visualize_label_distribution.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_short_names = {\n",
    "    'VisceralSC': 'VSC',\n",
    "    'LiTS': 'LiTS',\n",
    "    'BCV Cervix': 'Cvx',\n",
    "    'CHAOS': 'Chaos',\n",
    "    # 'CTOrg': 'CTO',\n",
    "    'AbdomenCT1k': 'A1K',\n",
    "    'Learn2Reg': 'L2R',\n",
    "    # 'SLiver07': 'SL7',\n",
    "    'AMOS': 'AMO',\n",
    "    # 'Empire': 'EMP',\n",
    "    'TotalSegmentator': 'TS',\n",
    "    'VisceralGC': 'VGC',\n",
    "    'BCV Abdomen': 'Abd'\n",
    "}\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_labels = {\n",
    "    'VisceralSC': [[\"spleen\", \"right kidney\", \"left kidney\", \"liver\", \"aorta\", \"pancreas\", \"trachea\", \"urinary bladder\"], 'VSC', 127],\n",
    "    'LiTS': [[\"liver\"], 'LiTS', 126], \n",
    "    'BCV Abdomen': [[\"spleen\", \"right kidney\", \"left kidney\", \"liver\", \"stomach\", \"aorta\", \"pancreas\"], 'Abd', 30],\n",
    "    'BCV Cervix': [[\"urinary bladder\"], 'Cvx', 30],\n",
    "    'CHAOS': [[\"liver\"], 'Chao', 20],\n",
    "    'AbdomenCT1k': [[\"spleen\", \"liver\", \"pancreas\"], 'A1k', 1000],\n",
    "    'Learn2Reg': [[\"liver\", \"urinary bladder\"], 'L2R', 8],\n",
    "    'TotalSegmentator': [['aorta', 'duodenum', 'esophagous', 'gall bladder', 'left kidney', 'pancreas', 'right kidney', 'stomach', 'trachea', 'urinary bladder'], 'TS', 1072], \n",
    "}\n",
    "\n",
    "test_labels = {\n",
    "    'VisceralGC': [[\"spleen\", \"right kidney\", \"left kidney\", \"liver\", \"aorta\", \"pancreas\", \"trachea\", \"urinary bladder\"], 'VGC', 40],\n",
    "    'AMOS':[[\"spleen\", \"right kidney\", \"left kidney\", \"gall bladder\", \"liver\", \"stomach\", \"aorta\", \"pancreas\", \"esophagous\", \"duodenum\"], 'TS', 200],\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "cp = sns.color_palette()\n",
    "colors = {**{n: cp[i] for i, n in enumerate(train_labels)}, **{n: cp[i+len(train_labels)] for i, n in enumerate(test_labels)}}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['aorta',\n",
       " 'duodenum',\n",
       " 'esophagous',\n",
       " 'gall bladder',\n",
       " 'left kidney',\n",
       " 'liver',\n",
       " 'pancreas',\n",
       " 'right kidney',\n",
       " 'spleen',\n",
       " 'stomach',\n",
       " 'trachea',\n",
       " 'urinary bladder']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels = [xx for x in train_labels.values() for xx in x if type(xx) == list]\n",
    "labels = [xx for x in labels for xx in x]\n",
    "labels = np.unique(labels).tolist()\n",
    "labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_829302/3544833302.py:86: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.\n",
      "  ax_train.set_xticklabels([labels_short[l.get_text()] for l in ax_train.get_xticklabels()])\n",
      "/tmp/ipykernel_829302/3544833302.py:87: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.\n",
      "  ax_test.set_xticklabels([labels_short[l.get_text()] for l in ax_test.get_xticklabels()])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set()\n",
    "\n",
    "fig, (ax_train, ax_test) = plt.subplots(1,2,figsize=(10,3), sharey=True)\n",
    "\n",
    "sizes = {\n",
    "    'VisceralSC': 20,\n",
    "    'LiTS': 20,\n",
    "    'BCV Cervix': 10,\n",
    "    'CHAOS': 10,\n",
    "    # 'CTOrg': 'CTO',\n",
    "    'AbdomenCT1k': 100,\n",
    "    'Learn2Reg': 5,\n",
    "    # 'SLiver07': 'SL7',\n",
    "    'AMOS': 30,\n",
    "    # 'Empire': 'EMP',\n",
    "    'TotalSegmentator': 100,\n",
    "    'VisceralGC': 20,\n",
    "    'BCV Abdomen': 10\n",
    "}\n",
    "\n",
    "labels_long_short = {\n",
    "    'aorta': 'Aorta (A)',\n",
    "    'duodenum': 'Duodenum (DD)',\n",
    "    'esophagous': 'Esophagus (Eso)',\n",
    "    'gall bladder': 'Gall Bladder (GB)',\n",
    "    # 'inferior vena cava': 'Inferior Vena Cava (IVC)',\n",
    "    'left kidney': 'Left Kidney (LK)',\n",
    "    'liver' : 'Liver (L)',\n",
    "    'pancreas': 'Pancreas (P)',\n",
    "    'right kidney': 'Right Kidney (RK)',\n",
    "    'spleen': 'Spleen (Spl)',\n",
    "    'stomach': 'Stomach (Sto)',\n",
    "    'trachea': 'Trachea (T)',\n",
    "    'urinary bladder': 'Urinary Bladder (UB)',\n",
    "    'mean': 'Mean (M)'\n",
    "}\n",
    "\n",
    "labels_short = {\n",
    "    'aorta': 'A',\n",
    "    'duodenum': 'DD',\n",
    "    'esophagous': 'Eso',\n",
    "    'gall bladder': 'GB',\n",
    "    # 'inferior vena cava': 'Inferior Vena Cava (IVC)',\n",
    "    'left kidney': 'LK',\n",
    "    'liver' : 'L',\n",
    "    'pancreas': 'P',\n",
    "    'right kidney': 'RK',\n",
    "    'spleen': 'Spl',\n",
    "    'stomach': 'Sto',\n",
    "    'trachea': 'T',\n",
    "    'urinary bladder': 'UB',\n",
    "    'mean': 'M'\n",
    "}\n",
    "\n",
    "train_label_distribution = {l: 0 for l in labels}\n",
    "for ds_name, (ds_labels, _, n) in train_labels.items():\n",
    "    bottom = np.array(list(train_label_distribution.values()))\n",
    "    ds_labels_plot = [np.sqrt(n) if l in ds_labels else 0 for l in labels]\n",
    "    ax_train.bar(labels, ds_labels_plot, bottom=bottom, color=colors[ds_name])\n",
    "    # ax2.bar(labels, ds_labels_plot, bottom=bottom, color=colors[ds_name])\n",
    "    for l in ds_labels:\n",
    "        train_label_distribution[l] += np.sqrt(n)\n",
    "    # ax_train.set_xticks(ax_train.get_xticks(), rotation=45)\n",
    "ax_train.tick_params(axis='x', rotation=90)\n",
    "\n",
    "test_label_distribution = {l: 0 for l in labels}\n",
    "for ds_name, (ds_labels, _, n) in test_labels.items():\n",
    "    bottom = np.array(list(test_label_distribution.values()))\n",
    "    ds_labels_plot = [np.sqrt(n) if l in ds_labels else 0 for l in labels]\n",
    "    ax_test.bar(labels, ds_labels_plot, bottom=bottom, color=colors[ds_name])\n",
    "    # ax2.bar(labels, ds_labels_plot, bottom=bottom)\n",
    "    for l in ds_labels:\n",
    "        test_label_distribution[l] += np.sqrt(n)\n",
    "    # ax_train.set_xticks(ax_train.get_xticks(), rotation=45)\n",
    "ax_test.tick_params(axis='x', rotation=90)\n",
    "\n",
    "ylim = ax_train.get_ylim()\n",
    "for ds_name, color in colors.items():\n",
    "    short_name = (train_labels | test_labels)[ds_name][-2]\n",
    "    n_samples = (train_labels | test_labels)[ds_name][-1]\n",
    "    ax_test.scatter([1,2],[-1000,-1000], label=f'{ds_name} ({short_name}, {n_samples})', color=color, marker='s')\n",
    "ax_test.set_ylim(ylim)\n",
    "# ax_test.set_ylim(ax_train.get_ylim())\n",
    "ax_test.legend(bbox_to_anchor=(1.01,1.032))\n",
    "\n",
    "ax_train.set_xticklabels([labels_short[l.get_text()] for l in ax_train.get_xticklabels()])\n",
    "ax_test.set_xticklabels([labels_short[l.get_text()] for l in ax_test.get_xticklabels()])\n",
    "# ax_train.set_yticklabels([int(x**2) for x in ax_train.get_yticks()])\n",
    "yticks = [np.sqrt(x) for x in np.arange(0,7200,400)]\n",
    "ax_test.set_yticks(yticks)\n",
    "yticklabels = [int(x**2) if int(x**2) in [400,1600,3600,6400] else '' for x in yticks]\n",
    "ax_test.set_yticklabels(yticklabels) # [int(x)**2 for x in ax_test.get_yticks()])\n",
    "\n",
    "# ax_train.set_ylabel('Number of Clients')\n",
    "# ax_train.set_ylabel('$\\sqrt{N}$')\n",
    "ax_train.set_ylabel('N')\n",
    "# ax_test.set_ylabel('N')\n",
    "ax_train.set_title('Train Label Distribution')\n",
    "ax_test.set_title('Test Label Distribution')\n",
    "fig.subplots_adjust(wspace=0.03)\n",
    "fig.savefig('/mnt/ssd/git-repos/FedUWAvg/images/label_distribution2.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "fba",
   "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}