16sqMMujfdivechRpSVCUKeJgfr8VgN1KO3BXsBp+mqBjCae8C5AjwGZY4j3EdG81rjcWnmAU61sPdQB4e6wR6Mnd85pXX7IjZGMADaE7Cfe8tOh/BaEzRBrdGVujEEYtQK7ocW1Cs4vpltXAsR1/h3DXwCSAghhBDSMDgBJIQQQghpGJwAEkIIIYQ0jLE0gBJbInJur7skwH66qJMTx0YJdBoh6uJj9LzS9L54ki7CAbAXcPW2zrN6zeQp+Ss6y/yDKl0tf0DnUcxAner9tOrxzthkIdNlSlVVfbESes55+1+inqxuz8xR/PVQ81eXp1evotCarFotIur5RvAWxHpjCVgHEZHOgr7HjEddDtc2AV5ZTmxjP6X4tyfKdUbw9Uu2UgSoqN9jtl7fObzfRETKmr7EFkNNWqd0dHG4BWvN/tIR+tHzOetCpssFpsEz0+k7OEWmcl3uYokeh5iDzdP0i5HUgobS5CASIW4Hw3QrNIDKBxB9H51bwoQV3NuoK/V+p1DDh+1qNH8XoFVGUItXQCx3nbgNZgzWebRy6G+nAysQSWYtHPeHj8ner7y5j1O4HyDW0ZtTRAQtlYuButMHkBBCCCGErAkngIQQQgghDYMTQEIIIYSQhsEJICGEEEJIwxhrEUiQsm/mXIkWiqPpqIjdKNyKlnGViC+71YcMlz6HoDdaDsYgVKTsvqpLyDsqncCm0NXCnfr73nanllpMHwMscDGNoTeNPl8R+IDzc48QQl80XC+ut58ljnB9EE+APzExMTRP75y6euU5mDI7C0V0JjXpcxUZekgXNiufPz0nhp6+L8tKx3YB9+1VrR0qjSJmEdvmKDbH9htlIc5gFlvpzV7T9S51onfv+osKxkuMayPoh4VLjkF1jOPdGxUcj2kRkQoWgXRh6FuGdUeemXQbzklgaKzgnKLEccBkKcEYaWuwyR3tvT1m4IMLiYOLZTBOzMIppxEwRtIMFi3gqiBMyyhjHaRtDkPP98AzRhr36z6BRae4sEJERFpQDjqhJ7C4zdxjTvuBUXYOi+pSXPCCK6LEzi8Gr3+URTcrcIZBCCGEENIwOAEkhBBCCGkYnAASQgghhDSM8YygpZIVzV0VQcjhmDjj/DKgxs/bFbqOug28jf7Aex+udVBl+f9UOkmfg/R1Os/elMkR9SUVaA+jccXeAtfQK4UY+4KbiOan3kbjCWqloG/geDdm4KOyxA3PdS4pmpA62qoApqGejlaVgWbJzv2TQEUTiMOls1q7u3x2weSRV/qcEsQx3UWts80XtAn89t07bL3QZBf0bGiO6nWBvbeHH7+RDOpQMYI8D2urDRpfA4nn5BmkE4xB+N4xDrcaPtzMXqfRZLYoHeEU9HUXTHsX4adjIrOdl0AjYpOibipAPCXBMUlHA2NMo7zLc1IGqsH2uZDfs4slrvFvEfeeqDMgT1K8970i9YcJxF1VoxH0vraytfHMpf3f+eF6vMm8pb927o+lnt5oIslB2w8N1ANj/Sxx9NBQjxbkWYFROmoCRWwfdHqrY/IoJvqrdSGEEEIIIY2CE0BCCCGEkIbBCSAhhBBCSMMYSwNYyapiIInaF03ijHOGNm+KQfvtubtVXySoWRGxm91LAp/Be/zYfQ1O0L6BkrzTZBniLvwE0tT8rRdVVfVFI+g5lmWeFnW4ThQlKV5Y1mn+6ryXPH0fajUwbTavT1FXasU0FcRyWsC1g8dfcEQ+ZURPy+F6nMW5eZWenJo0ebZa4K0JukLUaUbHsw49y6pBPdoY3lfrzwhlDxUw1nureXkE0JXmmU634FYwHm9i9VoJaJUreEaw0EHNk5Mn6pNAJmjUkI4+NgVdYIX+daABLCHOS8fED2+XgPcwtK83kiDmHt1sBq7BjEHuPTFce2rzcHLAQ9C/tE6X7el7IY161Yiel6BFLYsRfl+hfxeW9Xxkz5495pSlU1rvXMD4ifo99Ans9axGdmZaj4+ow+12db3y3E7TSvASLHur108NICGEEEIIWRNOAAkhhBBCGgYngIQQQgghDWMsDaC2HNqmvgti35/HOA2fOPvfXjSgi4ng7eP4QVkNAmieqrMqXcYXVTpJZk2WSbkNPoGmxb2A3bn3VuqYLh/aE21JzmtEOh2tl0gz2645+D2hRgm9mxJHR5ik6GM3XL+HGjfPIgy1M0YHh7aZUIfC0RWW4IeGmsAC6uXtd4raKAN834U+OPHW2+YUswcp6Fqu2q73106c/YSLru6nwfbbin1YxwE1oME0fP0FBIixAjR9S2jNCo3SdobC6ZaOwQlI90yZqLWz9xvuWdyBehmtXWKvvQUaMJSEppAHSsDQr1BEjAYsQe9O1BM794bZb3kgvRWj9zDtqFefFL1Hx9CL9YH+whqg52eW4ThnYyaDzybaWjOMUuUCOnw52rlFATGAXr1nQAO4s2fXC+yc0b/rp06f0nnCb0V3GTW1thcKWA/Q6eh0F3WDjig9g98C5TM7RiDyCSAhhBBCSMPgBJAQQgghpGFwAkgIIYQQ0jDG2ws4SP/9cgTfvyA3mMNj1H55MZyFI5x9JMcGNYDgleZMcWOm9y5N2npvX9RmVcV/6/OL12yey9fBBzkcAVqL6DU9NYCjELJVaWcr6HZGjYuISHtiuAaw6tTrrxLY0zGBv50K0I+UPe0f5e5UCR8WJewjmekYyUAXFx2dE27JXYFPW3ta+3finrwiIr0l9OtErzB97bhXpTlfHA0cirqgC3ZevdvkkTp+cQM5DvlucxnF089attXXHw9BjdMCiO16oE+KLdt+uFVp7KLmT3+PnmVetXFf1gLO6UHzdB0ZWgGiL9xTdQLqHabQf83Rx6L9K+oGIUaxni4D19bbaiGq2ebX2WMZdcbQN+h36oFeuxmkt2/TY8zMlF4L4O2L3gHvuwL9KVGvCT6QqCEUEemiVx7oHfH7hYU5k8fe3Xoc6k1rTeBCR2sPS9BDZ9N6riEisgTlgixbAqxbQB2uiEiGSx3W+HcdfAJICCGEENIwOAEkhBBCCGkYnAASQgghhDQMTgAJIYQQQhrGeItAZECrXemFE2nYaw8O16hkJf8NByyMW7wtAhZXhKBFmGYthohkM3oBSza9AzLVws5u56RKJ0vHTZ6xq68lGmNoFAh7W42Pbw7bRLpFT5Lzgm00FS3Lnjm+t4hOtGgiC6JcR3SbpvpWMeVWugxcaJI6ImUsFxdGoFlyWcDG5IU1P8XN6SswIE/bOs9t2VUmj0XRYuilxSWVxitB4+jEkSGjsBn18ksLuoz25KLJA82iB9vYM9reLHABxyiLQMbN8/ynOomLcyrcIF5/v+C4fne6sKgDxxyoB67dcTzTpYQIqSo0DoZ648olEZmExQotqFaE1SkJ3LK5034ZLoDAfoN6Yz1XPtWs5pFVmx+EcWBhJo5JXghFYx6PC7yw/20eLViQswcWRlw1recGOSxm8xaaRPisgEV0OOZkEHhpC1ZFiJj7YwkW6i3CAo7JljNZgPbaNqEXuKAxemcZfgcKGxMt2JigW8FmBnCTZblTLxwK1vh3HXwCSAghhBDSMDgBJIQQQghpGJwAEkIIIYQ0jLE0gEFiXyMSJIfvrHFrkr5bpavyFX1ABA2ga/QKmgXUKBhjUq0lSCbs+/PkKjBn3KbfyaOEJwENg5TLJs+Yas1SRH0XNrXjUO3s+XzRmCzXW6YSN1+pWMSibwyaQjsWhd3QuwB9CWplkhE2SI+gpUtAw9lD7SHkkaQ2Do0+L0UD0BrNnyN8M5/ApuEBNmbPE1uvzqLWpFSLOraxWNT8eWGMhrNoJhtBSzN/xuqD2/mkSrcmV+/bsBE3zwZyITpB1AXiFaMhN24Y72Fla0PERWJNnlN0MxeRHAIkSXWcZ1DPllPPdgbXCnrGEu5z1ASaQfxcyTrPGk2lt1GB8VoeGEtQ97sZJEnSjwscP6zeTyRB3SP2laNVRt5xldbizoD+rg2/lxPwfa+wOm0Eb2c0PUddIRqFi4hU0J9oLj0leh4w2bKmzXifZqDPA0m1TMGmA2g2LSISQYuYtWCTAZgH5amdpnVBIznY1U63rwmfABJCCCGENAxOAAkhhBBCGgYngIQQQgghDWMsDWASVzWAKHeIAX3vRKpEawBj3KMPKMEX0NHFhTD8hbaRbST6ktLWlCARNAkRLYTAPDAsa++fsnI0DEF7pxlVS+X4FNmajXDMeBhbp3UoAvUZm618CcmqXLSCnbQxfe4E8CGDmIqgY/KkqAGOMeVAniXWy9mc3urAarwGQWfoWsXB33QBdEGhRv947hz0Q8MDanRibkDgOahng2vt2n6cn9daxB0DepstsGBbVy7IOxD6IYEYzUFbh/ovEdu3VkoHvpKQTp1qtxP9IUicZALqNeHcbxNR66QKkF2DxNoZgzwvSlvOcJwTLjGp6aAG0MaQo88ELXICmvku6PNQV7hSpjoG+g+7EzXCLdTUO3RA42l1ceAD6GgA0cexiLrcdq7znMrtbzQqyufAOxDH06u26fnG6XnrZ1rBuoa0Rs9YOr8dZQnazQs0AuQTQEIIIYSQhsEJICGEEEJIw+AEkBBCCCGkYYy3F3CoVrVOqAH0NG5xlz7deAXq4qPju2QLwjkreB+h0MOTEC7DPoOwT6mAd095Ruv7ZOFtk2VSnVbpEHQZJWienG05LzV5ySVLjKvaT/TsQ42SiBjxD8ZIaTRsNo8U98wEX6oEQtfsh+pUC7U0lREpgfcZ6Ao9rR3aeKWgiUUto7fnbAC9De7ja/zTDJ4/IZyDnnYoLHXq1evqe6o3oBOqnP1FN4sL0e/hKaPp04ZrvIy3IhqCOeJW4+tX008piP5amb32HPSw0zC8TkF60nkMMQUDd1XivQAxan4XRtjHF+ppusA3tNSHDBxTbokQdaDMGp9IERsCdR6eqCsVEVnsaEHmzgm9nzjuZYuPmVqOr53ZH7jGfw/3dc6dTamNLjAM1z979+Dckt6j/O35eZWegvUE07BXcJ7ba+2An2sFeueYw171pTOJMXGYDPx79JkEnwASQgghhDQMTgAJIYQQQhoGJ4CEEEIIIQ2DE0BCCCGEkIYx3iIQqWR1VQUKXu2m8qHaqdKpXKPSUbR5dBVO11chYpVRbK+tGysQcYqIBDTxndei1ghi83JJmzmGwjZbGk5APbXQEw0lJTiidbOhPZeFeHS7hdlgfhgVLOqJYEQbEzDmNP1gN7DP4ZhYQjqDRVGOSLksOjqNYl808kXRsiN0N4s+cEEHHB9xBYyIBFihlKC5dI0RtLcoooI/Nc1iFOxOpw9QZF8MpKtL3Aka2wTbMKIxtmfaDHmU4Ibchb7EbkhGcIFPoa/buY65ySxA2uY5AXGKizzasMBjwjH7z9CAGs6x7QPm5c6l4ohbt3jHu7/MDTRQLcevd+MZvAZzWzr3ECy2MDGCRvHOIpClrv5tK6GcLNdzgRYuhHBMm5McFrhB2+PiPqwXLhIRsYbTWE9cELMI1yUisrCk5wapWbymj+9gHl4Iwb2eQ3tgk3tP6bCf1GIuLgIhhBBCCCFrwQkgIYQQQkjD4ASQEEIIIaRhjKUBjLGUVTUbbEzvaAClmlTJNLxTfy1aI1iFM06pMEc1GkCtrou4S3jXEWaAVqYQrcVCk9+A2hqBnclFpEp03YPUaQE841rURlAD6BHLUqo1drxGbYSINQk2nqMJ6rFsvhlucJ/peA9Gx5EM/V5EpEJNijFg1senxpDZUoLeEbVSVh7ixKE5aFxtqid8Qc3K8GtFg28RkYj35cBJ45ifbjTrUxdPA1hBGjV/kAbdaXT6BU3PJ2EYzzPY7H5C12u7M+xPQ9VbWC72o6PfrIwGFDR+eDxqqt0uGK9fRpH0DVaz3IoQDLJ6WaNocVFmDNo51IniOCZijYw7UE6ea/0z6ubO9OyY0+vp+UTW0oG1iLp8iJmlntXvBdBQp5Be7Om5wxlnvQBqJmfa2uh5YqKt0vPLer1AzzFxzuC+NHo+6DZjaC1idhYYvLfHGX74BJAQQgghpGFwAkgIIYQQ0jA4ASSEEEIIaRjjaQAHfABDQI1byzkB9AWyAw7YDce/7pRqHPQ0xk8P/LYcP7HEeJ8N94OK4GuFnl0iIiXoCBOjIMFzemJAvYUpZ/hm8D6Xtj/ahZAmaV83YXytRtg422ykjR5sTrtWlY7DAjcWh3Mi/G1VRqe/jdYD9ThwPIhDKuda0dcvQhymEMuYFhFpt7SuBevlqNOwouYIq/kb7mkYPAUW6DAHq37pKABH0wDWedBdSDnYlyZ+XGmmPqgN+qRp2Jh+ZhI1gDbTadAnBTDI68G1F5WndxzuaZhCLKBG1xVC4XiKwwIcjhrduiycn4UNJ0joxwG2WeK1AYwhqPFED1Tvty6Fse/04oJKv2NmWqWzCb0W4I3vgGeuiOyY1GNOPjWj0ovzb+o88doKW8+rpnS5WaqnOyfnteZvqWPH6Az9SiFKOj3wcgWRpadlboFPIt7HeI73e2R8RJUPoDl8TfgEkBBCCCGkYXACSAghhBDSMDgBJIQQQghpGGPuBRxlVfWA2jvP104TYO/fBDWAbnVQA4haj2r41yNs5xhQ4wS6iCroelWVNb8yHl3G62gEVylTV6tKuVhQi+XsDuueNSzHzSYLSd+fqqy0bqMoHM0oXFKGOjjof/StEnF0GJBnAVrEHh7u/qmFvm6wRyTo5FDfVzp+WmkYvld2gE15q+j4VKGWDDWw6DUI53taKLy2BOoZYA9v78ZttfDaqjX+vfGEsKq9uhDfv7pzPI0gnoL+YagBRL1SxDYWu18qeqWlsE82Su0c203rxQn6LDwlS2zfJTX7optRrEbf5+WB0kP0SczcLkLt1ZpfbQpBQl9/bLxR3RgbriPGsc8btiIcU4BHX6+jdXGnQBP69hnr97vjqutUugPjOF5KjppBT68JWrv5Zb1uoQeewRnuWSwiOeyFjX6uZYn+hfDbEp094GHOgvu3437cnhdjt4B7ZqAf4xj7ovMJICGEEEJIw+AEkBBCCCGkYXACSAghhBDSMMbUACayOmccpsdZATRLQb+TL0X79MRotXUB/PWMFAunsCDs8Oy2KkFdiz4nA+2VFTk5+hzYHzhJ34Zz9D7IsdJ7CvoF1eks6/R8IgE0lHbPZvRv9HwXsdEH22fMEFoHqqLod3yKuh7Pwy9if4OfHMpIHV1cClqpBPbQDLinJjQj+kOJ2H04UzRuK4fHaZ5rHYyISIQbokKhE6QrJ5bLLlQefROhnjn4a3k6zGj2uYa9XUeQL2U56hdL999XLuPpBrFvPV84CEHJQY8HTS7tVOc54QnllsH3D8KhCtj3NgZRZ4aU6BtZo48UceTi2F5YB1dLhT6bA3UaQ3u1XqRp0r/WBO59b8ypixHULuNe0SJW73zt9qtV+p0zWuv/ygmt+YvOvVqAnx76l+7coX0B50BnONW2PsSnTp9W6Q5oAFuw37DxFhQRgXhfXNbltkGXjL6K6A8rIlLA7wvu3260qk4/CugXqwFxLs5vhsEngIQQQgghDYMTQEIIIYSQhsEJICGEEEJIw+AEkBBCCCGkYVzAIpDzYm5cfRE8ETYIZhMtBq2q45CH3pz5/FGQZY3xc00dRKzgHNMoykSD0OCJMuUkVOQVlUzBQDIGLWoVsYtkBE19IywcqbZBesrkiYLryojxccGL3RDbLD4Z6Pt12NP+wjhfLho0e0J3NDJuZbAZN3RnmlnzzgwEw7gIBMXUKAi3xrYiJjZLFKVr9XwKi5MCLlY696ku1yySwaNtvZaX9YIm7OIp2Nw9g9heXNCbw4uIhATaD9JGGI6dInbRjF4EsrlG0JcCZkN4iEFMe0a3eL8EMODGhUlG5O457aPZsBlv6x2TQ42QHa+tHGEcwkUSdXh5VtXaC0d6W7AIpKyq/u+THYvrf/twuCxxQwPnknDsg5CQChaJLBf6N6XdsuNWL+oFGgXUfRIWK06bhZu2nm919XwCDf/bEzrPCOtNRUSWl3XdAyySKsFcvepCvZzfEjvm4oImbFDHFH9S38uxvVpOWY6+II5PAAkhhBBCGgYngIQQQgghDYMTQEIIIYSQhjGmBjAX6Zs115kWi9EFVvGUSseg08HbWbwE7ZzZ9BvesY9kCItmuKClKfCdO+jiHH1SDMdUuii1NiAJr+p07hhBJ5OQBk1fqTV/sdipD09323oJGlDvhAPgWoJjBF2ByeZAm1uD340nT/K+DqXXA41GacUgqH2aaut2xQ3RXRPiDNoA/ZVrNi9PUc8pIhFipIQ06q8yMKMWE6e+8aiuJ+o5nePBZDQFnVcbTVdBLJWmNiaiMf8FrRHqYNAUW6xeTekqR9CVbRSoR0Nt3oi5QNoxda/JFw1jkwT1fZ5Juk4XcD93YXhYKnQ/TiR2vG3DMJ6CbkpQQ+1cVp3FPf4O4BjuyN/Mh3iKkTM67YXnDGp9t0ACKBJjX6eH4wOOJ+eO0f2LMYI/n94l4TloDL0EY0wPMvVMm0s4poAYiaB3noYxvUS3cRFpZ2iMDWXCmF31rP49S3QL4G9JtwO/8zD2eRsTIBGOCSg+dXS2Ea4tGyh3nNGHTwAJIYQQQhoGJ4CEEEIIIQ2DE0BCCCGEkIYxpgYwyHhvmPV7/SRoDVsa3qfShWw3OVRBeweGAB5l1bxOi/YTcgVdOO8Ff6gI3wfUJ6EmUOzm1RJ1Pav4hk73rI7Q+P9gGRG7CzSD0L4iIiGZ1en0OpVOIpyDXoMiEsLVa6ZjXMQW33DykEtyXhOCulFPf5ZAf6bgt5iitK6wepIKPkpBb5Oi/x74Q3mGWhH0imkFG4ujbrCLOrn6v9+MrhCESp5PVZHrfHtLOo9uZ1GlUUbqbl4OPl/YGhU0cJrba8N+HNTqxi0RYPmgJtDDDkuoI/TOGe77hyTJCPpco1VFf9fhfntes3cLTA/X53mXkcG1os+b7z+4itd+RvMH32Nruc6d5sOt056ulO95eYo4mmGxXnjo+ycptLsTQ6gBxDx7MH7iMJXhgCsiVY3fb4Tx4QyMhVlqx4vtuf59XAKtYgStIsaciIi00PcSxnnwh0X/QvREFLGSvgT0jQneH6lTL7gRy+XV9qEPICGEEEIIWRNOAAkhhBBCGgYngIQQQgghDWNMDWApq7q+Oqcm55jqHSoZ5YM6HbSeT0QkSdArEPbcrU7rGka9D6mnLwmgrQqgCcPv03iV/t5o8UQE9m2Ngvvygpaisvseh6i1hTHqY2IA7ZWgrsq++68SvSexJK9BGfraJIKuUETa6Q0qnaY3DZy/LF1v++ANJE/zvtYP99iNie0b1E4FOKcEzVpw/i7KIF9Moz4NZXDR0cUlOWg/UFBlNu4dRVumrw33u0R9Y1XZmJkAn65iQutC0XuxBx5csXL0ewE1bngMegk6fQCaJuU/tskawBBCv63r9uBdOf5iqdP8of7VlOlp7UCgleFewJBGrWXh7K+71NPHLIBeC38qPP1mBpXNk+H1qo8v5xj43u7QauuFt+RWW1FmWSLJSh9C+di3InYfX+w+jJnc0eu1QQMIMjk52YHfNvTJdHTaXRiHSoiZOfCnzcy47+zfDkNbWsG4j3tlwx7xIiI90CJWEIftCWgfMBusnN/GlqAOE+oF/snubb8MHrIDnY+/Z8PgE0BCCCGEkIbBCSAhhBBCSMPgBJAQQgghpGFwAkgIIYQQ0jDGXARSiKyIMSMIJh0xsIg+pgTj4hjbKl0FnT6HXviQymmVzgSMoUULUF05JNQ9YBoWhVQR6+UZ3cJCAjBYDoICU2ufbEw54doEF8kkmIc1MBbRi2ZixHNATOuYXPfAHLoc6Ovo9vvG0s4GF4Hgt96G97pdcREDOpVOtG0ctmEhRAF5oPA2BxF6mjuLU6Duy4s6dnHBRivHOLXXigsB2nAteE6BJr1iNzyfaOtrP3tWx+GZM3MqjRuii4hRyKNxOmr2vUUgKW5cP3At67HIYhxGMXvearCOlbsiTvdVAYcUsECjB0J6L8/ELpXQ5+CY4yxuwlHICtthcQrmacZSx8QXF4eZxWI2j1JwIcFquhhde79uJOnqIhC4PSTN7X2IBuvY8miG3HYWV+CttgTtdAbGRlwAl0abZwRj4wDpFAb6CbgOxzdeFgpdj8LEUN0yILtQBBs5h3Nw3O+mzj0XcQzGhSYQy86ijgyutzWwWIdG0IQQQgghZE04ASSEEEIIaRicABJCCCGENIyxNIBBBt6Sj2BKi+/UAxSHxpmJaK2RiEg0WkPQERq9idZN+cog0CCgJgHSEY0ZQ8fkaE1E9bUEzFOs4XI0OiZ8l78TToBr9zaQRw9W023QXs61VSX0m0yv/jtaQ+uNJk1SSc9rU6oUTZxtI6BOEfV76JeK2iARMQafCZYDWik0Kg2OPqgqh4uG0Pi4BLPUlmNciud0u1ZrOsjkpL3nUI9Xgh5nBq4tz8E4umfHhrNnz6q00WNl+v5otezQFCNs3j4ghKnKzdXkDerrRtEfWs3gcP1ZHMFVOEHRV02ZnjSoC6K/FOrRK0GvBPWqXLPk4bqp1Jxj7wPU9GH74P2H7eWFA+oZ8RirMrTPRzC0B9uj5nbeEKpY9p2CW2D67hku15lwo/6s07W6crzMpRSM4DHO4IOOo3/HuJoEDXAG14LS5eDEYQ/GSxwjctA6o8mziEgF+rwIBXfM/QMbAjga2Q64Q3ehjGSEPAoYL8OATrCiETQhhBBCCFkLTgAJIYQQQhrGSK+AV14lRFkaWDeOpzqPm81j2d7Qb/1PcF883P8W7QBwT14PrCvOg+F1LTzCxj14XYyFAOZZ/9rDvgJGc4QRHvViA9S8Ahaxr4DF7H08aAOzdP6/G/8KbqWMwWXuuOTdfwUMrzng1QC+svCW0eMrywpfJ5UYp8NfYYnYV8Devrw6T318WTp7fSbw2qPmdYD3GhFlAvgKGK8Vy8C2EfHjXX+v01698bNBq4iV7zY6Dvtj4ZByNssiBou5kHLtvaG/LzHO4fue867V7BdslCqm4jYPSOM5ZidpjFkn3NDSA6teO1Q65wymyxFiY71YKePcvVad/3e9HAaHIXy9aM5w7uU6MxVzqxupkmfVBa/wjXxs+L7WuCe8iLUOwrgsIUjs768do43UAM+BpLGRETE3mRnravpIRKQEaUYycEw5xlg40gRwbu6cz1eU/7XGhGoDqdFprIfsYuxr8qx91qEeY5e7EYWOkKd3yNzcnGzfvn3dq4NliIj857ee3dByyCXA2fpDPDY6DldiUGS0yeBaXCo+gotLzh98A7yxSfW4ktjMsfDEW6c3tBxy+TJKHIY4wkhUVZUcP35cZmZmNt1wlVzaxBhlbm5O9u7d2zck3SgYh2QtNisOGYNkLTgWkkuBceJwpAkgIYQQQgi5cuAiEEIIIYSQhsEJICGEEEJIw+AEkBBCCCGkYXACSAghhBDSMDgBJIQQQghpGJwAEkIIIYQ0DE4ACSGEEEIaxv8HJibIA6AAtAYAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 800x200 with 4 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Found 3 examples with light issue in the dataset.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 600x200 with 3 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Found 1 example with low_information issue in the dataset.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 200x200 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "imagelab.report()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f64b1e8b-97f2-4ae3-913d-699541fa9ab6",
+   "metadata": {},
+   "source": [
+    "View more information about each image, such as what types of issues it exhibits and its quality score with respect to each type of issue."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "1a7bce39-a072-4284-b7e6-2b12c55949e8",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>odd_aspect_ratio_score</th>\n",
+       "      <th>is_odd_aspect_ratio_issue</th>\n",
+       "      <th>low_information_score</th>\n",
+       "      <th>is_low_information_issue</th>\n",
+       "      <th>light_score</th>\n",
+       "      <th>is_light_issue</th>\n",
+       "      <th>grayscale_score</th>\n",
+       "      <th>is_grayscale_issue</th>\n",
+       "      <th>dark_score</th>\n",
+       "      <th>is_dark_issue</th>\n",
+       "      <th>blurry_score</th>\n",
+       "      <th>is_blurry_issue</th>\n",
+       "      <th>is_exact_duplicates_issue</th>\n",
+       "      <th>is_near_duplicates_issue</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.854301</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.797130</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.915605</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.484206</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.926223</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.850131</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.976886</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.571952</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.818548</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.852996</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.532279</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.813629</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.844033</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.775992</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.401225</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.898399</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.958406</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.919564</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.503716</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59995</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.860407</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.794629</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.996078</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.523458</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59996</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.888932</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.939203</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.843293</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.498186</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59997</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.818150</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.960275</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.865067</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.444907</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59998</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.900018</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.892104</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.952069</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.528622</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>59999</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.858985</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.809504</td>\n",
+       "      <td>False</td>\n",
+       "      <td>1</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.932046</td>\n",
+       "      <td>False</td>\n",
+       "      <td>0.501550</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>60000 rows × 14 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       odd_aspect_ratio_score  is_odd_aspect_ratio_issue  \\\n",
+       "0                         1.0                      False   \n",
+       "1                         1.0                      False   \n",
+       "2                         1.0                      False   \n",
+       "3                         1.0                      False   \n",
+       "4                         1.0                      False   \n",
+       "...                       ...                        ...   \n",
+       "59995                     1.0                      False   \n",
+       "59996                     1.0                      False   \n",
+       "59997                     1.0                      False   \n",
+       "59998                     1.0                      False   \n",
+       "59999                     1.0                      False   \n",
+       "\n",
+       "       low_information_score  is_low_information_issue  light_score  \\\n",
+       "0                   0.854301                     False     0.797130   \n",
+       "1                   0.926223                     False     0.850131   \n",
+       "2                   0.818548                     False     0.852996   \n",
+       "3                   0.813629                     False     0.844033   \n",
+       "4                   0.898399                     False     0.958406   \n",
+       "...                      ...                       ...          ...   \n",
+       "59995               0.860407                     False     0.794629   \n",
+       "59996               0.888932                     False     0.939203   \n",
+       "59997               0.818150                     False     0.960275   \n",
+       "59998               0.900018                     False     0.892104   \n",
+       "59999               0.858985                     False     0.809504   \n",
+       "\n",
+       "       is_light_issue  grayscale_score  is_grayscale_issue  dark_score  \\\n",
+       "0               False                1               False    0.915605   \n",
+       "1               False                1               False    0.976886   \n",
+       "2               False                1               False    1.000000   \n",
+       "3               False                1               False    0.775992   \n",
+       "4               False                1               False    0.919564   \n",
+       "...               ...              ...                 ...         ...   \n",
+       "59995           False                1               False    0.996078   \n",
+       "59996           False                1               False    0.843293   \n",
+       "59997           False                1               False    0.865067   \n",
+       "59998           False                1               False    0.952069   \n",
+       "59999           False                1               False    0.932046   \n",
+       "\n",
+       "       is_dark_issue  blurry_score  is_blurry_issue  \\\n",
+       "0              False      0.484206            False   \n",
+       "1              False      0.571952            False   \n",
+       "2              False      0.532279            False   \n",
+       "3              False      0.401225            False   \n",
+       "4              False      0.503716            False   \n",
+       "...              ...           ...              ...   \n",
+       "59995          False      0.523458            False   \n",
+       "59996          False      0.498186            False   \n",
+       "59997          False      0.444907            False   \n",
+       "59998          False      0.528622            False   \n",
+       "59999          False      0.501550            False   \n",
+       "\n",
+       "       is_exact_duplicates_issue  is_near_duplicates_issue  \n",
+       "0                          False                     False  \n",
+       "1                          False                     False  \n",
+       "2                          False                     False  \n",
+       "3                          False                     False  \n",
+       "4                          False                     False  \n",
+       "...                          ...                       ...  \n",
+       "59995                      False                     False  \n",
+       "59996                      False                     False  \n",
+       "59997                      False                     False  \n",
+       "59998                      False                     False  \n",
+       "59999                      False                     False  \n",
+       "\n",
+       "[60000 rows x 14 columns]"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "imagelab.issues"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a9d4c1eb-bbf7-471c-b057-6161c8ac45ae",
+   "metadata": {},
+   "source": [
+    "Get indices of all **dark** images in the dataset sorted by their dark score."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "c54d1234-c076-4e07-bc4a-e8610f0d27bf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "indices = imagelab.issues.query('is_dark_issue').sort_values(by='dark_score').index.tolist()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d7f241fc-0674-49af-ac59-c2cde1a79b3f",
+   "metadata": {},
+   "source": [
+    "View the 5th darkest image in the dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "345a9996-850b-492e-ba45-afde3c1ca64e",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<PIL.Image.Image image mode=RGB size=32x32>"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset[indices[5]][0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "9efed1a8-3a2d-4f2b-9470-f8196a3d5791",
+   "metadata": {},
+   "source": [
+    "View global information about each issue, such as how many images in the dataset suffer from this issue."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "9f47f3b1-41d2-41cb-8816-25ae16ce2b29",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>issue_type</th>\n",
+       "      <th>num_images</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>near_duplicates</td>\n",
+       "      <td>40</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>dark</td>\n",
+       "      <td>29</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>light</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>low_information</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>blurry</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>grayscale</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>odd_aspect_ratio</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>exact_duplicates</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         issue_type num_images\n",
+       "0   near_duplicates         40\n",
+       "1              dark         29\n",
+       "2             light          3\n",
+       "3   low_information          1\n",
+       "4            blurry          0\n",
+       "5         grayscale          0\n",
+       "6  odd_aspect_ratio          0\n",
+       "7  exact_duplicates          0"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "imagelab.issue_summary"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "24c4ef0e-fb94-45c6-a76e-c74825c1b214",
+   "metadata": {},
+   "source": [
+    "**For more detailed guide on how to use CleanVision, check the [tutorial notebook](https://github.com/cleanlab/cleanvision/blob/main/examples/tutorial.ipynb).**"
+   ]
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}