results_plot.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6947f463",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F \n",
    "from torch.utils.data import DataLoader\n",
    "import torch\n",
    "import torch.optim as optim\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from datetime import datetime\n",
    "from torch.autograd import Variable\n",
    "import os\n",
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "import argparse\n",
    "import random\n",
    "from sklearn.metrics import f1_score, accuracy_score, roc_auc_score, precision_recall_curve, auc\n",
    "import sys\n",
    "import os\n",
    "\n",
    "from collections import OrderedDict\n",
    "import matplotlib.patches as mpatches\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "699552d0",
   "metadata": {},
   "outputs": [],
   "source": [
    "simband_ppg_roc_auc = [ \n",
    "[0.879, 0.878, 0.881],\n",
    "[0.891,0.889, 0.892],\n",
    "[0.914,0.913, 0.916],\n",
    "[0.870,0.868, 0.871],\n",
    "]\n",
    "simband_ppg_pr_auc = [ \n",
    "[0.841,0.838, 0.843],\n",
    "[0.847,0.844, 0.845],\n",
    "[0.865,0.868, 0.868],\n",
    "[0.799, 0.796, 0.803],\n",
    "]\n",
    "ucla_ppg_roc_auc = [ \n",
    "[0.918,0.916, 0.92],\n",
    "[0.92,0.918, 0.922],\n",
    "[0.924,0.922, 0.925],\n",
    "[0.872, 0.87, 0.875],\n",
    "]\n",
    "\n",
    "ucla_ppg_pr_auc = [ \n",
    "[0.768,0.762, 0.774],\n",
    "[0.778,0.772, 0.784],\n",
    "[0.773,0.768, 0.779],\n",
    "[0.607,0.601, 0.614],\n",
    "]\n",
    "\n",
    "\n",
    "stanford_roc_auc = [ \n",
    "[0.763,0.761, 0.764],\n",
    "[0.764,0.763, 0.766],\n",
    "[0.877,0.876, 0.878],\n",
    "[0.883,0.882, 0.884],\n",
    "]\n",
    "stanford_pr_auc = [ \n",
    "[0.582,0.577, 0.586],\n",
    "[0.613,0.609, 0.617],\n",
    "[0.726,0.723, 0.73],\n",
    "[0.726, 0.722, 0.729],\n",
    "]\n",
    "\n",
    "simband_ecg_roc_auc = [\n",
    "[0.724,0.722, 0.727],\n",
    "[0.721,0.718, 0.723],\n",
    "[0.747,0.744, 0.75],\n",
    "]\n",
    "\n",
    "simband_ecg_pr_auc = [\n",
    "     \n",
    "[0.621,0.617, 0.625],\n",
    "[0.675,0.671, 0.679],\n",
    "[0.732,0.729, 0.736],\n",
    "]\n",
    "\n",
    "ucla_ecg_roc_auc = [\n",
    "      \n",
    "[0.89, 0.887, 0.893],\n",
    "[0.905,0.902, 0.908],\n",
    "[0.927,0.925, 0.929],\n",
    "]\n",
    "\n",
    "ucla_ecg_pr_auc = [\n",
    "      \n",
    "[0.726,0.72, 0.732],\n",
    "[0.749,0.743, 0.755],\n",
    "[0.765,0.759, 0.772],\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b725051b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# plt.clf()\n",
    "# plt.figure(figsize=(10, 10))\n",
    "\n",
    "# # colors = ['#46237a', '#42ab8a', '#f9961e', '#f2712d', '#f94045']\n",
    "# colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "\n",
    "# methods = [\n",
    "# 'ResNet-34 (PPG)',\n",
    "# 'Deep Mut.',\n",
    "# 'SiamAF',\n",
    "# 'DeepBeat'\n",
    "# ]\n",
    "\n",
    "# fig = plt.figure()\n",
    "# # fig.suptitle('PPG dataset comparison AUROC-AUPRC')\n",
    "# ax1 = fig.add_subplot()\n",
    "# ax2 = ax1.twinx()\n",
    "\n",
    "# ax1.scatter(simband_ppg_pr_auc[0][0], simband_ppg_roc_auc[0][0], marker='^', label='Simband Dataset', c='w', edgecolor='gray', s=50)\n",
    "# ax1.scatter(ucla_ppg_pr_auc[0][0], ucla_ppg_roc_auc[0][0], marker='.', label='Institution B Dataset', c='w', edgecolor='gray', s=50)\n",
    "# ax1.scatter(stanford_pr_auc[0][0], stanford_roc_auc[0][0], marker='*', label='Stanford Dataset', c='w', edgecolor='gray', s=50)\n",
    "\n",
    "\n",
    "\n",
    "# ax1.legend(title=\"Datasets\", loc='upper left')\n",
    "\n",
    "\n",
    "# for i in range(len(methods)):\n",
    "    \n",
    "#     print(methods[i])\n",
    "#     edgecolors=None\n",
    "    \n",
    "# #     ax2.plot([simband_ppg_pr_auc[i][0], simband_ppg_pr_auc[i][0]], [simband_ppg_roc_auc[i][1], simband_ppg_roc_auc[i][2]], color='gray')\n",
    "# #     ax2.plot([simband_ppg_pr_auc[i][1], simband_ppg_pr_auc[i][2]], [simband_ppg_roc_auc[i][0], simband_ppg_roc_auc[i][0]], color='gray')\n",
    "    \n",
    "# #     ax2.plot([ucla_ppg_pr_auc[i][0], ucla_ppg_pr_auc[i][0]], [ucla_ppg_roc_auc[i][1], ucla_ppg_roc_auc[i][2]], color='gray')\n",
    "# #     ax2.plot([ucla_ppg_pr_auc[i][1], ucla_ppg_pr_auc[i][2]], [ucla_ppg_roc_auc[i][0], ucla_ppg_roc_auc[i][0]], color='gray')\n",
    "    \n",
    "# #     ax2.plot([stanford_pr_auc[i][0], stanford_pr_auc[i][0]], [stanford_roc_auc[i][1], stanford_roc_auc[i][2]], color='gray')\n",
    "# #     ax2.plot([stanford_pr_auc[i][1], stanford_pr_auc[i][2]], [stanford_roc_auc[i][0], stanford_roc_auc[i][0]], color='gray')\n",
    "    \n",
    "#     ax2.scatter(simband_ppg_pr_auc[i][0], simband_ppg_roc_auc[i][0], marker='^', label=methods[i], color=colors[i], edgecolors=edgecolors, s=50)\n",
    "#     ax2.scatter(ucla_ppg_pr_auc[i][0], ucla_ppg_roc_auc[i][0], marker='o', label=methods[i], color=colors[i], edgecolors=edgecolors, s=50)\n",
    "#     ax2.scatter(stanford_pr_auc[i][0], stanford_roc_auc[i][0], marker='*', label=methods[i], color=colors[i], edgecolors=edgecolors, s=50)\n",
    "    \n",
    "    \n",
    "# patches = []\n",
    "# for i in range(len(methods)):\n",
    "#     edgecolors=None      \n",
    "#     patches.append(mpatches.Patch(facecolor=colors[i], label=methods[i], edgecolor=edgecolors))\n",
    "\n",
    "# legend_properties = {'weight':'bold'}\n",
    "\n",
    "# plt.legend(handles=patches, loc='lower right', title=\"Method\")\n",
    "\n",
    "# ax1.set_xlabel('AUPRC')\n",
    "# ax1.set_ylabel('AUROC')\n",
    "\n",
    "# ax1.set_xlim(0.5, 0.95)\n",
    "# ax1.set_ylim(0.75, 0.95)\n",
    "# ax2.set_xlim(0.5, 0.95)\n",
    "# ax2.set_ylim(0.75, 0.95)\n",
    "# # ax1.set_xlim(0.5, 0.95)\n",
    "# # ax1.set_ylim(0.5, 0.95)\n",
    "# # ax2.set_xlim(0.5, 0.95)\n",
    "# # ax2.set_ylim(0.5, 0.95)\n",
    "# ax2.axis('off')\n",
    "\n",
    "\n",
    "\n",
    "# # plt.show()\n",
    "# plt.savefig('result_plots/ppg_comparisons.png', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "ba0b13e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "\n",
    "i = 0\n",
    "plt.plot([ucla_ppg_pr_auc[i][0], ucla_ppg_pr_auc[i][0]], [ucla_ppg_roc_auc[i][1], ucla_ppg_roc_auc[i][2]], color='k', zorder=0)\n",
    "plt.plot([ucla_ppg_pr_auc[i][1], ucla_ppg_pr_auc[i][2]], [ucla_ppg_roc_auc[i][0], ucla_ppg_roc_auc[i][0]], color='k', zorder=0)\n",
    "plt.scatter(ucla_ppg_pr_auc[i][0], ucla_ppg_roc_auc[i][0], color=colors[i], s=400)\n",
    "\n",
    "plt.xticks(np.arange(ucla_ppg_pr_auc[i][1], ucla_ppg_pr_auc[i][2], 0.004), np.arange(ucla_ppg_pr_auc[i][1], ucla_ppg_pr_auc[i][2], 0.004))\n",
    "plt.tick_params(axis='both', which='major', labelsize=20)\n",
    "plt.tick_params(axis='both', which='minor', labelsize=20)\n",
    "\n",
    "plt.savefig('result_plots/zomin.png', dpi=300)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "6e5c1bf0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.762, 0.766, 0.77 , 0.774])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.arange(ucla_ppg_pr_auc[i][1], ucla_ppg_pr_auc[i][2], 0.004)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "361a35a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x400 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "\n",
    "methods = [\n",
    "'ResNet-34 (PPG)',\n",
    "'Deep Mut.',\n",
    "'SiamAF',\n",
    "'DeepBeat'\n",
    "]\n",
    "\n",
    "plt.clf()\n",
    "fig, axs = plt.subplots(1, 3, figsize=(15, 4))\n",
    "plt.setp(axs, xlim=(0.57, 0.92), ylim=(0.75, 0.95))\n",
    "\n",
    "\n",
    "\n",
    "for i in [0, 1, 3, 2]:\n",
    "    axs[0].scatter(simband_ppg_pr_auc[i][0], simband_ppg_roc_auc[i][0], marker='^', label=methods[i], color=colors[i], edgecolors=None, s=100)\n",
    "    axs[0].set_title('Simband PPG dataset', fontsize = 15)\n",
    "for i in [0, 1, 3, 2]:\n",
    "    axs[1].scatter(ucla_ppg_pr_auc[i][0], ucla_ppg_roc_auc[i][0], marker='.', color=colors[i], edgecolors=None, s=150)\n",
    "#     axs[1].axes.get_yaxis().set_visible(False)\n",
    "    axs[1].set_title('Institution B PPG dataset', fontsize = 15)\n",
    "\n",
    "    \n",
    "for i in [0, 1, 3, 2]:\n",
    "    axs[2].scatter(stanford_pr_auc[i][0], stanford_roc_auc[i][0], marker='p', label=methods[i], color=colors[i], edgecolors=None, s=100)\n",
    "#     axs[2].axes.get_yaxis().set_visible(False)\n",
    "    axs[2].set_title('Stanford dataset', fontsize = 15)\n",
    "    \n",
    "axs[1].yaxis.set_visible(False) \n",
    "axs[2].yaxis.set_visible(False) \n",
    "axs[0].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[0].tick_params(axis='both', which='minor', labelsize=15)\n",
    "axs[1].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[1].tick_params(axis='both', which='minor', labelsize=15)\n",
    "axs[2].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[2].tick_params(axis='both', which='minor', labelsize=15)\n",
    "     \n",
    "patches = []\n",
    "for i in [0, 1, 3, 2]:\n",
    "    edgecolors=None      \n",
    "    patches.append(mpatches.Patch(facecolor=colors[i], label=methods[i], edgecolor=edgecolors))\n",
    "\n",
    "#'weight':'bold'\n",
    "legend_properties = {'size':12}\n",
    "\n",
    "legend = axs[0].legend(handles=patches, title='Methods', loc='upper left', prop=legend_properties)\n",
    "axs[1].plot((ucla_ppg_pr_auc[i][0], ucla_ppg_pr_auc[i][0]), (ucla_ppg_roc_auc[i][0], ucla_ppg_roc_auc[i][0]), label='95% CI', color='k', zorder=0)\n",
    "axs[1].legend(loc='lower center', prop=legend_properties)\n",
    "\n",
    "\n",
    "\n",
    "# axs[2].set_yticks(np.arange(0.76, 0.89, 0.02), [round(x, 2) for x in np.arange(0.76, 0.89, 0.02)])\n",
    "\n",
    "# from matplotlib.ticker import FormatStrFormatter\n",
    "# axs[0].yaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "# axs[1].yaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "# axs[2].yaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "# axs[0].xaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "# axs[1].xaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "# axs[2].xaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "fig.supxlabel('AUROC', fontsize=15)\n",
    "fig.supylabel('AUPRC', fontsize=15)\n",
    "\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.setp(legend.get_title(),fontsize='12')\n",
    "plt.savefig('result_plots/ppg_comparisons_v2.png', dpi=300)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "8237908f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "\n",
    "methods = [\n",
    "'ResNet-34 (PPG)',\n",
    "'Deep Mut.',\n",
    "'SiamAF',\n",
    "]\n",
    "\n",
    "plt.clf()\n",
    "fig, axs = plt.subplots(1, 2, figsize=(10, 4))\n",
    "plt.setp(axs, xlim=(0.6, 0.8), ylim=(0.71, 0.95))\n",
    "\n",
    "\n",
    "\n",
    "for i in range(len(methods)):\n",
    "    axs[0].scatter(simband_ecg_pr_auc[i][0], simband_ecg_roc_auc[i][0], marker='^', label=methods[i], color=colors[i], edgecolors=None, s=100)\n",
    "    axs[0].set_title('Simband ECG dataset', fontsize = 15)\n",
    "for i in range(len(methods)):\n",
    "    axs[1].scatter(ucla_ecg_pr_auc[i][0], ucla_ecg_roc_auc[i][0], marker='.', label=methods[i], color=colors[i], edgecolors=None, s=150)\n",
    "    axs[1].set_title('Institution B ECG dataset', fontsize = 15)\n",
    "    \n",
    "axs[0].legend(title='Methods', loc='upper left')\n",
    "\n",
    "axs[1].yaxis.set_visible(False) \n",
    "\n",
    "axs[0].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[0].tick_params(axis='both', which='minor', labelsize=15)\n",
    "axs[1].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[1].tick_params(axis='both', which='minor', labelsize=15)\n",
    "# from matplotlib.ticker import FormatStrFormatter\n",
    "# axs[0].yaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "# axs[1].yaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "# axs[0].xaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "# axs[1].xaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "\n",
    "# axs[2].set_yticks(np.arange(0.76, 0.89, 0.02), [round(x, 2) for x in np.arange(0.76, 0.89, 0.02)])\n",
    "\n",
    "\n",
    "patches = []\n",
    "for i in range(len(methods)):\n",
    "    edgecolors=None      \n",
    "    patches.append(mpatches.Patch(facecolor=colors[i], label=methods[i], edgecolor=edgecolors))\n",
    "\n",
    "legend_properties = {'size':12}\n",
    "\n",
    "fig.supxlabel('AUROC', fontsize=15)\n",
    "fig.supylabel('AUPRC', fontsize=15)\n",
    "\n",
    "\n",
    "legend = axs[0].legend(handles=patches, title='Methods', loc='upper left', prop=legend_properties)\n",
    "plt.setp(legend.get_title(),fontsize='12')\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.savefig('result_plots/ecg_comparisons_v2.png', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "563032d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "\n",
    "# methods = [\n",
    "# 'ResNet-34 (PPG)',\n",
    "# 'Deep Mut.',\n",
    "# 'SiamAF',\n",
    "# 'DeepBeat'\n",
    "# ]\n",
    "\n",
    "# plt.clf()\n",
    "# fig, axs = plt.subplots(2, 3, figsize=(8, 4))\n",
    "# plt.setp(axs, ylim=(0.5, 1))\n",
    "\n",
    "# axs[0][0].set_ylabel('AURPC')\n",
    "# axs[1][0].set_ylabel('AUROC')\n",
    "\n",
    "\n",
    "# for i in range(len(methods)):\n",
    "#     axs[0][0].bar(i, simband_ppg_pr_auc[i][0], color=colors[i])\n",
    "#     axs[0][0].plot((i, i), (simband_ppg_pr_auc[i][1], simband_ppg_pr_auc[i][2]), c='k')\n",
    "#     axs[0][0].set_title('Simband')\n",
    "    \n",
    "#     axs[1][0].bar(i, simband_ppg_roc_auc[i][0], color=colors[i])\n",
    "#     axs[1][0].plot((i, i), (simband_ppg_roc_auc[i][1], simband_ppg_roc_auc[i][2]), c='k')\n",
    "\n",
    "# for i in range(len(methods)):\n",
    "#     axs[0][1].bar(i, ucla_ppg_pr_auc[i][0], color=colors[i])\n",
    "#     axs[0][1].plot((i, i), (ucla_ppg_pr_auc[i][1], ucla_ppg_pr_auc[i][2]), c='k')\n",
    "#     axs[0][1].set_title('Institution B')\n",
    "    \n",
    "#     axs[1][1].bar(i, ucla_ppg_roc_auc[i][0], color=colors[i])\n",
    "#     axs[1][1].plot((i, i), (ucla_ppg_roc_auc[i][1], ucla_ppg_roc_auc[i][2]), c='k')\n",
    "    \n",
    "# for i in range(len(methods)):\n",
    "#     axs[0][2].bar(i, stanford_pr_auc[i][0], color=colors[i])\n",
    "#     axs[0][2].plot((i, i), (stanford_pr_auc[i][1], stanford_pr_auc[i][2]), c='k')\n",
    "#     axs[0][2].set_title('Stanford')\n",
    "    \n",
    "#     axs[1][2].bar(i, stanford_roc_auc[i][0], color=colors[i])\n",
    "#     axs[1][2].plot((i, i), (stanford_roc_auc[i][1], stanford_roc_auc[i][2]), c='k')\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "88fcd9c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "\n",
    "# methods = [\n",
    "# 'ResNet-34 (PPG)',\n",
    "# 'Deep Mut.',\n",
    "# 'Proposed',\n",
    "# 'DeepBeat'\n",
    "# ]\n",
    "\n",
    "# plt.clf()\n",
    "# fig, axs = plt.subplots(2, 3, figsize=(8, 4))\n",
    "# plt.setp(axs, ylim=(0.5, 1))\n",
    "\n",
    "# axs[0][0].set_ylabel('AURPC')\n",
    "# axs[1][0].set_ylabel('AUROC')\n",
    "\n",
    "# for idx, rec in enumerate([simband_ppg_pr_auc, ucla_ppg_pr_auc, stanford_pr_auc]):\n",
    "#     for i in range(len(methods)-1):\n",
    "#         axs[0][idx].plot((i+1, i+2), (rec[i][0], rec[i+1][1]), color='gray', zorder=0)\n",
    "# #         axs[0][idx].set_title('Simband')\n",
    "\n",
    "#     for i in range(len(methods)):\n",
    "#         axs[0][idx].scatter(i+1, rec[i][0], color=colors[i])\n",
    "        \n",
    "# for idx, rec in enumerate([simband_ppg_roc_auc, ucla_ppg_roc_auc, stanford_roc_auc]):\n",
    "#     for i in range(len(methods)-1):\n",
    "#         axs[1][idx].plot((i+1, i+2), (rec[i][0], rec[i+1][1]), color='gray', zorder=0)\n",
    "\n",
    "#     for i in range(len(methods)):\n",
    "#         axs[1][idx].scatter(i+1, rec[i][0], color=colors[i])\n",
    "\n",
    "# # for i in range(len(methods)):\n",
    "# #     axs[0][1].bar(i, ucla_ppg_pr_auc[i][0], color=colors[i])\n",
    "# #     axs[0][1].plot((i, i), (ucla_ppg_pr_auc[i][1], ucla_ppg_pr_auc[i][2]), c='k')\n",
    "# #     axs[0][1].set_title('Institution B')\n",
    "    \n",
    "# #     axs[1][1].bar(i, ucla_ppg_roc_auc[i][0], color=colors[i])\n",
    "# #     axs[1][1].plot((i, i), (ucla_ppg_roc_auc[i][1], ucla_ppg_roc_auc[i][2]), c='k')\n",
    "    \n",
    "# # for i in range(len(methods)):\n",
    "# #     axs[0][2].bar(i, stanford_pr_auc[i][0], color=colors[i])\n",
    "# #     axs[0][2].plot((i, i), (stanford_pr_auc[i][1], stanford_pr_auc[i][2]), c='k')\n",
    "# #     axs[0][2].set_title('Stanford')\n",
    "    \n",
    "# #     axs[1][2].bar(i, stanford_roc_auc[i][0], color=colors[i])\n",
    "# #     axs[1][2].plot((i, i), (stanford_roc_auc[i][1], stanford_roc_auc[i][2]), c='k')\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "6cd591a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# def enlarge_ci_by_factor(minci, maxci, pos, factor=1):\n",
    "#     return pos-factor*(pos - minci), factor*(maxci - pos)+pos\n",
    "\n",
    "\n",
    "# plt.clf()\n",
    "# plt.figure(figsize=(10, 10))\n",
    "\n",
    "# colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "# methods = [\n",
    "# 'ResNet-34 (ECG)',\n",
    "# 'Deep Mut.',\n",
    "# 'Proposed'\n",
    "# ]\n",
    "\n",
    "# fig = plt.figure()\n",
    "# # fig.suptitle('ECG dataset comparison AUROC-AUPRC')\n",
    "# ax1 = fig.add_subplot()\n",
    "# ax2 = ax1.twinx()\n",
    "\n",
    "# ax1.scatter(simband_ecg_pr_auc[0][0], simband_ecg_roc_auc[0][0], marker='^', label='Simband Dataset', c='w', edgecolor='gray', s=100)\n",
    "# ax1.scatter(ucla_ecg_pr_auc[0][0], ucla_ecg_roc_auc[0][0], marker='o', label='Institution B Dataset', c='w', edgecolor='gray', s=100)\n",
    "\n",
    "\n",
    "# ax1.legend(title=\"Datasets\", loc='upper left')\n",
    "\n",
    "# for i in range(len(methods)):\n",
    "#     edgecolors=None\n",
    "#     ax2.scatter(simband_ecg_pr_auc[i][0], simband_ecg_roc_auc[i][0], marker='^', label=methods[i], color=colors[i], edgecolors=edgecolors, s=100)\n",
    "#     ax2.scatter(ucla_ecg_pr_auc[i][0], ucla_ecg_roc_auc[i][0], marker='o', label=methods[i], color=colors[i], edgecolors=edgecolors, s=100)\n",
    "    \n",
    "# #     ax2.text(simband_ecg_pr_auc[i][0], simband_ecg_roc_auc[i][0], f'+-aaa')\n",
    "# #     ax2.plot([ucla_ecg_pr_auc[i][0], ucla_ecg_pr_auc[i][0]], enlarge_ci_by_factor(ucla_ecg_roc_auc[i][1], ucla_ecg_roc_auc[i][2], ucla_ecg_roc_auc[i][0]), color='k', zorder=0,  solid_capstyle='round')\n",
    "# #     ax2.plot(enlarge_ci_by_factor(ucla_ecg_pr_auc[i][1], ucla_ecg_pr_auc[i][2], ucla_ecg_pr_auc[i][0]), [ucla_ecg_roc_auc[i][0], ucla_ecg_roc_auc[i][0]], color='k', zorder=0,  solid_capstyle='round')\n",
    "\n",
    "# #     ax2.plot([simband_ecg_pr_auc[i][0], simband_ecg_pr_auc[i][0]], [simband_ecg_roc_auc[i][1], simband_ecg_roc_auc[i][2]], color='k', zorder=0,  solid_capstyle='round')\n",
    "# #     ax2.plot([simband_ecg_pr_auc[i][1], simband_ecg_pr_auc[i][2]], [simband_ecg_roc_auc[i][0], simband_ecg_roc_auc[i][0]], color='k', zorder=0,  solid_capstyle='round')\n",
    "    \n",
    "# patches = []\n",
    "# for i in range(len(methods)):\n",
    "#     edgecolors=None         \n",
    "#     patches.append(mpatches.Patch(facecolor=colors[i], label=methods[i], edgecolor=edgecolors))\n",
    "\n",
    "# plt.legend(handles=patches, loc='lower left', title=\"Method\")\n",
    "\n",
    "# ax2.axis('off')\n",
    "# ax1.set_xlabel('AUPRC')\n",
    "# ax1.set_ylabel('AUROC')\n",
    "\n",
    "\n",
    "# ax1.set_xlim(0.5, 0.95)\n",
    "# ax1.set_ylim(0.5, 0.95)\n",
    "# ax2.set_xlim(0.5, 0.95)\n",
    "# ax2.set_ylim(0.5, 0.95)\n",
    "# ax2.axis('off')\n",
    "\n",
    "# ax1.set_xticks(np.arange(0.5, 0.95, 0.04), np.round(np.arange(0.5, 0.95, 0.04), 2))\n",
    "\n",
    "# # plt.show()\n",
    "# plt.savefig('result_plots/ecg_comparisons.png', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "4b545093",
   "metadata": {},
   "outputs": [],
   "source": [
    "simband_ppg_roc_auc_label01 = [  \n",
    "[0.5,0.5, 0.5],\n",
    "[0.458, 0.456, 0.459],\n",
    "[0.9,0.898, 0.901],\n",
    "\n",
    "]\n",
    "simband_ppg_pr_auc_label01 = [  \n",
    "[0.703,0.702, 0.705],\n",
    "[0.349, 0.345, 0.352,],\n",
    "[0.841, 0.838, 0.844],\n",
    "]\n",
    "ucla_ppg_roc_auc_label01 = [  \n",
    "[0.5, 0.5, 0.5],\n",
    "[0.425, 0.423, 0.426],\n",
    "[0.907, 0.905, 0.909],\n",
    "]\n",
    "\n",
    "ucla_ppg_pr_auc_label01 = [ \n",
    "[0.167, 0.164, 0.171],\n",
    "[0.11, 0.109, 0.112,],\n",
    "[0.744, 0.739, 0.749],\n",
    "]\n",
    "\n",
    "stanford_pr_auc_label01 = [  \n",
    "[0.196, 0.193, 0.2,],\n",
    "[0.161, 0.159,0.163],\n",
    "[0.565,0.56, 0.569],\n",
    "\n",
    "]\n",
    "stanford_roc_auc_label01 = [  \n",
    "[0.5,0.5, 0.5],\n",
    "[0.433,0.432, 0.434],\n",
    "[0.8,0.799, 0.801],\n",
    "]\n",
    "\n",
    "simband_ecg_roc_auc_label01 = [\n",
    "[0.509, 0.509, 0.51],\n",
    "[0.5,0.5, 0.5],\n",
    "[0.729,0.726, 0.732],\n",
    "]\n",
    "\n",
    "simband_ecg_pr_auc_label01 = [\n",
    "[0.401,0.393, 0.41],\n",
    "[0.703,0.702, 0.705],\n",
    "[0.661,0.658, 0.665],\n",
    "]\n",
    "\n",
    "ucla_ecg_roc_auc_label01 = [\n",
    "[0.491,0.49, 0.492],\n",
    "[0.5,0.5, 0.5],\n",
    "[0.899, 0.897, 0.902],\n",
    "]\n",
    "\n",
    "ucla_ecg_pr_auc_label01 = [\n",
    "[0.194, 0.19,0.198],\n",
    "[0.571,0.57, 0.572],\n",
    "[0.736,0.731, 0.742],\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "57b58d61",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x400 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "\n",
    "methods = [\n",
    "'ResNet-34 (PPG)',\n",
    "'Deep Mut.',\n",
    "'SiamAF',\n",
    "]\n",
    "\n",
    "plt.clf()\n",
    "fig, axs = plt.subplots(1, 3, figsize=(15, 4))\n",
    "plt.setp(axs, xlim=(0.0, 0.9), ylim=(0.4, 0.95))\n",
    "\n",
    "\n",
    "\n",
    "for i in range(len(methods)):\n",
    "    axs[0].scatter(simband_ppg_pr_auc_label01[i][0], simband_ppg_roc_auc_label01[i][0], marker='^', label=methods[i], color=colors[i], edgecolors=None, s=100)\n",
    "    axs[0].set_title('Simband PPG dataset',  fontsize = 15)\n",
    "for i in range(len(methods)):\n",
    "    axs[1].scatter(ucla_ppg_pr_auc_label01[i][0], ucla_ppg_roc_auc_label01[i][0], marker='.', label=methods[i], color=colors[i], edgecolors=None, s=150)\n",
    "#     axs[1].axes.get_yaxis().set_visible(False)\n",
    "    axs[1].set_title('Institution B PPG dataset',  fontsize = 15)\n",
    "\n",
    "    \n",
    "for i in range(len(methods)):\n",
    "    axs[2].scatter(stanford_pr_auc_label01[i][0], stanford_roc_auc_label01[i][0], marker='p', label=methods[i], color=colors[i], edgecolors=None, s=100)\n",
    "#     axs[2].axes.get_yaxis().set_visible(False)\n",
    "    axs[2].set_title('Stanford dataset',  fontsize = 15)\n",
    "\n",
    "axs[2].yaxis.set_visible(False) \n",
    "axs[1].yaxis.set_visible(False) \n",
    "\n",
    "axs[0].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[0].tick_params(axis='both', which='minor', labelsize=15)\n",
    "axs[1].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[1].tick_params(axis='both', which='minor', labelsize=15)\n",
    "axs[2].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[2].tick_params(axis='both', which='minor', labelsize=15)\n",
    "  \n",
    "     \n",
    "patches = []\n",
    "for i in range(len(methods)):\n",
    "    edgecolors=None      \n",
    "    patches.append(mpatches.Patch(facecolor=colors[i], label=methods[i], edgecolor=edgecolors))\n",
    "\n",
    "legend_properties = {'size':12}\n",
    "\n",
    "fig.supxlabel('AUROC', fontsize=15)\n",
    "fig.supylabel('AUPRC', fontsize=15)\n",
    "\n",
    "legend = axs[0].legend(handles=patches, title='Methods', loc='upper left', prop=legend_properties)\n",
    "plt.setp(legend.get_title(),fontsize='12')\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.savefig('result_plots/ppg_comparisons_1perc_v2.png', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "746f754c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "\n",
    "methods = [\n",
    "'ResNet-34 (ECG)',\n",
    "'Deep Mut.',\n",
    "'SiamAF',\n",
    "]\n",
    "\n",
    "plt.clf()\n",
    "fig, axs = plt.subplots(1, 2, figsize=(10, 4))\n",
    "plt.setp(axs, xlim=(0.15, 0.8), ylim=(0.4, 0.95))\n",
    "\n",
    "\n",
    "\n",
    "for i in range(len(methods)):\n",
    "    axs[0].scatter(simband_ecg_pr_auc_label01[i][0], simband_ecg_roc_auc_label01[i][0], marker='^', label=methods[i], color=colors[i], edgecolors=None, s=100)\n",
    "    axs[0].set_title('Simband ECG dataset')\n",
    "for i in range(len(methods)):\n",
    "    axs[1].scatter(ucla_ecg_pr_auc_label01[i][0], ucla_ecg_roc_auc_label01[i][0], marker='.', label=methods[i], color=colors[i], edgecolors=None, s=150)\n",
    "    axs[1].set_title('Institution B ECG dataset')\n",
    "    \n",
    "\n",
    "axs[0].legend(title='Methods', loc='upper left')\n",
    "\n",
    "axs[1].yaxis.set_visible(False) \n",
    "\n",
    "axs[0].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[0].tick_params(axis='both', which='minor', labelsize=15)\n",
    "axs[1].tick_params(axis='both', which='major', labelsize=15)\n",
    "axs[1].tick_params(axis='both', which='minor', labelsize=15)\n",
    "\n",
    "patches = []\n",
    "for i in range(len(methods)):\n",
    "    edgecolors=None      \n",
    "    patches.append(mpatches.Patch(facecolor=colors[i], label=methods[i], edgecolor=edgecolors))\n",
    "\n",
    "legend_properties = {'size':12}\n",
    "fig.supxlabel('AUROC', fontsize=15)\n",
    "fig.supylabel('AUPRC', fontsize=15)\n",
    "\n",
    "legend = axs[0].legend(handles=patches, title='Methods', loc='upper left', prop=legend_properties)\n",
    "plt.setp(legend.get_title(),fontsize='12')\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.savefig('result_plots/ecg_comparisons_1perc_v2.png', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "d6a2edaf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x1000 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize=(10, 10))\n",
    "\n",
    "colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "methods = [\n",
    "'ResNet-34 (PPG)',\n",
    "'Deep Mut.',\n",
    "'SiamAF'\n",
    "]\n",
    "\n",
    "fig = plt.figure()\n",
    "# fig.suptitle('PPG dataset 1% training labels comparison AUROC-AUPRC')\n",
    "ax1 = fig.add_subplot()\n",
    "ax2 = ax1.twinx()\n",
    "\n",
    "ax1.scatter(simband_ppg_pr_auc_label01[0][0], simband_ppg_roc_auc_label01[0][0], marker='^', label='Simband Dataset', c='w', edgecolor='gray', s=100)\n",
    "ax1.scatter(ucla_ppg_pr_auc_label01[0][0], ucla_ppg_roc_auc_label01[0][0], marker='o', label='Institution B Dataset', c='w', edgecolor='gray', s=100)\n",
    "ax1.scatter(stanford_pr_auc_label01[0][0], stanford_roc_auc_label01[0][0], marker='*', label='Stanford Dataset', c='w', edgecolor='gray', s=100)\n",
    "\n",
    "ax1.legend(title=\"Datasets\", loc='upper left')\n",
    "\n",
    "for i in range(len(methods)):\n",
    "    edgecolors=None\n",
    "    ax2.scatter(simband_ppg_pr_auc_label01[i][0], simband_ppg_roc_auc_label01[i][0], marker='^', label=methods[i], color=colors[i], edgecolors=edgecolors, s=100)\n",
    "    ax2.scatter(ucla_ppg_pr_auc_label01[i][0], ucla_ppg_roc_auc_label01[i][0], marker='o', label=methods[i], color=colors[i], edgecolors=edgecolors, s=100)\n",
    "    ax2.scatter(stanford_pr_auc_label01[i][0], stanford_roc_auc_label01[i][0], marker='*', label=methods[i], color=colors[i], edgecolors=edgecolors, s=100)\n",
    "    \n",
    "#     ax2.plot([stanford_pr_auc_label01[i][0], stanford_pr_auc_label01[i][0]], [stanford_roc_auc_label01[i][1], stanford_roc_auc_label01[i][2]], color='gray')\n",
    "#     ax2.plot([stanford_pr_auc_label01[i][1], stanford_pr_auc_label01[i][2]], [stanford_roc_auc_label01[i][0], stanford_roc_auc_label01[i][0]], color='gray')\n",
    "    \n",
    "#     ax2.plot([ucla_ppg_pr_auc_label01[i][0], ucla_ppg_pr_auc_label01[i][0]], [ucla_ppg_roc_auc_label01[i][1], ucla_ppg_roc_auc_label01[i][2]], color='gray')\n",
    "#     ax2.plot([ucla_ppg_pr_auc_label01[i][1], ucla_ppg_pr_auc_label01[i][2]], [ucla_ppg_roc_auc_label01[i][0], ucla_ppg_roc_auc_label01[i][0]], color='gray')\n",
    "    \n",
    "#     ax2.plot([simband_ppg_pr_auc_label01[i][0], simband_ppg_pr_auc_label01[i][0]], [simband_ppg_roc_auc_label01[i][1], simband_ppg_roc_auc_label01[i][2]], color='gray')\n",
    "#     ax2.plot([simband_ppg_pr_auc_label01[i][1], simband_ppg_pr_auc_label01[i][2]], [simband_ppg_roc_auc_label01[i][0], simband_ppg_roc_auc_label01[i][0]], color='gray')\n",
    "    \n",
    "patches = []\n",
    "for i in range(len(methods)):\n",
    "    edgecolors=None        \n",
    "    patches.append(mpatches.Patch(facecolor=colors[i], label=methods[i], edgecolor=edgecolors))\n",
    "\n",
    "plt.legend(handles=patches, loc='center left', title=\"Method\")\n",
    "\n",
    "ax2.axis('off')\n",
    "ax1.set_xlabel('AUPRC')\n",
    "ax1.set_ylabel('AUROC')\n",
    "\n",
    "ax1.set_xlim(0.0, 0.9)\n",
    "ax1.set_ylim(0.4, 0.95)\n",
    "ax2.set_xlim(0.0, 0.9)\n",
    "ax2.set_ylim(0.4, 0.95)\n",
    "\n",
    "\n",
    "# plt.show()\n",
    "plt.savefig('result_plots/ppg_comparisons_1perc.png', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "cc29f993",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x1000 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize=(10, 10))\n",
    "\n",
    "colors = ['#e8dab2', '#4f6d7a', '#dd6e42', '#c0d6df']\n",
    "methods = [\n",
    "'ResNet-34 (ECG)',\n",
    "'Deep Mut.',\n",
    "'SiamAF'\n",
    "]\n",
    "\n",
    "fig = plt.figure()\n",
    "# fig.suptitle('ECG dataset comparison 1% training labels AUROC-AUPRC')\n",
    "ax1 = fig.add_subplot()\n",
    "ax2 = ax1.twinx()\n",
    "\n",
    "ax1.scatter(simband_ecg_pr_auc_label01[0][0], simband_ecg_roc_auc_label01[0][0], marker='^', label='Simband Dataset', c='w', edgecolor='gray', s=100)\n",
    "ax1.scatter(ucla_ecg_pr_auc_label01[0][0], ucla_ecg_roc_auc_label01[0][0], marker='o', label='Institution B Dataset', c='w', edgecolor='gray', s=100)\n",
    "# ax1.scatter(stanford_pr_auc, stanford_roc_auc, marker='*', label='Stanford Dataset', c='w', edgecolor='gray')\n",
    "\n",
    "ax1.legend(title=\"Datasets\", loc='upper left')\n",
    "\n",
    "for i in range(len(methods)):\n",
    "    edgecolors=None\n",
    "    ax2.scatter(simband_ecg_pr_auc_label01[i][0], simband_ecg_roc_auc_label01[i][0], marker='^', label=methods[i], color=colors[i], edgecolors=edgecolors, s=100)\n",
    "    ax2.scatter(ucla_ecg_pr_auc_label01[i][0], ucla_ecg_roc_auc_label01[i][0], marker='o', label=methods[i], color=colors[i], edgecolors=edgecolors, s=100)\n",
    "#     ax2.scatter(stanford_pr_auc[i], stanford_roc_auc[i], marker='*', label=methods[i], color=colors[i], edgecolors=edgecolors, s=100)\n",
    "    \n",
    "#     ax2.plot([ucla_ecg_pr_auc_label01[i][0], ucla_ecg_pr_auc_label01[i][0]], [ucla_ecg_roc_auc_label01[i][1], ucla_ecg_roc_auc_label01[i][2]], color='gray')\n",
    "#     ax2.plot([ucla_ecg_pr_auc_label01[i][1], ucla_ecg_pr_auc_label01[i][2]], [ucla_ecg_roc_auc_label01[i][0], ucla_ecg_roc_auc_label01[i][0]], color='gray')\n",
    "    \n",
    "#     ax2.plot([simband_ecg_pr_auc_label01[i][0], simband_ecg_pr_auc_label01[i][0]], [simband_ecg_roc_auc_label01[i][1], simband_ecg_roc_auc_label01[i][2]], color='gray')\n",
    "#     ax2.plot([simband_ecg_pr_auc_label01[i][1], simband_ecg_pr_auc_label01[i][2]], [simband_ecg_roc_auc_label01[i][0], simband_ecg_roc_auc_label01[i][0]], color='gray')\n",
    "    \n",
    "    \n",
    "patches = []\n",
    "for i in range(len(methods)):\n",
    "    edgecolors=None          \n",
    "    patches.append(mpatches.Patch(facecolor=colors[i], label=methods[i], edgecolor=edgecolors))\n",
    "\n",
    "plt.legend(handles=patches, loc='center left', title=\"Method\")\n",
    "\n",
    "ax2.axis('off')\n",
    "ax1.set_xlabel('AUPRC')\n",
    "ax1.set_ylabel('AUROC')\n",
    "\n",
    "ax1.set_xlim(0.0, 0.9)\n",
    "ax1.set_ylim(0.4, 0.95)\n",
    "ax2.set_xlim(0.0, 0.9)\n",
    "ax2.set_ylim(0.4, 0.95)\n",
    "\n",
    "\n",
    "# plt.show()\n",
    "plt.savefig('result_plots/ecg_comparisons_1perc.png', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "299fa038",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base_pytorch",
   "language": "python",
   "name": "base_pytorch"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}