DenoisingAutoEncoder.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "initial_id",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-10-22T20:02:03.773388787Z",
     "start_time": "2023-10-22T20:02:02.850688019Z"
    }
   },
   "outputs": [],
   "source": [
    "import cv2\n",
    "from imagecorruptions import corrupt, get_corruption_names\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "import glob\n",
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "from torch.utils.tensorboard import SummaryWriter\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a392730c91745d41",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-10-21T18:31:32.511816890Z",
     "start_time": "2023-10-21T18:31:32.487517819Z"
    }
   },
   "outputs": [],
   "source": [
    "DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
    "writer = SummaryWriter(f\"runs/DAE\", flush_secs=1)\n",
    "IMG_FOLDER= \"JPEGImages\"\n",
    "image_list = glob.glob(f\"{IMG_FOLDER}/*.jpg\")\n",
    "train_image_list = image_list[:int(len(image_list)*0.7)]\n",
    "val_image_list = image_list[int(len(image_list)*0.7):]\n",
    "IMG_SIZE = (64, 64)\n",
    "EPOCHS = 500\n",
    "BATCH_SIZE = 512"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c85c1fb5d4277ed4",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-10-21T18:31:32.910475123Z",
     "start_time": "2023-10-21T18:31:32.909327213Z"
    }
   },
   "outputs": [],
   "source": [
    "def display_img_tensors(img, cor_img, pred):\n",
    "    fig, axes = plt.subplots(1, 3, figsize=(15, 5))\n",
    "    img = img.detach().cpu().permute(1,2,0).numpy()\n",
    "    cor_img = cor_img.detach().cpu().permute(1,2,0).numpy()\n",
    "    pred = pred.detach().cpu().permute(1,2,0).numpy()\n",
    "    \n",
    "    axes[0].imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))\n",
    "    axes[0].set_title(\"Image\")\n",
    "\n",
    "    axes[1].imshow(cv2.cvtColor(cor_img, cv2.COLOR_BGR2RGB))\n",
    "    axes[1].set_title(\"Corrupted Image\")\n",
    "\n",
    "    axes[2].imshow(cv2.cvtColor(pred, cv2.COLOR_BGR2RGB))\n",
    "    axes[2].set_title(\"Predicted Image\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "fa8c0b27174ac869",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-10-21T18:31:33.269029703Z",
     "start_time": "2023-10-21T18:31:33.266056059Z"
    }
   },
   "outputs": [],
   "source": [
    "class ImageDataset(Dataset):\n",
    "    def __init__(self, image_list, img_size=IMG_SIZE):\n",
    "        super().__init__()\n",
    "        ### where you define your data source\n",
    "        self.image_list = image_list\n",
    "        self.img_size = img_size\n",
    "        self.corruption_types = get_corruption_names()\n",
    "        self.corruption_types.remove(\"glass_blur\")\n",
    "        \n",
    "    def __getitem__(self, index):\n",
    "        image = cv2.imread(self.image_list[index])\n",
    "        image = cv2.resize(image, self.img_size)\n",
    "        \n",
    "        c_name = random.sample(self.corruption_types,1)[0]\n",
    "        corrupt_image = corrupt(image, corruption_name=c_name, severity=1)\n",
    "        image = image / 255\n",
    "        corrupt_image = corrupt_image / 255\n",
    "        image = torch.from_numpy(image).permute(2, 0, 1)\n",
    "        corrupt_image = torch.from_numpy(corrupt_image).permute(2,0,1)\n",
    "        \n",
    "        return corrupt_image.to(torch.float32), image.to(torch.float32)\n",
    "    \n",
    "    def __len__(self):\n",
    "        return len(self.image_list)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "712aafb586d623c5",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-10-21T18:31:33.666012645Z",
     "start_time": "2023-10-21T18:31:33.662748099Z"
    }
   },
   "outputs": [],
   "source": [
    "train_dataset = ImageDataset(train_image_list)\n",
    "val_dataset = ImageDataset(val_image_list)\n",
    "\n",
    "train_dataloader = DataLoader(train_dataset, batch_size=BATCH_SIZE, shuffle=True)\n",
    "val_dataloader = DataLoader(val_dataset, batch_size=BATCH_SIZE, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "80b43c88421b7199",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-10-21T18:31:34.122859075Z",
     "start_time": "2023-10-21T18:31:34.115747116Z"
    }
   },
   "outputs": [],
   "source": [
    "class DenoisingAE(nn.Module):\n",
    "    def __init__(self, input_shape=IMG_SIZE):\n",
    "        super().__init__()\n",
    "        self.encoder = nn.Sequential(\n",
    "            #inp = 64x64\n",
    "            nn.Conv2d(3, 64, kernel_size=3, padding=1), \n",
    "            nn.BatchNorm2d(64),\n",
    "            nn.ReLU(True),\n",
    "            nn.Conv2d(64, 64, kernel_size=3, padding=1), \n",
    "            nn.BatchNorm2d(64),\n",
    "            nn.ReLU(True),\n",
    "            nn.MaxPool2d(kernel_size=2),  \n",
    "            \n",
    "            nn.Conv2d(64, 128, kernel_size=3, padding=1), \n",
    "            nn.BatchNorm2d(128),\n",
    "            nn.ReLU(True),\n",
    "            nn.Conv2d(128, 128, kernel_size=3, padding=1), \n",
    "            nn.BatchNorm2d(128),\n",
    "            nn.ReLU(True),\n",
    "            nn.MaxPool2d(kernel_size=2),  \n",
    "            \n",
    "            nn.Conv2d(128, 512, kernel_size=3, padding=1), \n",
    "            nn.BatchNorm2d(512),\n",
    "            nn.ReLU(True),\n",
    "            nn.Conv2d(512, 512, kernel_size=3, padding=1), \n",
    "            nn.BatchNorm2d(512),\n",
    "            nn.ReLU(True),\n",
    "            nn.MaxPool2d(kernel_size=2), \n",
    "        )\n",
    "        \n",
    "        self.decoder = nn.Sequential(\n",
    "            nn.ConvTranspose2d(512, 128, kernel_size=3, padding=1, stride=2, output_padding=1), \n",
    "            nn.ReLU(True),\n",
    "            nn.ConvTranspose2d(128, 128, kernel_size=3, padding=1 ),  # Upscale to 4x4\n",
    "            nn.ReLU(True),\n",
    "            nn.ConvTranspose2d(128, 64, kernel_size=3, padding=1, stride=2, output_padding=1),  \n",
    "            nn.ReLU(True),\n",
    "            nn.ConvTranspose2d(64, 64, kernel_size=3, padding=1 ),  \n",
    "            nn.ReLU(True),\n",
    "            nn.ConvTranspose2d(64, 3, kernel_size=3, padding=1, stride=2, output_padding=1),  # Upscale to 64x64\n",
    "            nn.Sigmoid(),\n",
    "        )\n",
    "    def forward(self, image):\n",
    "        image = image.to(DEVICE)\n",
    "        enc = self.encoder(image)\n",
    "        return self.decoder(enc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "43b57a4a8b08022e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-10-21T18:31:34.756211121Z",
     "start_time": "2023-10-21T18:31:34.721526117Z"
    }
   },
   "outputs": [],
   "source": [
    "model = DenoisingAE()\n",
    "model = model.to(device=DEVICE)\n",
    "model.train()\n",
    "opt = torch.optim.Adam(model.parameters())\n",
    "crit = nn.MSELoss()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "dc1cfcd011b4258a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-10-21T23:28:19.766308474Z",
     "start_time": "2023-10-21T18:31:36.630852454Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 500/500 [4:56:43<00:00, 35.61s/it]  \n"
     ]
    }
   ],
   "source": [
    "c = 0\n",
    "for e in tqdm(range(EPOCHS)):\n",
    "    for corrupt_img, img in train_dataloader:\n",
    "        corrupt_img, img = corrupt_img.to(DEVICE), img.to(DEVICE)\n",
    "        pred = model(corrupt_img)\n",
    "        loss = crit(pred, img)\n",
    "        writer.add_scalar(\"Loss/train\", scalar_value=loss.item(), global_step=c)\n",
    "        c+=1\n",
    "        \n",
    "        opt.zero_grad()\n",
    "        loss.backward()\n",
    "        opt.step()\n",
    "    writer.flush()\n",
    "    writer.add_image(f\"og after {e} epocs\", img[0])\n",
    "    writer.add_image(f\"pred after {e} epocs\", pred[0])\n",
    "    torch.save(model.state_dict(), \"DAE.pt\")\n",
    "\n",
    "writer.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bd15f18aeebb283",
   "metadata": {},
   "source": [
    "### I used tensorboard to track my loss hence attaching a screenshot of the graph here, have included tensorboard data in zip file I uploaded\n",
    "![Loss Graph](img.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e53a528d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<All keys matched successfully>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.load_state_dict(torch.load('DAE.pt'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "776d7935",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.eval()\n",
    "for corrupt_img, img in val_dataloader:\n",
    "    pred = model(corrupt_img)\n",
    "    display_img_tensors(img[0], corrupt_img[0], pred[0])\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "08f1e984",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}