Training_Synthetic_Dataset.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "from typing import Tuple\n",
    "from torch.utils.data import DataLoader\n",
    "import preprocess_dataset\n",
    "import field_geometry\n",
    "import losses\n",
    "from main_network import DecompositionNetwork\n",
    "import network\n",
    "import train_eval\n",
    "import visualize_primitive\n",
    "import importlib\n",
    "from typing import Dict\n",
    "import random\n",
    "import numpy\n",
    "\n",
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "random.seed(42069)\n",
    "torch.manual_seed(42069)\n",
    "numpy.random.seed(42069)\n",
    "\n",
    "rect_amount = 1\n",
    "circ_amount = 0\n",
    "image_size = torch.tensor([64, 64])\n",
    "data = preprocess_dataset.Rectangle_Dataset(\n",
    "    rect_amount=rect_amount,  \n",
    "    circ_amount=circ_amount,\n",
    "    img_size=image_size,\n",
    "    mode=\"hardcoded\"\n",
    "    )\n",
    "train_size = int(0.8 * len(data))\n",
    "test_size = len(data) - train_size\n",
    "train_dataset, test_dataset = torch.utils.data.random_split(\n",
    "    data, [train_size, test_size]\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Hyperparameters:\n",
    "LEARNING_RATE = 1e-3\n",
    "DEVICE = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
    "BATCH_SIZE = 32\n",
    "NUM_EPOCHS = 20\n",
    "LOAD_MODEL = False\n",
    "NUM_WORKERS = os.cpu_count()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dataloader = DataLoader(\n",
    "    train_dataset,\n",
    "    batch_size=BATCH_SIZE,\n",
    "    shuffle=True,\n",
    "    num_workers=NUM_WORKERS,\n",
    "    drop_last=True,\n",
    ")\n",
    "test_dataloader = DataLoader(\n",
    "    test_dataset,\n",
    "    batch_size=BATCH_SIZE,\n",
    "    shuffle=True,\n",
    "    num_workers=NUM_WORKERS,\n",
    "    drop_last=True,\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fbe715abc40>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from field_geometry import compute_circle_distance_field, compute_rotated_rectangle_distance_field\n",
    "\n",
    "\n",
    "r= torch.tensor([\n",
    "    [\n",
    "        [0.25, 0.75],\n",
    "        [0.25, 0.75]\n",
    "    ]\n",
    "])\n",
    "\n",
    "c=torch.tensor([\n",
    "    [\n",
    "        [0.5, 0.5, 0.3],\n",
    "    ]\n",
    "])\n",
    "\n",
    "r_field_out = compute_rotated_rectangle_distance_field(r, torch.tensor([0.0]), image_size).squeeze()\n",
    "r_field_in = compute_rotated_rectangle_distance_field(r, torch.tensor([0.0]), image_size, True).squeeze()\n",
    "\n",
    "c_field_out = compute_circle_distance_field(c, image_size).squeeze()\n",
    "c_field_in = compute_circle_distance_field(c, image_size, True).squeeze()\n",
    "\n",
    "r_field = (r_field_out + r_field_in).unsqueeze(0)\n",
    "c_field = (c_field_out + c_field_in).unsqueeze(0)\n",
    "\n",
    "field_all = torch.cat((r_field, c_field), dim=0)\n",
    "field = field_all.min(dim=0).values\n",
    "\n",
    "_, ax = plt.subplots(1, 1, figsize=(10, 10))\n",
    "ax.imshow(c_field.squeeze(), cmap=\"gray\")\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([1, 2, 2])\n",
      "tensor(16.)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from field_geometry import compute_circle_distance_field, compute_rotated_rectangle_distance_field\n",
    "\n",
    "\n",
    "_, ax = plt.subplots(1, 2, figsize=(10, 10))\n",
    "\n",
    "r = torch.tensor([\n",
    "    [\n",
    "        [0.1, 0.4],\n",
    "        [0.1, 0.4]\n",
    "    ]\n",
    "])\n",
    "\n",
    "for mask, field in train_dataloader:\n",
    "    print(r.shape)\n",
    "    r_field = (compute_rotated_rectangle_distance_field(r, torch.tensor([0.0]) ,image_size)) == 0\n",
    "    # r_field = r_field * -1\n",
    "    ax[0].imshow(r_field.squeeze(), cmap=\"gray\")\n",
    "    ax[1].imshow(field[0], cmap=\"gray\")\n",
    "    print(field[0, 32, 32])\n",
    "    visualize_primitive.plot_rotated_rectangles(ax[1], r, torch.tensor([0]), image_size)\n",
    "    break\n",
    "\n",
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Rectangle': 4, 'Circle': 8}\n"
     ]
    }
   ],
   "source": [
    "torch.cuda.empty_cache()\n",
    "# triangles and squares are not defined yet\n",
    "\n",
    "model = network.Model(\n",
    "    img_size=image_size,\n",
    "    conv_channels=(1, 12, 6),\n",
    "    dense_layers=(2048,),\n",
    "    num_recs=4,\n",
    "    num_circs=8\n",
    ")\n",
    "optim = torch.optim.Adam(model.parameters(), lr=LEARNING_RATE)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "KeyboardInterrupt\n",
      "\n"
     ]
    }
   ],
   "source": [
    "(cov_loss, cons_rect_train_loss, cons_circ_train_loss) = train_eval.train_model(\n",
    "    NUM_EPOCHS, \n",
    "    train_dataloader, \n",
    "    train_eval.extract_synthetic_data,\n",
    "     model, \n",
    "     NUM_EPOCHS//2, \n",
    "     BATCH_SIZE, \n",
    "     optim, \n",
    "     DEVICE\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x2000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "importlib.reload(visualize_primitive)\n",
    "importlib.reload(field_geometry)\n",
    "\n",
    "for mask, dist_field in test_dataloader:\n",
    "\n",
    "    model.cpu().eval()\n",
    "    with torch.no_grad():\n",
    "        r, rp, r_rot, c, cp = train_eval.unpack_model_output(\n",
    "            model.num_dict, \n",
    "            model(mask.float().reshape(-1, 1, *image_size))\n",
    "            )\n",
    "        z_rec = (rp > 0.5)\n",
    "        z_circ = (cp > 0.5)\n",
    "        torch.set_printoptions(precision=2, sci_mode=False)\n",
    "\n",
    "    # plot\n",
    "    _, (ax0, ax1, ax2, ax3) = plt.subplots(1, 4, figsize=(20,20))\n",
    "    ax0.imshow(mask[0], cmap=\"gray\")\n",
    "    ax1.imshow(mask[0], cmap=\"gray\")\n",
    "    ax2.imshow(mask[0], cmap=\"gray\")\n",
    "    ax3.imshow(mask[0], cmap=\"gray\")\n",
    "\n",
    "    visualize_primitive.plot_rotated_rectangles(ax1, r[0, z_rec[0]], r_rot[0, z_rec[0]], mask[0].size())\n",
    "    visualize_primitive.plot_rotated_rectangles(ax2, r[0, ~z_rec[0]], r_rot[0, ~z_rec[0]],  mask[0].size())\n",
    "    visualize_primitive.plot_rotated_rectangles(ax3, r[0], r_rot[0], mask[0].size())\n",
    "    visualize_primitive.plot_circles(ax1, c[0, z_circ[0]], mask[0].size())\n",
    "    visualize_primitive.plot_circles(ax2, c[0, ~z_circ[0]], mask[0].size())\n",
    "    visualize_primitive.plot_circles(ax3, c[0], mask[0].size())\n",
    "    ax0.set(title=\"GT Mask\")\n",
    "    ax1.set(title=\"Selected\")\n",
    "    ax2.set(title=\"Unselected\")\n",
    "    ax3.set(title=\"Select all\")\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'dist_threshold' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb Cell 9\u001b[0m in \u001b[0;36m<cell line: 7>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     <a href='vscode-notebook-cell://ssh-remote%2Bpool13/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb#X11sdnNjb2RlLXJlbW90ZQ%3D%3D?line=10'>11</a>\u001b[0m     \u001b[39mwith\u001b[39;00m torch\u001b[39m.\u001b[39mno_grad():\n\u001b[1;32m     <a href='vscode-notebook-cell://ssh-remote%2Bpool13/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb#X11sdnNjb2RlLXJlbW90ZQ%3D%3D?line=11'>12</a>\u001b[0m         r, rp, r_rot, c, cp \u001b[39m=\u001b[39m train_eval\u001b[39m.\u001b[39munpack_model_output(\n\u001b[1;32m     <a href='vscode-notebook-cell://ssh-remote%2Bpool13/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb#X11sdnNjb2RlLXJlbW90ZQ%3D%3D?line=12'>13</a>\u001b[0m             model\u001b[39m.\u001b[39mnum_dict, \n\u001b[1;32m     <a href='vscode-notebook-cell://ssh-remote%2Bpool13/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb#X11sdnNjb2RlLXJlbW90ZQ%3D%3D?line=13'>14</a>\u001b[0m             model(mask\u001b[39m.\u001b[39mfloat()\u001b[39m.\u001b[39mreshape(\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m, \u001b[39m1\u001b[39m, \u001b[39m*\u001b[39mimage_size))\n\u001b[1;32m     <a href='vscode-notebook-cell://ssh-remote%2Bpool13/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb#X11sdnNjb2RlLXJlbW90ZQ%3D%3D?line=14'>15</a>\u001b[0m             )\n\u001b[0;32m---> <a href='vscode-notebook-cell://ssh-remote%2Bpool13/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb#X11sdnNjb2RlLXJlbW90ZQ%3D%3D?line=17'>18</a>\u001b[0m     IoU_score \u001b[39m=\u001b[39m train_eval\u001b[39m.\u001b[39;49mIoU(r, r_rot, rp, c, cp, \u001b[39m0.5\u001b[39;49m, mask, device\u001b[39m=\u001b[39;49m\u001b[39m\"\u001b[39;49m\u001b[39mcpu\u001b[39;49m\u001b[39m\"\u001b[39;49m, mode\u001b[39m=\u001b[39;49m\u001b[39m\"\u001b[39;49m\u001b[39mbatch\u001b[39;49m\u001b[39m\"\u001b[39;49m)\n\u001b[1;32m     <a href='vscode-notebook-cell://ssh-remote%2Bpool13/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb#X11sdnNjb2RlLXJlbW90ZQ%3D%3D?line=18'>19</a>\u001b[0m     IoU_list\u001b[39m.\u001b[39mappend(IoU_score)\n\u001b[1;32m     <a href='vscode-notebook-cell://ssh-remote%2Bpool13/home/stud/schwemme/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/Training_Synthetic_Dataset.ipynb#X11sdnNjb2RlLXJlbW90ZQ%3D%3D?line=20'>21</a>\u001b[0m \u001b[39mprint\u001b[39m(\u001b[39msum\u001b[39m(IoU_list) \u001b[39m/\u001b[39m \u001b[39mlen\u001b[39m(IoU_list))\n",
      "File \u001b[0;32m~/Documents/GitHub/Deep-Learning-Network-2D-Natural-Image-Decomposition/train_eval.py:200\u001b[0m, in \u001b[0;36mIoU\u001b[0;34m(r, r_rot, rp, c, cp, prob_threshold, mask, device, mode)\u001b[0m\n\u001b[1;32m    198\u001b[0m \u001b[39m# apply the probability mask to the unselected primitives\u001b[39;00m\n\u001b[1;32m    199\u001b[0m d_rect \u001b[39m=\u001b[39m d_rect\u001b[39m.\u001b[39mmasked_fill(rp\u001b[39m.\u001b[39mreshape(b, p_rect, \u001b[39m1\u001b[39m, \u001b[39m1\u001b[39m) \u001b[39m<\u001b[39m prob_threshold, \u001b[39mFalse\u001b[39;00m)\n\u001b[0;32m--> 200\u001b[0m \u001b[39mprint\u001b[39m(dist_threshold)\n\u001b[1;32m    202\u001b[0m d_rect_union \u001b[39m=\u001b[39m d_rect\u001b[39m.\u001b[39many(dim\u001b[39m=\u001b[39m\u001b[39m1\u001b[39m)\n\u001b[1;32m    203\u001b[0m _, ax \u001b[39m=\u001b[39m plt\u001b[39m.\u001b[39msubplots(\u001b[39m1\u001b[39m, \u001b[39m2\u001b[39m, figsize\u001b[39m=\u001b[39m(\u001b[39m15\u001b[39m, \u001b[39m15\u001b[39m))\n",
      "\u001b[0;31mNameError\u001b[0m: name 'dist_threshold' is not defined"
     ]
    }
   ],
   "source": [
    "import importlib\n",
    "\n",
    "importlib.reload(train_eval)\n",
    "\n",
    "\n",
    "IoU_list = []\n",
    "for mask, dist_field in test_dataloader:\n",
    "    mask = mask.unsqueeze(1)\n",
    "\n",
    "    model.cpu().eval()\n",
    "    with torch.no_grad():\n",
    "        r, rp, r_rot, c, cp = train_eval.unpack_model_output(\n",
    "            model.num_dict, \n",
    "            model(mask.float().reshape(-1, 1, *image_size))\n",
    "            )\n",
    "\n",
    "\n",
    "    IoU_score = train_eval.IoU(r, r_rot, rp, c, cp, 0.5, mask, device=\"cpu\", mode=\"batch\")\n",
    "    IoU_list.append(IoU_score)\n",
    "\n",
    "print(sum(IoU_list) / len(IoU_list))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#(cov_loss, cons_rect_train_loss, cons_circ_train_loss)\n",
    "_, ax = plt.subplots(1, 3, figsize=(15, 5))\n",
    "cov_loss_stacked = torch.tensor([cov_loss]).flatten()\n",
    "cons_loss_rect_stacked = torch.tensor([cons_rect_train_loss]).flatten()\n",
    "cons_loss_circ_stacked = torch.tensor([cons_circ_train_loss]).flatten()\n",
    "\n",
    "\n",
    "# ax[0].set_ylim([0, cov_loss_stacked.max()])\n",
    "ax[0].set_ylim([0, cov_loss_stacked.max()])\n",
    "ax[1].set_ylim([0, cons_loss_rect_stacked.max()])\n",
    "ax[2].set_ylim([0, cons_loss_circ_stacked.max()])\n",
    "\n",
    "# ax[3].set_ylim([0, intersect_loss_stacked.max()])\n",
    "ax[0].set_title(\"Union Coverage Loss\")\n",
    "ax[0].plot(cov_loss_stacked)\n",
    "ax[1].set_title(\"Rectangle Consistency Loss\")\n",
    "ax[1].plot(cons_loss_rect_stacked)\n",
    "ax[1].set_title(\"Circle Consistency Loss\")\n",
    "ax[1].plot(cons_loss_circ_stacked)\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mean_losses = []\n",
    "for union_cov, rect_cons, circ_cons in zip(cov_loss, cons_rect_train_loss, cons_circ_train_loss):\n",
    "    mean_losses.append(\n",
    "        (\n",
    "            sum(union_cov) / len(union_cov) \n",
    "            + sum(rect_cons) / len(rect_cons) \n",
    "            + sum(circ_cons) / len(circ_cons)\n",
    "        )\n",
    "    )\n",
    "mean_losses = torch.stack(mean_losses).detach().cpu()\n",
    "fig, ax = plt.subplots(1, 1, figsize=(6, 6))\n",
    "ax.set_title(\"All losses combined\")\n",
    "ax.plot(mean_losses)\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.12 ('cv')",
   "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": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "8223595c013b6b73f6cb6a05bbf47ca9e18582c2a390050935c3934650c768f8"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}