{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "import pickle \n",
    "import matplotlib.pyplot as plt \n",
    "import seaborn as sns\n",
    "import random\n",
    "import os\n",
    "np.set_printoptions(suppress=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# collect[-1]['Observation'] = sstgcnn_['obs'][key]\n",
    "# collect[-1]['GroundTruth'] = sstgcnn_['tarj'][key]\n",
    "# collect[-1]['S-GAN'] = sgan_['pred'][key]\n",
    "# collect[-1]['S-STGCNN'] = sstgcnn_['pred'][key]\n",
    "# collect[-1]['S-Implicit'] = simplicit_['pred'][key]\n",
    "# collect[-1]['Trajectron++'] = trajpp_['pred'][key]\n",
    "# collect[-1]['ExpertTraj'] = expertraj_['pred'][key]\n",
    "dset = 'eth' # change dest to 'eth', 'zara1', 'hotel' \n",
    "with open('./aligned/vis_'+dset+'.pkl', 'rb') as f:\n",
    "    data = pickle.load(f)\n",
    "output_dir= './' + dset + 'Vis'\n",
    "if not os.path.exists(output_dir):\n",
    "  # Create a new directory because it does not exist \n",
    "  os.makedirs(output_dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def obtain_limits(model_data, normal_xy_ratio=True):\n",
    "    x_min = y_min = np.inf\n",
    "    x_max = y_max = -np.inf\n",
    "    for trajs in model_data.values():\n",
    "        sub_x_min = trajs[..., 0, :].min()\n",
    "        sub_x_max = trajs[..., 0, :].max()\n",
    "        sub_y_min = trajs[..., 1, :].min()\n",
    "        sub_y_max = trajs[..., 1, :].max()\n",
    "        x_min = min(x_min, sub_x_min)\n",
    "        x_max = max(x_max, sub_x_max)\n",
    "        y_min = min(y_min, sub_y_min)\n",
    "        y_max = max(y_max, sub_y_max)\n",
    "    x_len = x_max - x_min\n",
    "    y_len = y_max - y_min\n",
    "    \n",
    "    if normal_xy_ratio:\n",
    "        if x_len/y_len > 3.5/4.5:  # the x/y ratio of plot region in the figure is about 3.5/4.5. \n",
    "            # y_len is too small\n",
    "            y_center = (y_max + y_min) / 2\n",
    "            y_len = x_len / 3.5 * 4.5\n",
    "            y_min = y_center - y_len / 2\n",
    "            y_max = y_center + y_len / 2\n",
    "        else:\n",
    "            # x_len is too small\n",
    "            x_center = (x_max + x_min) / 2\n",
    "            x_len = y_len / 4.5 * 3.5\n",
    "            x_min = x_center - x_len / 2\n",
    "            x_max = x_center + x_len / 2\n",
    "        \n",
    "    return [x_min, x_max], [y_min, y_max]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(100, 2, 12)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_2577516/3556512815.py:51: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-Implicit'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:51: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-Implicit'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:60: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-GAN'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:60: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-GAN'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:68: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-STGCNN'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:68: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-STGCNN'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:76: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['Trajectron++'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:76: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['Trajectron++'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:84: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['ExpertTraj'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:84: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['ExpertTraj'][:, 0].reshape(-1),\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x500 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(100, 2, 12)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_2577516/3556512815.py:51: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-Implicit'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:51: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-Implicit'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:60: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-GAN'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:60: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-GAN'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:68: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-STGCNN'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:68: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['S-STGCNN'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:76: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['Trajectron++'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:76: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['Trajectron++'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:84: UserWarning: \n",
      "\n",
      "`shade_lowest` has been replaced by `thresh`; setting `thresh=0.05.\n",
      "This will become an error in seaborn v0.13.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['ExpertTraj'][:, 0].reshape(-1),\n",
      "/tmp/ipykernel_2577516/3556512815.py:84: FutureWarning: \n",
      "\n",
      "`shade` is now deprecated in favor of `fill`; setting `fill=True`.\n",
      "This will become an error in seaborn v0.14.0; please update your code.\n",
      "\n",
      "  sns.kdeplot(x = data[key]['ExpertTraj'][:, 0].reshape(-1),\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x500 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cmap = ['g', 'y', 'b', 'k', 'r']\n",
    "\n",
    "indices = {}\n",
    "#Paper indices\n",
    "indices['eth'] = [0,95]\n",
    "indices['hotel'] = [419,701]\n",
    "indices['zara1'] = [2031,315,1290]\n",
    "# indices['eth'] = [\n",
    "#     0, 14, 21, 24, 33, 44, 45, 50, 58, 60, 79, 92, 93, 95, 105, 106, 112, 113,\n",
    "#     121, 113, 121, 122, 127, 140, 153, 159, 164, 166, 167, 170, 172, 177\n",
    "# ]\n",
    "# indices['hotel'] = [\n",
    "#     30, 87, 173, 189, 196, 198, 221, 329, 364, 419, 501, 510, 528, 560, 572,\n",
    "#     620, 701, 719, 733, 818, 850, 861, 870, 965, 1007, 1008, 1013, 1031, 1046,\n",
    "#     1052\n",
    "# ]\n",
    "# indices['zara1'] = [\n",
    "#     25, 38, 130, 240, 315, 347, 461, 648, 679, 687, 693, 720, 781, 858, 947,\n",
    "#     1004, 1053, 1091, 1130, 1178, 1290, 1320, 1359, 1691, 1702, 1884, 1927,\n",
    "#     2031, 2186, 2220\n",
    "# ]\n",
    "\n",
    "for key in indices[dset]:\n",
    "    fig, ax = plt.subplots(1, 5, figsize=(20, 5))  #,sharex=True, sharey=True)\n",
    "\n",
    "    # Deyao: i notice that for the selected examples, eth and zara1 are better visualized when we switch x and y axis\n",
    "    if dset in ['eth', 'zara1']:\n",
    "        for model in data[key].keys():\n",
    "            data[key][model] = np.flip(data[key][model], axis=-2)\n",
    "            \n",
    "    ### 이전 경로와 이후 GT 경로 그려줌\n",
    "    # for i in range(5): \n",
    "    #     if i == 0:\n",
    "    #         ax[i].plot(data[key]['Observation'][0],\n",
    "    #                    data[key]['Observation'][1],\n",
    "    #                    '--o',\n",
    "    #                    label='Observation')\n",
    "    #         ax[i].plot(data[key]['GroundTruth'][0],\n",
    "    #                    data[key]['GroundTruth'][1],\n",
    "    #                    '--x',\n",
    "    #                    label='GroundTruth')\n",
    "    #     else:\n",
    "    #         ax[i].plot(data[key]['Observation'][0],\n",
    "    #                    data[key]['Observation'][1], '--o')\n",
    "    #         ax[i].plot(data[key]['GroundTruth'][0],\n",
    "    #                    data[key]['GroundTruth'][1], '--x')\n",
    "\n",
    "    x_limit, y_limit = obtain_limits(data[key])\n",
    "    \n",
    "    print(data[key]['S-Implicit'].shape)\n",
    "    \n",
    "    sns.kdeplot(x = data[key]['S-Implicit'][:, 0].reshape(-1),\n",
    "                y = data[key]['S-Implicit'][:, 1].reshape(-1),\n",
    "                shade=True,\n",
    "                shade_lowest=False,\n",
    "                color=cmap[0],\n",
    "                alpha=0.8,\n",
    "                label='S-Implicit',\n",
    "                ax=ax[0])\n",
    "\n",
    "    sns.kdeplot(x = data[key]['S-GAN'][:, 0].reshape(-1),\n",
    "                y = data[key]['S-GAN'][:, 1].reshape(-1),\n",
    "                shade=True,\n",
    "                shade_lowest=False,\n",
    "                color=cmap[1],\n",
    "                alpha=0.8,\n",
    "                label='S-GAN',\n",
    "                ax=ax[1])\n",
    "    sns.kdeplot(x = data[key]['S-STGCNN'][:, 0].reshape(-1),\n",
    "                y = data[key]['S-STGCNN'][:, 1].reshape(-1),\n",
    "                shade=True,\n",
    "                shade_lowest=False,\n",
    "                color=cmap[2],\n",
    "                alpha=0.8,\n",
    "                label='S-STGCNN',\n",
    "                ax=ax[2])\n",
    "    sns.kdeplot(x = data[key]['Trajectron++'][:, 0].reshape(-1),\n",
    "                y = data[key]['Trajectron++'][:, 1].reshape(-1),\n",
    "                shade=True,\n",
    "                shade_lowest=False,\n",
    "                color=cmap[3],\n",
    "                alpha=0.8,\n",
    "                label='Trajectron++',\n",
    "                ax=ax[3])\n",
    "    sns.kdeplot(x = data[key]['ExpertTraj'][:, 0].reshape(-1),\n",
    "                y = data[key]['ExpertTraj'][:, 1].reshape(-1),\n",
    "                shade=True,\n",
    "                shade_lowest=False,\n",
    "                color=cmap[4],\n",
    "                alpha=0.8,\n",
    "                label='ExpertTraj',\n",
    "                ax=ax[4])\n",
    "\n",
    "    for i in range(5):\n",
    "        ax[i].xaxis.set_tick_params(labelsize=14)\n",
    "        ax[i].yaxis.set_tick_params(labelsize=14)\n",
    "        ax[i].set_xlim(x_limit)  # deyao: Here the limit changed\n",
    "        ax[i].set_ylim(y_limit)\n",
    "\n",
    "    lines_labels = [ax.get_legend_handles_labels() for ax in fig.axes]\n",
    "    lines, labels = [sum(lol, []) for lol in zip(*lines_labels)]\n",
    "    fig.legend(lines,\n",
    "               labels,\n",
    "               loc=\"upper center\",\n",
    "               ncol=7,\n",
    "               fontsize=15,\n",
    "               bbox_to_anchor=(0.5, 1.1))\n",
    "    plt.tight_layout()\n",
    "    plt.savefig(output_dir+ '/case_' + dset + '_' + str(key) + '.png',\n",
    "            dpi=300,\n",
    "            bbox_inches='tight')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"
  },
  "kernelspec": {
   "display_name": "scit",
   "language": "python",
   "name": "scit"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}