Discrepancy_GenerateDataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generating a synthetic disrepancy dataset\n",
    "\n",
    "Here we go through the process of generating the synthetic discrepancy dataset.\n",
    "\n",
    "The results will be stored in: `DIR_DATA/discrepancy_dataset/dataset_name/variant_name`\n",
    "\n",
    "We provide the generated discrepancy dataset used to train our discrepancy network, so there is no need to re-generate it.\n",
    "It should be located in `DIR_DATA/discrepancy_dataset/cityscapes/051X_semGT__fakeSwapFgd__genNoSty` and can be downloaded from the project page.\n",
    "\n",
    "The generation process is non-deterministic as it choses random instances and random classes.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# %run -i env_default.py\n",
    "%run -i src/notebook_base_light.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Experiment class\n",
    "\n",
    "First we create a new Experiment class and its associated config. The `gen_name` entry is the name of the discrepancy dataset variant: here we choose a new name.\n",
    "\n",
    "```python\n",
    "gen_name = 'my_discrepancy_variant',\n",
    "swap_fraction = 0.33, # random fraction of object instances which get altered (default is 0.5)\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.a05_differences.experiments import Exp0521_SwapFgd_ImgAndLabelsVsGen_semGT\n",
    "from src.pipeline.config import add_experiment\n",
    "\n",
    "class Exp0551_NewDiscrepancyVariant(Exp0521_SwapFgd_ImgAndLabelsVsGen_semGT):\n",
    "\tcfg = add_experiment(Exp0521_SwapFgd_ImgAndLabelsVsGen_semGT.cfg,\n",
    "\t\tname = '0551_NewDiscrepancyVariant',\n",
    "\t\tgen_name = 'my_discrepancy_variant',\n",
    "\t\tgen_img_ext = '.webp', # better compression\n",
    "\t\tswap_fraction = 0.75, # instead of default 0.5\n",
    "    )\n",
    "\n",
    "exp = Exp0551_NewDiscrepancyVariant()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cityscapes: found images in .webp format\n",
      "Discovered 2975 frames - DatasetCityscapesSmall(2975 frames, /cvlabsrc1/cvlab/dataset_Cityscapes/1024x512, s=train)\n",
      "Cityscapes: found images in .webp format\n",
      "Discovered 2975 frames - DatasetCityscapesSmall(2975 frames, /cvlabsrc1/cvlab/dataset_Cityscapes/1024x512, s=train)\n",
      "Cityscapes: found images in .webp format\n",
      "Discovered 500 frames - DatasetCityscapesSmall(500 frames, /cvlabsrc1/cvlab/dataset_Cityscapes/1024x512, s=val)\n",
      "Cityscapes: found images in .webp format\n",
      "Discovered 500 frames - DatasetCityscapesSmall(500 frames, /cvlabsrc1/cvlab/dataset_Cityscapes/1024x512, s=val)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "DatasetCityscapesSmall(2975 frames, /cvlabsrc1/cvlab/dataset_Cityscapes/1024x512, s=train)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load the Cityscapes dataset\n",
    "exp.init_default_datasets()\n",
    "\n",
    "# they are now available here:\n",
    "exp.datasets['train']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Construct the pipeline\n",
    "\n",
    "The steps involved in generating the dataset are shown in `exp.discrepancy_dataset_init_pipeline`. \n",
    "\n",
    "The object label substitution is done by the functions `tr_synthetic_swapFgd_labels`, `tr_swap_labels` located in [src/a05_differences/transforms.py](src/a05_differences/transforms.py). \n",
    "The chosen function is stored in `exp.synthetic_mod` and can be overwritten if needed.\n",
    "\n",
    "Set `write_orig_label=True` to also output the original unchanged labels. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "exp.discrepancy_dataset_init_pipeline(write_orig_label=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m\n",
       "\u001b[0mexp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdiscrepancy_dataset_init_pipeline\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0muse_gt_labels\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mwrite_orig_label\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mSource:\u001b[0m   \n",
       "        \u001b[0;32mdef\u001b[0m \u001b[0mdiscrepancy_dataset_init_pipeline\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0muse_gt_labels\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwrite_orig_label\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0;34m\"\"\"\u001b[0m\n",
       "\u001b[0;34m                @param use_gt_labels: True: the starting labels which we will be altering are the GT semantics of Cityscapes\u001b[0m\n",
       "\u001b[0;34m                        False: Load starting labels from ch_labelsPred\u001b[0m\n",
       "\u001b[0;34m                \"\"\"\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpix2pix\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mPix2PixHD_Generator\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcfg\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'pix2pix_variant'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0;32mif\u001b[0m \u001b[0muse_gt_labels\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_load_correct_labels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mTrsChain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0;31m# load cityscapes labels and instances\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0mTrChannelLoad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'labels_source'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'labels_source'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0mTrChannelLoad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'instances'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'instances'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0;31m# convert to trainIDs\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0mTrSemSegLabelTranslation\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfields\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlabels_source\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'pred_labels_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m     \u001b[0mtable\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mCityscapesLabelInfo\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtable_label_to_trainId\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_load_correct_labels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mTrChannelLoad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mch_labelsPred\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'pred_labels_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_alter_labels_and_gen_image\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mTrsChain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;31m# load original labels\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_load_correct_labels\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;31m# alter labels\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msynthetic_mod\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;31m# synthetize image\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mbind\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpix2pix\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_generator_np\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpred_labels_trainIds\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'labels_fakeErr_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgen_image\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'gen_image'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_synthetic_and_show\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mTrsChain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_alter_labels_and_gen_image\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrColorimg\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'pred_labels_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrColorimg\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'labels_fakeErr_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrChannelLoad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'image'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'image'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrShow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0;34m[\u001b[0m\u001b[0;34m'image'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'gen_image'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0;34m[\u001b[0m\u001b[0;34m'pred_labels_trainIds_colorimg'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'labels_fakeErr_trainIds_colorimg'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'semseg_errors'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_synthetic_and_save\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mTrsChain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_alter_labels_and_gen_image\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;31m# saving as image does not like np.bool\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrByField\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'semseg_errors'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mlambda\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muint8\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0;36m255\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;31m# write to disk\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrChannelSave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mch_discrepancy_mask\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'semseg_errors'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrChannelSave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mch_labelsFake\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'labels_fakeErr_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrChannelSave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mch_reconstruction\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'gen_image'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;31m# fake labels colorimg\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrColorimg\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'labels_fakeErr_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mTrChannelSave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mch_labelsFake_colorimg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'labels_fakeErr_trainIds_colorimg'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0;31m# write the original labesl (as they were before alteration)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                \u001b[0;32mif\u001b[0m \u001b[0mwrite_orig_label\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtr_synthetic_and_save\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0mTrColorimg\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'pred_labels_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0mTrChannelSave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mch_labelsPred\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'pred_labels_trainIds'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                                \u001b[0mTrChannelSave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mch_labelsPred_colorimg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'pred_labels_trainIds_colorimg'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m                        \u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mFile:\u001b[0m      /cvlabdata2/home/lis/dev/detecting-the-unexpected-github/src/a05_differences/experiments.py\n",
       "\u001b[0;31mType:\u001b[0m      method\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "exp.discrepancy_dataset_init_pipeline??"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test frame\n",
    "Visualization of the process on an example frame.\n",
    "\n",
    "| ----- image --- | --- gen_image  -----|  \n",
    "| original labels - | - altered labels - | - disrepancy mask |\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"display:flex; justify-content: space-evenly;\">\n",
       "<div style=\"width:100%;\"><img src=\"data:image/webp;base64,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\n",
       "\" /></div>\n",
       "<div style=\"width:100%;\"><img src=\"data:image/webp;base64,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\n",
       "\" /></div>\n",
       "</div>\n",
       "<div style=\"display:flex; justify-content: space-evenly;\">\n",
       "<div style=\"width:100%;\"><img src=\"data:image/webp;base64,UklGRhZBAABXRUJQVlA4IApBAADwVgGdASoABAACPm00lkikIqKhI5GJsIANiWVu/BFZbYcG8J9/sfyw8K+N/L35n+7/tl/ff3Q+eWxf2r+7/p7+3ftx90H8TuK+C/6f/I9D/mT/Zf3T/F/s58wf8T/0f8r7nfz1/1/8j+//0CfqX/tv8P/ofgr/w/3L91H9o/8H/G9gX8x/xH7Te8z/lf+//qPdR/ef9b/2f9V/r/kA/qP+F/+nrp+xp/nP+f7BH84/zn/q9dn9xvhM/sH/G/cb4HP2k/+/+39wD/4eoB/4es/6j/5P+4eyz51+/f6v8uOiX9vfjE/TssdbL7wfxfXp/H/4D7ZvSHgBfl/9F3XfUvMC7sf9r1TPov+Z6BeIB5Lf9D8VfN9+1+oB/KP7x6uP+H/9v8550/sf2F/2L9O/1//vB///cu/Zf//jJszN7vlIeb4CwwAShssAplU3ykPN8BYOUOYVyYAdh35i2fs/tRXnl4q2/E+XEXMwfXYmswpNgaS+kF1mrXKmYPrsTWXpQCV3eOJaUNlgABwXNtxalXI9LxZWVHJSWIGQn47KjkpHc36U+MQ19TnDyAKMwxJanPUoi8tyg/Vjsy3SMFWOxiGVAJNpUWOjsr4NaiAHW+YurEDIT8dlRyUliBkJ+RKXvkTJAaSvquwr+QgZByksQMhPx1G3fRJD/24IQQraBkadLXIUDHyLBST8dlRyUliBkJ+Oyo5KTlDXjIVEfMsko6ynrWlRyUliBkJ+Onqrrwqa+aYIzNXNI2KjcaHc5SxM9fQQOo6gWmsQMhPx2VHJSWIGQn4yxuIvrHikCaSIqzrPRQ7RIqj/D4FwDhkJ+Oyo5KR3WXZhEHG7C8Rx3oZOq61LMzUPBuTHpYgZCfjsqOSksQMhOy+m7VS+mF9Z8bflKXqh3kZaRGkY1TDUw7KX5Ytn5n/+Klr+dqXevLpVmyKfvoU0w1MOyo5KStmh5yxz8dZslHEffX7kNwMnzkIpWdHHHG90d5t8g5RAVUYMnldageqqzbulNQ8cWN/S3HVRUb71VL6t5Pp1iBkJ+Oyo5KHrkMFK9bLzFXgZHQbpR6XCn5Ct67yLKylqocQDMdhlN2EOHrGEfO2H1mBVIQMfqkGQ9XxKXDZ93VdIODi07m5Bt0amHZUclJYgW3WygARpP1EV/6N4BShM4j01tpLF+A5k+kXLYctd6ALwcskCJs4ZBUacV52f9jUXZS9KwW4PjzsY2iDDRTmkuzhRuxVS9+8CSDBHzyN6qb0/SY/5HKutBpAHYYjfNVPUGkZ9ciHAEU540ni/jsqOSksQMg/8rvZCr9AZo0rgAc0z8cElcuyLVQ7480AOtrBOyuy+cpL5oPdkdUWUv/6MIsKQnKbGuCDx46fV5P0IkoHv49EZm/U71bfRlSH4WWR2U/BhaEQYmY3ON5agZByh8u4iCAWHU8qg6Yahy2PgI+UzPgMH9HzB4yhqYdlRZZYIpQzf/XD8sNb9+NZ8IQ5ftFjdA2sqg5gwiDkFgYaUrkKl3kFGij7DVAzd3w/UZIvdbKjkevCFW7RRm4gZCfjHEZQfTFoyFH+fE87gpaC0mViu8yfx2VHPXfIqQTUiLjerdncT1y2gdO7rehZG/ve7JsSH6OzdLhmaUeSE4MrENaDR20ziBpQIwF0QfDw+kziTCjYNi1ZnbfoIry+CUliAHmBJKj4QMhC4EpILeg/L67iawbpeIICo8MEBaI5if4qWLicGLYGmwlHUPPXpewAYr8kJh4e/Hq9qnICbGr0EqjNEX2BjpBIaL0LFhNTDG608+9c51osPCtdFykvlNGdOfsTyiDVrt7yKXFY3sOJIMIa/IWyW9TB3z84nyFokKkAH3/VVSQPz5c1T+IwUWL48YwL1cDfotxlXV8syJVZkBPXZStyiMCYrJctdBB82yzdqaUt1ofQiyCn8ONpPVDcs5w+h0tgTanvuOr+XU0NucZEyTWdpv6BDCWKCyOnsRQF6gtXsEMRsySqyYenUAq6iG88CINV94PWK7RcxJJokUrqn0AwlntKqMfuD1RlVUnKuxE26Vk/pjbTQGnL0F8RG2DCjqaca5RBqs2PyUL41CBc0tCxhS9WEScW8myPZksLcNv5DQyUJUOOAo1pJ39MSeWivNhFxBesiDFdnj3Ejg2HfpfkjOIV974LV1FPyQIeWgHJ4vUYkS7YGDAhkDMz9URVqPwoHACRWUj45W6ssreNEfZ4XzuW91aUfUicQeIwpUdH+jgwsvOiUk2ekm1GtsJUTKUwkrl2RKI5u6RU1q4Qmg1ZeV6gJBFmwkn/+fDoQjCagj3lU9Xhgf2tEvtfJKXV6T6MlaLNUNJA4uS9YZaQ+1ST6oaLXHcBTSGE3M3d3J1YHi+e8Uw201bfCf/gNxK8VMLTS5pBLJtzjCcLPr9QEFjtBx5MpfSxcGYLmju7t1HOmIchup68nzQEE7yxcN9MlAnz3XG+0nOFgf/zgbWhZkknYESDZxgNZLxVUCEPk+s/SN45v/nyZU1MQK+AieUNGEH7QtJ7tJKCbNJhfKx4iJ21pWvcI0BUh90S3cz6Cs3dhL7DZcvhnyjgh4gD599YCBq7oykdw5akhqs4LpALxp4fMgfxJsC368B1x3/zpHgw5qMtxnIFjmPr5syB9bGf+V0AAQ8ChCFTtT1sR0QZ4e1xlVh+URwZkB+nZpZbaQA74rZIP4twBNALNyJZrdBg7yx9lla+NT0plZ5lXYOczgMJqtyXOM1FC73FtW4Er7AlHPAmHZUclJYgZCfjsqOSksQMcrKvGwMpiOWWQtoOszDRssdkfeAQ4WO8iO7v0m9jDFlR4L/u9OCfjsqOSksQMhPx2VHJSWIEZjFeQHPhRixSXIaT6t70wT8dlRyUliBkJ+Oyo5KSxAyE/HZUclJYgZCfbIdfDrJiMdcMfQgM2SgAE1YeaYamHZUclJYgZCfjsqOSksQMhPx2VHJSWIGQn8C9cTwLjDX1G/nfIOUliBkJ+Oyo5KSxAyE/HZUclJYgZCfjsqOSkscs3ndkIGQcpLEDIT8dlRyUliBkJ+Oyo5KSxAyE/HZUclEnOcq/CBkJ+Oyo5KSxAyE/HZUclJYgZCfjsqOSksQMhO8oyPnpYgZCfjsqOSksQMhPx2VHJSWIGQn47KjkpLEDIQvf12ph2VHJSWIGQn47KjkpLEDIT8dlRyUliBkJ+Oyo4nF/MwT8dlRyUliBkJ+Oyo5KSttigU+QVNhxTXFG6zQksQMhPx2VHJSWIF41dHhkJ+Oyo5KSxAyE/HZUclCkrlES+kN7SMp4ZCfjsqOSksQMhO8oyPnpYgZCfjsqOSksQMhPx2Qcc7FDFLqqbw4/Mc5SWIGQn47KjkpK5pdtOcMhPx2VHJSWIGQn4x+02arV0Rexx3GM1QKNdRX9vAsG8cq8LGtXhlj9y7n1nSBlpENHfuGo5yksQMhPx2U2yjWgZCfjsqOSihzBjPf0NpEQSAc2+OfzrpjCWZlzWwugSugsHig7vM4hIa8MpADeYOCDvaiuu2aYYHkQbOxc5zUuPC5n+1ltzusIZexk14rZwL9yfjsiSuE2hce6OVMwfXYmsuKULmlyIZ5AIr9wDiwEhh1LFBsyiRcyJZH9tGyd+YQgsOfchPx2VHJSYdwkM3aO06v2qXWqEyI9nheDhj5Fn65d0fUPJ9uke8BYAAP70XLfn/BsaufG30ABInZ4czHqhAqE5XS5p00xWO4Bh9BNZIylklBP+wAFHLHy3icsKvAuzNwe9CbxOclmQuaum7yNRvvcb+JgDpMWn4CLWVSH5agcsuLl1K+gyMzzgbOuxCUfVzaaMPY6mYZ0nmW5DAejfxQZiVvhi4JBQpHHj+hliVULaCfEOuZIc3GrV3RSQS+q1CGm0WoBp5iISA2emNk+YiEgNnpjZPmIhIDZ6Y2T5iISA3FkYWb9kNgk5zJP/DPNJ1+EHAAi53z/Wek6o97p8fLDLio/Z8PX9okHJQufXA3uK35ijO17fg2pNnGLRKdHCT7u5ilvOrZFzGLWQZy0TgAERm3OqB6WxFykZnq+Dk2muABINDkThh+hXayh3SZ3GhCRh8gx4GayvJec9jcoelobYGyIXpf1+tmoaM8Wvwc+kJ0Qt6iQlD+cJqkiwq/AA9tRMqJSys2c3VZwA9bzId2Trb8oeV8oeV8oeV8oeV8oeV8oeV8oeV8oeV8oeVyH+YxmY1R+AUxeVwZTeEAuEigxqXmf71//X8P/7iH/2+ACpRuc4HsF6srTpNOlk8GpGmmrCj2m5MOB4hWlUg7ae46jywGqtJWOSZ8/4wL488BHlyQXrCR0H0tMWlRg4A38VG/yDqGki5l3jEOrJpPpr4e1/bt9NGhbwXiTKFQINXz+wt0vLhM90ug+mZgxOAAhVUBr000E5NtYMzkeJLWPFggLxSCfL9scuFDIS1uHFdoADvBneVYPQl1AGPxnewIsdJSCqfgqTHyw153lJHwq/94TCDFNvkzAKlLKYAAA+jxvNlRainMKSTzCJT3ovYkuu2ceBJ7mOii+GkI95lcyWUJqVNadhzINssn7GS/fAbsl+tvJBQQymfHwKCLx1QyrS8/yeeX4cSfPD9inRsMe2cpaO4wVjLsFZ2ZOUD2VUe7mkX0tQe0GnbvQ/jrLvSO4/4KceTf+D8MI22oMuqixY8DCAv8leGAdLPjzwCO3JghB4hUMo9WfK1UhZcZze6AAAwD/f8c6akbrVXBZcqbLFMUxRpTDyLMBXc1agzaw6iMK+bEdhdQf+v5Q2qMi8VUg9rURCAupxXr25JI1NZ3qX5JD+tmuvr4udWSBiN6oM48sy3Ve5NQWBx0L+kFxrgFJdgCRDCRYGAoR1SOC3ohOfqpaUqc84mxFbHuWKaQc24ARRNwBuoBxM10774Of/UqXP60ydu/dA9QITHnb/dXVBzkskmIm1+HTse6+s2vqyR5S8/63S6AtyQSwQhiA7aQ5xUKd+gjSnYfcKTGZCZucVCZLaCDUCbgPTa8ukN/NHtf8eAiG9GM6pcD9r8z5gBcb4kfHVfhqZnibu+Q/TeVxg1sIThhelQ1Ex97hMXmaES2mSYMPDg75cXHAOL2ZaYCL4O3QXcKow55o5c838AFV78YiK8uZiy+e9DSeM0FcQWRIczMKSTrBphoVcpwv/g3Oz517xF5fJxVgmfwLzQHce0Yn8hk55CR15sJf4+PiIaszBKvxrclc6UXy6F0ToKfbjihLB/4FSjWH7weTFn3DLvtkWF8/uOmXwxzmtXLnEpn4c557B8z0iVHclgLpwwuo953lgR/Qt9Xf6Pdg1p98FKIwgaSGo4+AC+6pfJ4twPYi01INtp2qvEodW3XEx14pPfxQQcUwez06jwXiWJ5re8p//SE6roOxNasrl3KjQ7ogtpe3k4oAsqSS5FRsJT5bFQjjo6DjOnyCKOpbP73vxscFsr4juPZ6rxD2Dwb0P8aEQ3Ojwh0QMhcsWEnflARihydqqsSWHGQAAAEF9N4KeWoNFuAqqtT8cMkytHHEkukx88yB1Cb2al9UjqVD52Y/IOYdsmjz7o7GM8iovv3UVDcgSTsQpuSPZtMfuqeSQNOCUmX1sVLcal7mSjPaIt/IMcnkifslqkgaNF8moRUZXFhha23Nmv741ynfrsRjhXGyxCTvDOFNiMyOnYK67rmjV+vgODB6pfUsfzdgMlZV1DyQk4hkGJfHIsatufZ5dEk011CxHKHLWLaWWSYLMSGA8Z1/M7tdJSkS5w05xZJxQRB4k272pC/UHtMH/feNL4s9uJGCC9h2aFmZQWL6CXQi++VFvs+VmEN0BuzG08gLYTOAk4za/we56435NpiUsT6ayPmtoeX8vuK0GLIfn/AcGVAJojfKpTaOkvkH9W48ywNeaesRaVpYwojgrxk04eMqU0rrDN6lS43/dGsttn+2uzPmWtQahyskOfS4nie0+qRGu+Zb5OBZCoEK/WxGU0B1o8WKReEnhS0IiaDCKw+P6n4UynaKbeh+ICTtKD6u5h6/b/K4P7rYjaUMEl2yf/WPib2hnRiufHSI1Vm4RyXUXbSfkBBFmGzagtoI81Evzqd8w7tZlONoF0zBE/JX6RbzmuonnD8ZAJGGIhtj7EOq3Kx/hir42pfauqJz9ByeBAt5YHWaeZQEw9X14J5KAGHwVC4ZJxENXOwRCJoNcv+v7rMWdYCCljBQkbmNwatfzKDemMHRw1GyLhVEiIzXq9lhoY3A3ZO70KU9fHC1uWeHZX4bEX9BGLLQp+/LSR/M7ePe1US9EbI3LZe1BYdBu7nJoPogz4/QrsLE/Mu7ukx8NobL9fHqUJXHq3sIVPNZdZTTS8ynzv755+Y3NeT2uzGnEV04ALfJd9Faf1eH4JX7XYu+MB1lsqcr5KVc+OSjj7YnpsF0E9i85ej54g4i4GewbdnOkwr5Zau6/fmR9X3HIBvu2UCLrMsMUGFKgdCvLXwQm8Pe+ZR4v3XnexvYWWvcCcZ+giybQOQBF5M6by6WE2YnZlzuacMk5hLy4NpwBF0gPh7X0SvpaLwOAwU1hjAREsXtsnIDUOzH9ShfEn4Ld9fUmH4t/m65sUuvxTELGjOEhKpCBH20lpQdLl4zDvWdIFbPUK/eDVqBxUYr4MDJMGMM9SbRdPMWMCVudeWWtfTgjQ1FyxFrdWi2fiYZgLzzXN+5OwwF3EypEBWuJrVSRaa9kqHED3rlnpDb4iuwD1nlz0yrAT35f1UVlO5MxPHzdhaPLH1LhYvaILI73po6J00KqrywvmqexuF3E14iyONhvEwL3uzDwil9h3y0tnrlivDjTq4sXcachKXS7cZp78MZxm01wflXRcjXLreYdATs+oYxF1FRrxULOV8mqVmy00R/YkCWDgBjcDyt954LuJnx+nJO8/GrtwwWht8WuyARKChEtIFbtq9OWVLHSkFTRR4vsbDkQNyozk9kH2GPUnXPAmaUuGU6dVUhozR7C6TLkXCuBsbuHZjDi2LEiSy2KxqsmUBG3JAQ8LfSVCQgylkMrGascavUYgeSncOb7QQ4JAJb7rP2TJO7WG5T8Y870BIVzHUNMUE5R/BDZer/eh4UDzU1CfIcbTJKLlPYMmqnHfGu2vNG3EQXD3038G/8IYTF08MV7hwzTqDqM4IFaEG7eFoFXfXmhbAjVgmzpqK2s2e2BAtzCV/B56s4YZWLHfkrDWU8Z0TmWUOE5A5Bt7ZMdg6dptrO8J20JBCCHRsv6AbK3wtavwgGW5g8s6kjnTvA6duZ0seDBD221px7UnkvAFgsRYRKwyeAokRDVhn54ocBI6LyRF7Y2ST5y+Ny6hMLLY3VuwvMwYf+Vi/pMnTqOhLaloXFVYj3IBGSgoGtlm15GoVwv9ae+P5rVGRMYqXhgk79cTgCHt0KelHjXqa3Seh7MuPxo+jN4A3c9YO9bL8nvNV56BchfgLeaXgy+Wl/WICT0zac0LEZP4DG9Cr26xLMQxCcsdJWdplZfSoWSN2S0/hcQkjSILaZi13gORCdbz3o17DiqB9uEHswpVsI7UwUTMoz1TRRmL65nmQhaWRtpu98OgARE19AOj4qQdDTUMOzobdxOtsHAfKOMZxxR4u2Ie31zc0hwf8AHyVxJxEiIgYUgJdS+DPrxy5EUAjcU3fKXnZHIxP0+AG4o/UfLujcy1fkEeFNVDUlmT2z/18rywm866Z0u+8549ywyfwslM2pjuwyig1+dsT7TrBFIV6phs21fgBlpF591DN942MnGtOikcGeZdVlSOZ8xTTE9bAQsFAiXJ01PIE2R7ZPD/h0QRhxikS7yZOhzqIM6wrzSEQMbsoC3P41rnp7ttojH732Ph9ZiY4oYkiQCg/8MuHv7Rg1tTiCAugOhNYjfwRk0xOAOB0gPFA/y3DYv7pL0I54Ns7cWKMYX5aHpToIMLRHJlS9Oxzxp8hqXG302Zp8UPy6Xy0AMuiJLAgE4y78mzjwIAVgD9tcCgChq5gTtMDetTN/6fleKF9cTRdwIj7rnrzCVGGqM6WACJ6FsUh1vdvUfp2GTDQmT1IuVwz+n2a9ISAtwGVnsexuLDNndDbD2qEjePdzJSjWrV9x5pL21XNQlI0C7QlrLNCdqj367o5vlqXGJ2fYiolJVw3LIQLzdEclIIoAiiiFKsKTnLwkXvZ3XLVZB2XBsKcafQpdDkaFo36c2C2g8TctKq70IuKPMSD/mLi5RwvBmcu9HEMHiGhogxBaol3mR4tAAdoU9966TuIj7W8ROtTqW7TSVUQifdPpWirHtp5C3fE8UB+o1zBiGaJbe8gj+UwNVaQWMJ1GLPxqR+HML+Ak3PpIfT8IHX3+hFCtkWr+x2V9iWPrch8csLy6ooJet9fzPezQGaI0t0QIXdohSopBHKmfOB/DxLXU9URp7BZ0IFm5USMudgQFWVCPATxHEA66dEJeVOCvg0SvJ9Z54PFP39ZDWlBJ2BT51gohIG9BaeXtyQ/z4Vvx197iqBl9cCVicPqURsfojsQZijNuBmbZFHNA/16DTjB0wb6qTnJGAI60VicRAetTNYLoE/XD3vXJZQ3xlS82C6X3r89UI3ogHCPTcVJnb/AtfvJ81fUJN0DiDRGBRHUIGQH2T7d4zuFq17AMyavS56sHjolA65gF68fn6Og58BxR22PtGVoTbi+oGdjIOa93BXk4KYlIYcO/LO5Iuf92yXimITjoP1OfHwfN+yXTbOpbstI2KoZgipwWXtabG8VYsXPGNssx8a16phlVa66dT6kDzQOL/Kb3z7h2mDiwxmw1WxTEwOdsUqbwL/Ii+d1ZgnqvqZ1JqKHo1UZTknjHjEU2zgQQOAF2E6/V3fLX6O8dAGMZX6Ktay62I+aZNp0Hdx3oNsYJ/t3ZavJbVACfrwVAiN8GAWNHRRPXqq6pmJcUEFSojeJywxiMUujFmqxXdoeI8FE3QtuGeskhtcI+LxlMr4COZ793yLgu7wUU0TX/F3FNwX40ZcEGvgkW5YJcBRfYx/qszfefOMndDal5uuzJVQhhQh3Rc6XJhVA2IhytfkimR8QeifOu41aC2qBMeQ8kEVOwZkGkCCdWdZiHpvuj7M4CkLRVzh+Ern0ENH7P10/X2tGzARZseh3FLVstktktkduIjKmJknJEfVWJBW91UtgNsp/l6V7XLwQrdddDV69O+Di1ia1CXxvciB6bFHTiRUvE3YPoDNzf61zB40VCMvAvj04PmkVtqdtAuFUUxeK3c8C1JqvSmnVcd8WAdMvAmSYOOm5vGQlSGI7BxyJxq92CkXKpyVsJcAsCrnyYeEMuOy77oaZgHaeqVHh4LhPy7s3J+/GOK+Kp+3odq6Buf/d8TxUpeCU2MQCKJ7dcf0eTtMGSu90ySATRHDUbHOyowc6DLOWIziq+1NCL2hjfVZwCH+NiF/9TyxKmzHgfBDoM+FOCw/txel36lphZC/VucjQJRnPMfYP16DX1+zId7Cn5MlVSEVO6ZZ2vbbyIBBMiB1R22iN9Zr1xey8OSZgSxiYdeyAoyCYOypzlGCDXH9NRwEn4vVhOAjYHaAnDzkfiRB5cOYfNW4Vhois87yWLwg1fBHK2bWynAizR1qvTI7n+QD2ZX20q89AbSuqhPCDHbv4nO27s/+Ak8rourCK3ln3JtCuBka5Fk2BEGgiMFvygIEiQAE1NhpV37FihnSIVbuzZThil5Ut2pqdH6euN56kRd/KHlFI0bdUXdMuFRIyNZ2ztfM8OOhBaKZb9w0ux1fy+4pYT9XI44OZjcL5F/v4xog2Ja3eap5e2SD3mCpZf6wtwBXNC1cDDDLlVZQGEoQzsm3tmEIt3Powsa7cXoS071qAVBugjEsAkzw/ysmdqclh97Wsg8CqwAyNS4ijWHhin8qD/EDSyIXD1NCIE1/I1elWk630A7WYTcDFvePimGPT4tzzYeEaS92TIMOVIb5vISwMSetEZYaqdsZZsOjZq5yTqKvh4zmnpcp/KzyjcdfUIkXER1EU/3+TF9/+1OoJ02qFC92xsvomuFgnAQRyDsjWfnk79762ktclj4cZ41jHuWFnIuF4gnbP8sRmoGfGVk/Yh6IxGMCKc1TWseLmp2Wcp3ZDJeJ4Mi6/3aJbs0iOJj8ZXPmAKzn3QwKMqjDrAK/f5+9T3uW6Gqs2QC9ziix2qnMtnUw/LezOR9RobAK83PmcncA9SK7IeofQKqScyIROUfAZ2iFU7vuY0Utm5SGWCsnk2AZ4PbIciNWkFAG39t9jV5tck9NCwTrp5u9JhXjrSAKxP33HaZIZU3UsnI96LICA13x3ivgA0UJP6G200qGh2QR5P+egpYmGh8dX/9cUzRPWInQTJgfnaiGu422psUyTQJm0EDMMkvAVGR70b8YEsF3PYW4ganfIBHnNYgW/tI43lgIIaNv/IonIOwBzbEbbyGg1Iw+es4tmeDcr+aQ9YsQ4tTHQgOPrdH//eO10e0zo/8aUZXVqbw//N6PidXM/FgCojl/FCnvWjws3+Em+bXCBE6i9+crEXt0G0zhbLeTdO6lBbbQ7CTQNoITG0DmSmHluyVgN491A1AY8LnYlziKuC1dVRoRxHVvp3qgbfeFXe8v/FBQVH+ELyA1fRzM/EGCd4O6Eo4l9V79jBdlDT5MPSC1o9cJUXQEGzK3UCv30EfReRylrGUYeiFmlWRYP4ILjuLxC2ZL2L9Vhc0dH0Y3AoQYnPm0ut2HRVM6Zuu3clCyUuvPgmNMfdjCq+W2mq+bI+TnSeqw15xbAQ1F/NXiBYWRjnVGr16leRFPde8rTX0c0XyhmXYoSWjjccF8+R/a1Wdm0XZfxN4NyuHWAljjjnh3Tibj0gmtl0JyQBJYvd2P7oM3SbEzXMoNWhouh5RKDECfnxSmYfOSCRAu5a/Muo83R6e1wPDyPOK35CW95IjuQ8Lu63EKpuvGkmfBTRg2PLY/eJeR7DLjg5N44BtYgcFi2MZLChqIU6kerAM4L5vZVu7xko3f7gMvsykrHeZqJHENvJ0YsstEh9vQLXIY4IQcoP9+s3DY8dZuHOqTHV8SyWVYDCggBd4ca/hba6FTUqX+wjmQSSwXwvUeZkheZf1/2JINz3z/OYa9CQraQNbaHMW12AoNdHlJbwtApIgNNEaly0mq9OlvgQRbdLpkmf4f8w2x+GMF+YrWxczsEwSBtvccw1HeFLDqSAxA+AXjJ8ZSfdHBVR1n2+RN2Xe0fleg5SQB/P0TA1Xuc/CVmw4M2SUD6jY52PQ5gSSQJwF+912MsNXfst5OsOZ6Dv71dthiVE7ZDR3WvBZCfFyE1g1OXss9+G132wEUCfzTMxqBUSv0vrPFQtyHetSsjKgaKbH1X/xoD//cH3LHF0jMEWe8/mBAEwFIqBLlVSbjmBWfFiJ4lseigd98RhR6bPYtnC2TICcZ6z7bwcOxhehrjfpmqtr7v4kvf0JG0DKjEQoQWBoTJNfvyPX1X0E7PB4lVukPUBPUKGs8Is8OWqRl4o2/l3vAujJsgiwoFTys4fp5Jq1OisbYAgrnMLBG9LFGDT2IRIp4qUSrxu3xOg8wJovcyRNfr6FkRwk3/1++SNLCXM46MG5Hpolu7f0HVh450ifR+u2HHDjz1RfAYGsz9LHHkjj6T4MQVf1qXtIr/N4E6VOIfhOJcESGNQlXEobMGXaCc9imwi1mW908ZwBxj+1GeEkgsYq/H4wxCIXTJ32qE+bpU0uJvqGOpvKY68tQ8VMfieJZMqdi3jaEiIfvOZWA3nbaqn93ZXXf4W3NmtzspXY8jvYySh3BsOHKnnWyB7iKPKp7Y5blU5nO+HWbJrLESFr9sZY3oo+/CVE93Q0DyqmfeWtdj2JeR3sZIzzksURNp8QIn3F/gUjH9T+CURLjWGT0Ch9+V6M7xVtA+X3Y9NaQwZd/CVmxJYTTcCT0IbNDMGVmx8XhktXYp4X1IHSUm+6pdekW8hInZ+Kkh9yriW/RfEs8MqlmenwSl9F8QhlsNmCoV/xccO1u1DxAjw/F6qTJwDhXzAQPt0xczXM3I/71n/LHal2KPMDRiCCTbW3xPo4MFcI+6PtcQazFQ7jggGrd4d8vXefWmtcfMFnFR60woikd8XtjAxJ6feglVxshL/jRS70S5+0vhWA83PveXfeliYUW33AZG9AgcbZzPV1bRv6awvf6C3b8mRa5c2w8cEllCw2gAlyDTa7eA47zxyzFxAyLvqVjCghA1HOcZO8W/Y0BmOME9BZILlfdVELApfJjunceDeHSKQDSrud0t3Wd2WU7EHwCL3s0lSOnMXGmIjQJbC5ZsWque2OpoYmODd3fUOv2QC19yTIM4aEoDAu0ZC8AmoM5da31o5jXR64VEpAJiRbTFRwvblJsazz6Tt/z/2bm2T3z1sqx3V8C+2dt0medsVYIKFroMW9ox9X2Y+vLy6JLUAHwH+oViuMt3FTHdjuBHE//IfBtGgkm2wUvCDzo9J9XmQm8ICvvEAzwIu6uBjTHgqOA5u/dQETcCbVRu+F71D/3FPCbEOszdYVcBSbZLddCqauuAd+zD5Ig1SHLJZPxI3Eaf5LOHN9BKqjBMLF26iWcLrOrUMj8kparIYegd2u65gKa0nQKyoIby4NMYmnvScw782NnnjAv4uvTLR4sah5Uw4EPU+CTSsWyUovMFwcOOkdbW6VwwSrZjtisMHQFlyYqX46jtyy2OSC1/tx/Ak2xMYOzqtqrMvKSiV1Psi0RECJx/GkH4mpDv0vln6STOPkUoAOgctF6W6K59klPH6x6g0MCYKML82+eUT/njQ/Pgu/jDFDjG5Bsrmm0i2ZWbGuy/yTM6DvmsLr+cSTRYgV3bg1FUGiAUAVUfhGN/CnkfiGvhnWRX02+TJ75mfOlLOYmE16DXDWjm+/4nl3y7zGJtKyxGVGVkuaukMOs3p7gU8s6x3FEqHBjwGA7Vo4T2Q5a4UNgc6cEaQWIpTDeVD9cueEFe7oomx+iVo2Q2dGH5CxeRgssE8Xi4tCkSkAfgx3vot8i3v9lf4qQ9cOXymE5vVzGHDZFcfDZcL1JL3ZuooDSXsAOO/BglVby6jiw4KVSUg+7gYQwTSeyWtQUFi34rag8IemCI9KXgKH/jHExY8R9oLM3mPKaF0BdbMcUBMsmHNbFM8fYgmUWCpftTO9KVm5uEdzksayKmmLd7UkL/g5ti5RgP4HC/2hqMQuMOSz2pfesfxWL6gNacI/Sbklg//YcQRkufE+faNXkU7c4KPkDCFo6xalNqx1aGCbEql5aTRHvKwAMvj7p3ocPHyyulejZJM7/gDufwEp11WTHvK9SP8iu/XIOjZTLq4vMWDTmjFTdcubja8PqBCluCuxu1mgFdQYS7XBa1eB03mq7tfEy4t0QqpP0z09IFcc0YvT5DZ6IXdluqGDfrFQZDXnf99dphseIzcrqlzBlPurgR+1COLR2R/LI8hDw08TGiaXSvpiMFSyi2tkne4kGRVH8W2niudJfXUjzL9jn9VtIuoi2hNDhY4XaPCli8C3pP/tkKZGQReavYjJl+y8ipw8oOczwCpHLUtS099couX3DxJkcTaxr8PeKb9nlJdp4JqAHqTaCIxsY+NWPVOxKOhf+tpeChmzQymx7G1ImNR1bERIH+W9NOSIbuSbxdAMMp4gwbQiCHIUppL46WWKciZpyNMqtDVzYv4w3HA1f6Q5X4BcZhbKYwdnY/gxZiuRo19Y3ghBFGc+gtT1Dqwc83sITdTyeA4Jl6+qFO2veTC0+ROY9HuOv91vfD3r2NmDAjgGBJ2eS8Fe69lCiNf/zY+tv5Y08ohJFqIUIcBgAMMPrPa9ib5hTUGTulKX1fn5/v3pflPs2pFfUSCQiN7C71hVrFVlmTnp5r3vBheQawxe8sr2tMpurPi8KqXUEC40iFCEz8c6QwIjG6UvNH5AacjESFMx0AKXNuMe+CxbFIMjNCrOTb3aofNEdDkHfKWv/0m6EoCbTZ8g+DclFzV+cdfDaIlroNke3ifiIXu7wWSCEA7CQL7CiijF2/0XiyJp9WNSZCK5T99GecFUgAwFPYzhayjVXDqCq4G2HDVNb+t/u1CpeYpXzW2Z13s3o42GoekyiA2hw0eLf/zh/KYKcj7MUIaOWj+3YeUKW1Q9ix1Xpbff3EvVw/sFqLrE7x1qkxVv19Rzm3IrG4D02pJKGLL+0LWIB/+TUDnnZLDnlk2RPyRigUM7kwotalU/Ch+FlFo7clZxij5/akCNmgSug/i6o+VGeh2XDjmoRPslytDvPsyb7zIZqcsxaFbGHBHHGnrBrkJNLSkC31Zl5HnbaKChd8Wa74QQCU3C86eqY+jQFZCGC2xLgMzPbTlBzSE7F8ir+zedkGouEpyrHrgbTeasm4epomGWdzlTUh8ImGTBde95xRJCbrA+CFKD4Cn13KA3SbfU8x/7byMtFjAmXLSXsnG0+aVD+no5vMvn03kdsT+RSLFV+lRqLgjs9FwBPmhnIrrRW65X9ShPfghOeTfiEuqx2ldKEt4a0v4H8azBl5bHuiRxqxcLMZHjCkVWrkVEWuRc9Fdib1IojE+jyLlB4YebbMta4YE4434MgMyr2e+RNa8+89RYr8mPLV8b/BGAW4K657c1ZUbOXM8Zu2aTvHXkowO8KI67jisieeHtD3BTpPQvr+xm2TruAzX2ItX7oYi5xiWXGGRyeWoPRRq98GnGgZ1o78/CnLtZ8xYNjn2uBRJkBtS3piLOsQUF66I2qjCoaMGNJ4YcWlx2H5bnXbh4KOQ9FFttpHuD/i7xt9JGENmGSsucT51rLxGDlVcgZPolX7djsOGwshVCqwDSY8xhscC9vxjsezz8JczorBGJnWV811ao6tSrCPYAaSJ/9L39cxOAANZMgU0U2pE/GK0fIQZ43t4FcHXOugmdVmI0PbRMcGRG+7zaEdjQYhk4TIcoxNkpM8QckOVw4Vj/KbxtE1M7MTn1+LRmZoh8KjS1BizHfWPC8OP8DG39SBfJdT9m2feqyUeqpV/YsW8/ZE4Xr+rSTjioa69A8bLPpGGvM+agWF+JOGJLtLNLFjTwETpEHER98EqWknm2pCjZsmsdBH8oHBBZLzibC335DD/OdSN+jdnA0ScNt+w6Kf6AM7HPFh6cY9Dq/14pyVoYbYP6SW8hIFz0pmASFvXeiNUac7qlySlGR4xxcbeyBAL2IUpP1zoBQXbugKuiV9bwqJES5yLFBLMtzvztumMSrzglsopiM1+dkzmn9anirhSIYgEMsgnssehS6HvEuVoKIOHYkcamslnPa84R+Ulom0Bdw+yZp0VSbmcA9RooHZ+tN3ur6h0TAWPpWzMTvBQA8KuipPK+/74FMuyggUxSaYYgTvXLyCbMIfP3tMZGAYhyL9e9uORGkphZuFG12NcOeSrCjC7IAzSAiiMMNCBmoytuvUq8WejrZ5fkcfek/8fq05QndUX64sBCqtRR4JaD0+I2r3C5sDBVNVd/EEUpTmaxbtHbQ1iKLgIUhe4KD/B9ldfNC/kN2YDu8TgPy819/ueyDRe35F1KmAUcn942Kb8HBxiv7Y5/js/MLs7KBkzfjjrO1tn4Cb6I2dzBRo0sxU8wu+WO4IHiqFqAedvmhHSlyISSiFJQbVwMQkLVvsmzfnk3O8TMQsUjzsRykVp1IJRWTwrwyDz1+QD767HqeBYtomMWxkL3Jur8BygwTKcQTwLZCUF3FZtXa8nW1oPE/F9RpC/iGbvPL/ZXUEWUJAeRyex8aqXl/36+hilLwErp06rWa8xQqGaurbKN5nlAv0nx6KSxXh67HWV5Rg7OCIrlURr2RvWkWcm+qTPX4iiep2pRtFFhyhpXkyAeROEmLov+ZjPnp6Cz8VdaoXHNnvbozodW3ypYVBXKS2vYWOnWlYlrKrgTFVvBDsayOpxwoxGyiUZnAbIfejDGoU/Yyfu6Dczje+Op6T/PAhvJoqMEeMXqlfWKEC5o6NB+0PQZbBNCp3KcebYZAxrxrJI31yzHUVUCD/93x7HvU2Fivkrdj8svd/T0y11E9XwKVtOV9anh+PHogNJCtDNRLFU7Cxg2S7ejZJfMAg6m9lyuQjFkDEsk/SCmRCmnoL7JGPf26T7Tkle4ndADbhKen7M+KCln/rDdQPpt4OUSwauY7XysnzUSXnfr9F9yk8dRplGWJdOgaFIzt6O0qPYFjWDxGwVt5lDD186ECjYMvPrzgsbNOOmBHHO/+BpaMT6JSL1D29kQcKrfshx7zrSxIl5dH9P+iayhYRIGfw1iT5RBpTypmrCG0cesKxP0k7CH32NsDaGxSRy8QIrQ2F9yVBiAt9mRe/VA709pc7uJZmiYwUCflZU64I3BJ/t2nzu30kwkGPRek6MLRxZ/LqfcqsAa/yBd5BuyssM94EVlq2EUZoio8UQN9nERc/jsDFmLVsE7RFzEUt+VwyQ4vRADUlQOY5RSgcIYo/oZV/TJLYxQOPG220/WQNxRzHYo4jQ0uIxLkt98y1eIEnS/YUVs5qVDPxGWyJ/DBBOtaBd8DwgRskrX2arduFLoknUn+6DHJz1yVFlfoVB7a2dd4s28Im41xZ66Y492Jxqesxfw6zO/NtnrwPVi+MT9Q3gWElhlZGaphCP0VZwmdy8Wgfm8RCg/UzTBu4Nerm6tuKdmGCFNaCGOFqiXNonGTd9pPPuljxP2/XV5S8k1NtAkHpZ/gNARuNHcwJVzWEeY9xJfCWCZmbPkyEHMsHHMHR/2Ez+dQ6IXkC/teO5Ja59RHJrkgc2PRYr9+Y4xHpYGfhB/CLhsntm5hfSvZ89t328zRnyT7HIEoZwM/hSC8CgFfe20TJQZeSCn+3iOEob22eD+wVaaOBm0caskbvRqWGZNNoXNG7X8cGDQ/AioDZUDFQJCC3urzDUsq1wwvUlHI+Th2/o1k1BqZXk9eHwOybPfuNoF2LxHfbg5NcL7AYnLG+ib4Gadg5YyWDUd1m+rRwcFkWc5Og4Q3CoRsfmzxGhtL6emeEMGPsems7KvQXI/YUKRy05TYT/lmACqimq6lpyYdECGpNVKjAaRO51uQH3eZZdfaxFwF2IZxq2ZzTijCb+gJUz9AZU5aMM16r9+pd3XedHlisBAe2jMBL4xvx7axHJAJ/dym2cDUs5Xy1BFh3qBnKwc1XoimlXMoINH/dBVGqZvrusqnW0ZBiuGqJLaTN8Pp9j8XT+IJAMjjkgmqStTCKvohYrV55ihUcumuOPUm3i4oE1ZVsC5DmggTus8Wwgs65A7fCRqs9z+3v2f91nu0X0fb5OvilcMaDG7Jbxq99AgWnzrnd3oySuD4N2VZqQdPs5tmtPhzaZqNldC61mquPYzdXwKEEAWPXoYuaprUNATsuGcYsYCPYQ5XqnbuixoAP0dzPKyykC0IGVbFdOXB17xOk1VPp+rMoG/35orwGspmIP5OCL3yXKgbja2dhxD+rcL4FzVHT1xpRseWKRlLs0nURVVarCI+604zXF16lUR3WasYmgrUdwqkBUnMkIR7jdhXcknC7B1Aam194M/riv1jXALRNAnj1koIKUM5POJ2H7hoRhjXcdbfBe6GmkuTvWre2BXlwkJC8aRsHeY9W1KdLhJS9VS0R0U9vRMqmrx8b+IpA5B+dVRNAEOmX74IFunxx18evGCfBT8FKi5jgBxGEM2bG1I8SE8HzFbqMeuGOHyEjkFtaOPEDkSMBVERSjKd4QYPXj0W/GXUH+jW9C2o1/FfLQa/jpuw5T0oczaQvAtgpHkU7Z84c3TyqnTVFOQfZNgKGciYs9ir+KGX6VsvTsC+L8qsV+EcdDKPDC/scjdcqwfbLlepu0RlhTv7a5t7V/VD76yJApSWQsxlPyAwIiiXDh7X7D4qHaXUyO7RhQcxqOfbmCl3CMZEGSM9IdVnkNM7UVjakO5qVYyKImOxO/ntioSaZE3UBF2MZ2oZBYT1hfdpURsYV8cEnZCRZkWUNHbiPglMWq/Y3fK/DFwk9qJ0R2hci4IScm3xsvQ7yNdKj9aYJRgfyg2yVY6vlzxg+c6Bru1rOx35bR8y051aeyXtCpa57RguFIRFF5NIeAxfC85rN2ETrOXDyQHB9Vs6vtb4sTGOidubwrM9r+tj8wDBC3UAZ1vBmuEecwwkyKhnW9FdLUDV5VSoKahLv+M0ix+xB9afxULTy7Ty7n0rMjXeJ3HfQFPZuhRDI857KegY4Zt23tTsjm+SCD+9+oash+ZixQ5DoSszfJyS84TL0ycfKyKltGBCW1zDoZstdoL2cUsxv19muGiqSgLo7l9mVcGRpn5BO+yJYfnqI7/k/eSkQ2NMT5jkyrBHL2X10qLU4vkmI2RC7Pf4s/IHPbxVgwvqB+e4IltcWSw8cz1Cmi3Hh/4vxiR5brjlZl74uUOtDTnP/NMlORfMEI2ZO7rUG2OY5aOnNXqQaAnh1w2dGQzGTaUKtjaY04CMK8P7wm20BcdfI0CeYBWBk785qFS7jvL0ssuTjQdMnmu144ROWkZE74Th4eQbR+Tmp28YshTGiVp1Y5wbT14raMxbueh+pJ7xHQYC5E2EJwg76Zxk10oUR9d1PSKQFkKn/hK6uYCKIzndnI/DmF/ASd5IEwUcsn4WY3wYj5iLcgtJR8uWF6CJtudcnArJ6JO1LMZIhb2PyLUU5GkZ7lXAee+W/Y1d4xRSJuoDDPmEqgD4Vl/aMzEtUYeDZi2Ou2pSYQF62DZIFAAAAAAAAAANL1jo4iBnbH4ONQgpNnk3HtOB9ItUnh3wYL+tLHtCzDd0ABb8Ehcz6iQYEOCOatkRPf1pCwDtlryOedXfzZjZ3WAKtlTw2eq//X+pYY4ZJzMkCsuvqhFiYlZkdqnk1TziQKRAz/oshZtrqtjrmTOAFZCFMFVGRHgUkqiWIVLXbaRBswgkDpdgDnGPDo/bSFAsQAUVW3xzCh87igekFkrqkSPTrO7jK8ZwtVO4D1hmPpWPwDXGSiofShLvlDy+Gn+AXuocUgjJuOTtilpmSCQ37YlLXW4UrGRWG6zSkCzwpfoSCe7fEYR8RbyN3CGGwViYgVcnKje6BNp9Cnf3HPAYAlNGEqOC0QUaeUTNEBE+UEh7E01ce4e7fg78vcvvV13ekJu1uYjVe5v3hG3kVonxoegOORcGffKv1EZ1pN1ANALwyP/7VaekuHfMnun6SFocxejscoDtBrP1kwvMF6PUaslwXcI16U4vt7WUCK0jHD1WwRi9I11d9V6xbaHmGn0slTyr9gcGYq2wxeRAtBtN/trYapG87wEk/XjxxLoN8R7S0QGBO6botx4RtbpVJ3KjZgyJATC07soouRP4HQBAiX/XI8z57gpGhJk5ubqzZ/ZVBzS6grMOasKI8WHvfEWidlxRcsuv1XS2C2pLfZsUUoRnM69k6rzIACEw/ToD9XaQRTmiaeK0jiZ0rq4o8TtGsG1FL2lwHze5xSJzgMm/OnP/qEgWmoDfrVvijfKvpmp/IdIHEyGH80oraES/YNYXMDcWu/YikA06bBYpIIRRVxU230PW2eWuoq/FbYZEykbzRjiIibjVXHFAMvqxZw7gyvlb3wX/tos9BDMwdVevZT6VZxJ+zOZC39ZWA6qeqx7uQOyDbKJ5RQAAAYQBoJ5lGYCnEFCmvbAymX7ZqLR37pyDsDuj64AtZDPKPtryvgSgcz0pHTw/cO5NFVQEe4zJekkxFiwUJytrrCHwDEhrwkEqTYwUAgU0OwjMdJpCP656X2lbO3h5lKC7Bae4rJCbSXUntqA7gfQLEHlCdzN2e3aLsYk+181nrqNADAw8zmNqH0GBC1rjOtrF3f0e/CzCrd7SOr+Ftz+UMi/n26CjqE5k+J3lLg6oHg62v4KxXl+WuppILZEm6ZIMebX9BkB0Nq63QJaBYt3+AN9NJhrkQVAYrlEi1rN5p4oncqY7WdqONWZDmZgpAAAAAAAAADg8VNWF+GMInAKn2m/V+PLf6LfqN9CpyCZEWzQk0HLgOtBRuUZ20YdLeYOvPrn90a2EAAq2Zi/Ix+BFJImwWCtdoUTjrwP0NWsLZB8/vwm9QU5GMuTarrc5Gu2o3JXOCUQAAAAAVAwcvjVLbC+A516vCCcpDervduyX85v0mkGbtIQRzAAFCl5ky0aIHUcgMft8ZGGxZwAGprRbRUcpbsYkNVBtiU7a297m8zjs/AO36q1AOBfDB7XMH7xcE7FHtNh2G/ZYjEN/Ztci9wZ4Ft/GuwOxWBIbgUkM39+5/d/zf/IUAAAAABXd9+HxW0jkZSn9w5Je+R9w55NFgX2YG2DS8g72wIRILtcCtQfDXefe4C+q+wQV+30gpO6kk8I8qqFAAAAC44VpigAAAB+aXF5CAAAAHvang4AAAAvpyYKFAAAAFVerrMAAJMj0A8jOVTqLHxBzbNSIBh7hyfv/z65pjqkGNmcAf6lu2zrDMsBObB8rs3U9x14S16oVKYAS8o8rTJ4LbQFfYj3ylOsGqCOahCpwG+yIYSzblxXivLq5fIFYAcXC7xzwY8AF9tHjtwAIFECkRpv8BHhcHO2+mO1BD/6wUGl9YtbB9djMjE4W+PVeSCZLn46m49gy25pBG+CFYhvvV3zGgTbO4sJLyz6DCbE7qGMhq+o1Eo7pwACHpVlmAAZKZ9atfaNr7uC2pgf/70IWhgnmEYc+rITmJ3yRtk21Z70gZ8KD8Qd1LKOfmCD6qUHGq3oRCmqDRbMwDwrsvH7CHLoVFm1Axl5QAAvrDEKFAARkBvrbNmj3fW9jlO1sC3kjx8Cp2HPJuUD7ZbbsCKWOtvEH4PsqbFFr8igDNOu3pWJPNd0zZiWjBoqDK7QrGmevci7JYuGiXrGOAQoDWcs4xBBef6d6k46IKbQAZcMHnppgorqtOuyBE32lLFiStKNbGU3OxgiA2aTZ9CSEkYSLDIktmeDGX5oVcgFz/XXEUg28+k67GBrA3qtjtzWiOfccP37oJxSvDENJDn7JrQ0UZoRv19Lacfx2/SoNNnHRn2/9adWseNjJ4A6mV3sgpLwxnC4IpveqGSUI49eEFyTRQxuTqs3UpMl3vc54KHnVTsKD6JjFI5s09vD7mIcQkj0BeLAt1CE8YbGLpRAlJQjj14QoLosgpMNAMRQyJgH+046Og4zMM6cPe1l1ZuyWlKxHCtvdXCVW9qKg7RPasEr3L/F0JC1a45ExhGX/BqalXsmy9HLtw57a64rdL8ahxtyRiabrGBcU3R+EgSw1wkbZXsSC5urqdqC3ShrwUfImCi7zBQrTmuV3XZvrvhusYFxTdH4RmBnx9xaBare1FQdkn6B9JfmQzngP/7zvXBOwECiP44Vb1VaR6MBvrbNCAmYYXu8V7mbvQ69H7BiBvihoUCESXTeA3ePik1Hmq76Ih+RwaOZR1H4ob2JNkhxSZV/wh07ZbMGocysNY8AAAAW+8cCJiAPJx0gLqzLvWgITpH52B6DgkG/ZKbswY9tTivGRccxNCek1anrGYdZlzWF/rhMoF+KBfpFvSXM1ZNXn3dVxG87R4fSuDpvIV/twVHApKFDUzSet9EJ2QQfHw+upWdF/KGv6Sb7JIg2VF9kmUpiG8PrGGLd/d+KRff7R1A2fgjUdCaTtuCsZz6gEEHx8GZ9yllcxqjZcOG3b3XQ18FavOiSB3yyXQFDhosgplHjMgphTZOGGyMP+EOkMjOM0q7+IeWPMh9JG6FYzAYB9WA5fNFJ/iY2qGzDUhmcg+2JHgH5eihyK8qmwXnIciMTs1ms1ms0CuZ4oCns780DC9l9wfews+FxbnCGo40EgMcf7+6nQMM8Ii8ZRDWWZrvTgqEu1MU9ueXa+RM0gdM34t/j/4jb82adVCCGVRllO3MM1ENQgPF/ILIPNm6LuaboZ8fcWgW2H73LhoRr3fSwc2xG7mqL8QgTGEFTPdAMKt6qtJKqEXxmZaNjfa4hWribxn1AdT/6rCPcWR45hMbF2b9tRn6F/ia5GgBBO0YdSHRj/mo2Ol3dUzw7MKOwku5MkUgB3AAAAAAAAABnj6EYCZ2Wu5TsC3kUl7aqgEPgi9xf6NsIPGpwvfXFtnevYMK4ydndu8hTV+im2JgJxTyKHdpWAlNjxWlk0ZewLeU1LfP3OFSS3p5xMv5FIOirI8m9B37TOIECsHGWfRkJfRXAH+nxujFPVB25F1xhSxCC66D4oftPm58+JNnFeaaygSIofZiZ7ZAP4A7eUtDHveQjTV390H1jMkE+wlntBSx+28cGcb7VCU4AAAA=\n",
       "\" /></div>\n",
       "<div style=\"width:100%;\"><img src=\"data:image/webp;base64,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\n",
       "\" /></div>\n",
       "<div style=\"width:100%;\"><img src=\"data:image/webp;base64,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\n",
       "\" /></div>\n",
       "</div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dset = exp.datasets['train']\n",
    "# disable default loading\n",
    "dset.set_channels_enabled()\n",
    "# run pipeline\n",
    "fr = dset[4]\n",
    "fr.apply(exp.tr_synthetic_and_show)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# the same process but write to a file\n",
    "fr.apply(exp.tr_synthetic_and_save)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate whole set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "exp.discrepancy_dataset_generate(dsets=exp.datasets.values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "\n",
    "Complete code to generate the synthetic dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.a05_differences.experiments import Exp0521_SwapFgd_ImgAndLabelsVsGen_semGT\n",
    "from src.pipeline.config import add_experiment\n",
    "\n",
    "class Exp0551_NewDiscrepancyVariant(Exp0521_SwapFgd_ImgAndLabelsVsGen_semGT):\n",
    "\tcfg = add_experiment(Exp0521_SwapFgd_ImgAndLabelsVsGen_semGT.cfg,\n",
    "\t\tname = '0551_NewDiscrepancyVariant',\n",
    "\t\tgen_name = 'my_discrepancy_variant',\n",
    "\t\tgen_img_ext = '.webp', # better compression\n",
    "\t\tswap_fraction = 0.75, # instead of default 0.5\n",
    "    )\n",
    "\n",
    "exp = Exp0551_NewDiscrepancyVariant()\n",
    "exp.init_default_datasets()\n",
    "exp.discrepancy_dataset_generate(dsets=exp.datasets.values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training\n",
    "\n",
    "The Experiment class (`src/pipeline/experiment.py`) is used to train the networks:\n",
    "\n",
    "```python\n",
    "from src.a05_differences.experiments import MyExperimentVariant\n",
    "MyExperimentVariant.training_procedure()\n",
    "```\n",
    "\n",
    "Weights will be written to `$DIR_EXP/0551_NewDiscrepancyVariant`\n",
    "Checkpoints are saved every epoch:\n",
    "* `chk_best.pth` - checkpoint with the lowest loss on eval set\n",
    "* `chk_last.pth` - checkpoint after the most recent epoch\n",
    "* `optimizer.pth` - optimizer data (momentum etc) after the most recent epoch\n",
    "\t\n",
    "The directory will also contain:\n",
    "* `[date]_log` - predictions for sample evaluation frames indexed by epoch\n",
    "* `training.log` - logs from the logging module, if the training procedure failed, the exception will be written there\n",
    "\n",
    "The loss is written to tensorboard, in order to display:\n",
    "\n",
    "```bash\n",
    "\ttensorboard --logdir $DIR_EXP/0551_NewDiscrepancyVariant\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Exp0551_NewDiscrepancyVariant.training_procedure()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}