Object Localization.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "be011338",
   "metadata": {},
   "source": [
    "# *Import Libraries:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "-D9Cu5Ki1sjx"
   },
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np \n",
    "import cv2\n",
    "import timm\n",
    "import torch \n",
    "from torch import nn \n",
    "import albumentations as A\n",
    "from tqdm.notebook import tqdm\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "import matplotlib.pyplot as plt "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "huIQiSlVKmK6"
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('object-localization-dataset')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8oM1P_LN2WzF"
   },
   "source": [
    "# *Configurations:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "l-Wz3fOX1yP6"
   },
   "outputs": [],
   "source": [
    "CSV_FILE = 'object-localization-dataset/train.csv'\n",
    "DATA_DIR = 'object-localization-dataset/'\n",
    "\n",
    "DEVICE = 'cuda'\n",
    "\n",
    "BATCH_SIZE = 16\n",
    "IMG_SIZE = 140\n",
    "\n",
    "LR = 1e-3\n",
    "EPOCHS = 40\n",
    "MODEL_NAME = 'efficientnet_b0'\n",
    "\n",
    "NUM_COR = 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "p7xh_WG929Pf"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>img_path</th>\n",
       "      <th>xmin</th>\n",
       "      <th>ymin</th>\n",
       "      <th>xmax</th>\n",
       "      <th>ymax</th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>train_images/mushroom_51.jpg</td>\n",
       "      <td>24</td>\n",
       "      <td>23</td>\n",
       "      <td>202</td>\n",
       "      <td>183</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>mushroom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>train_images/eggplant_37.jpg</td>\n",
       "      <td>34</td>\n",
       "      <td>34</td>\n",
       "      <td>88</td>\n",
       "      <td>201</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>eggplant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>train_images/mushroom_20.jpg</td>\n",
       "      <td>49</td>\n",
       "      <td>86</td>\n",
       "      <td>183</td>\n",
       "      <td>185</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>mushroom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>train_images/eggplant_51.jpg</td>\n",
       "      <td>51</td>\n",
       "      <td>59</td>\n",
       "      <td>191</td>\n",
       "      <td>164</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>eggplant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>train_images/eggplant_26.jpg</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "      <td>179</td>\n",
       "      <td>168</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>eggplant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>train_images/eggplant_62.jpg</td>\n",
       "      <td>67</td>\n",
       "      <td>22</td>\n",
       "      <td>177</td>\n",
       "      <td>215</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>eggplant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>train_images/cucumber_45.jpg</td>\n",
       "      <td>11</td>\n",
       "      <td>31</td>\n",
       "      <td>217</td>\n",
       "      <td>208</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>cucumber</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>train_images/mushroom_37.jpg</td>\n",
       "      <td>93</td>\n",
       "      <td>13</td>\n",
       "      <td>158</td>\n",
       "      <td>193</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>mushroom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>184</th>\n",
       "      <td>train_images/eggplant_44.jpg</td>\n",
       "      <td>21</td>\n",
       "      <td>59</td>\n",
       "      <td>192</td>\n",
       "      <td>171</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>eggplant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185</th>\n",
       "      <td>train_images/mushroom_16.jpg</td>\n",
       "      <td>43</td>\n",
       "      <td>20</td>\n",
       "      <td>197</td>\n",
       "      <td>182</td>\n",
       "      <td>227</td>\n",
       "      <td>227</td>\n",
       "      <td>mushroom</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>186 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         img_path  xmin  ymin  xmax  ymax  width  height  \\\n",
       "0    train_images/mushroom_51.jpg    24    23   202   183    227     227   \n",
       "1    train_images/eggplant_37.jpg    34    34    88   201    227     227   \n",
       "2    train_images/mushroom_20.jpg    49    86   183   185    227     227   \n",
       "3    train_images/eggplant_51.jpg    51    59   191   164    227     227   \n",
       "4    train_images/eggplant_26.jpg    40    70   179   168    227     227   \n",
       "..                            ...   ...   ...   ...   ...    ...     ...   \n",
       "181  train_images/eggplant_62.jpg    67    22   177   215    227     227   \n",
       "182  train_images/cucumber_45.jpg    11    31   217   208    227     227   \n",
       "183  train_images/mushroom_37.jpg    93    13   158   193    227     227   \n",
       "184  train_images/eggplant_44.jpg    21    59   192   171    227     227   \n",
       "185  train_images/mushroom_16.jpg    43    20   197   182    227     227   \n",
       "\n",
       "        label  \n",
       "0    mushroom  \n",
       "1    eggplant  \n",
       "2    mushroom  \n",
       "3    eggplant  \n",
       "4    eggplant  \n",
       "..        ...  \n",
       "181  eggplant  \n",
       "182  cucumber  \n",
       "183  mushroom  \n",
       "184  eggplant  \n",
       "185  mushroom  \n",
       "\n",
       "[186 rows x 8 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(CSV_FILE)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "_Dq7GKkk3ocN"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2812250e310>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "row = df.iloc[2]\n",
    "image = cv2.imread(DATA_DIR + row.img_path)\n",
    "image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
    "pt1 = (row.xmin, row.ymin)\n",
    "pt2 = (row.xmax, row.ymax)\n",
    "bnd_box_img = cv2.rectangle(image, pt1, pt2, (255, 0, 0), 2)\n",
    "plt.imshow(bnd_box_img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "j9IVE-2_34ab"
   },
   "outputs": [],
   "source": [
    "train_df, valid_df = train_test_split(df, test_size = 0.2, random_state=123)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NuHvzBs13_jt"
   },
   "source": [
    "# *Augmentations:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "8EeJAGT83-q1"
   },
   "outputs": [],
   "source": [
    "train_augs = A.Compose([\n",
    "    A.Resize(IMG_SIZE, IMG_SIZE),\n",
    "    A.HorizontalFlip(p=0.5),\n",
    "    A.VerticalFlip(p=0.5),\n",
    "    A.Rotate()\n",
    "], bbox_params=A.BboxParams(format = 'pascal_voc', label_fields = ['class_labels']))\n",
    "\n",
    "valid_augs = A.Compose([\n",
    "    A.Resize(IMG_SIZE, IMG_SIZE),\n",
    "], bbox_params=A.BboxParams(format = 'pascal_voc', label_fields = ['class_labels']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3GcEm3hr4KZ6"
   },
   "source": [
    "# *Custom Dataset:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "w-APc4_p4FiU"
   },
   "outputs": [],
   "source": [
    "class ObjLocDataset(torch.utils.data.Dataset):\n",
    "    \n",
    "    def __init__(self, df, augmentation):\n",
    "        self.df = df\n",
    "        self.augmentation = augmentation\n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.df)\n",
    "    \n",
    "    def __getitem__(self, idx):\n",
    "        \n",
    "        row = self.df.iloc[idx]\n",
    "        \n",
    "        xmin = row.xmin\n",
    "        ymin = row.ymin\n",
    "        xmax = row.xmax\n",
    "        ymax = row.ymax\n",
    "        \n",
    "        bboxes = [[xmin, ymin, xmax, ymax]]\n",
    "        \n",
    "        img_path = DATA_DIR + row.img_path\n",
    "        image = cv2.imread(img_path)\n",
    "        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
    "        \n",
    "        if self.augmentation:\n",
    "            data = self.augmentation(image = image, bboxes = bboxes, class_labels = [None])\n",
    "            image = data['image']\n",
    "            bboxes = data['bboxes'][0]\n",
    "        \n",
    "        # (h, w, c) -> (c, h, w)\n",
    "        image = torch.from_numpy(image).permute(2, 0, 1) / 255.\n",
    "        bboxes = torch.Tensor(bboxes)\n",
    "        \n",
    "        return image, bboxes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "Z3wIJmZ34SU8"
   },
   "outputs": [],
   "source": [
    "trainset = ObjLocDataset(train_df, train_augs)\n",
    "validset = ObjLocDataset(valid_df, valid_augs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "MiuNQ1845M0a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total example in train set : 148\n",
      "Total example in valid set : 38\n"
     ]
    }
   ],
   "source": [
    "print(f\"Total example in train set : {len(trainset)}\")\n",
    "print(f\"Total example in valid set : {len(validset)}\")      "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "lI0s9VPA4dye"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x281226350a0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img, bbox = trainset[5]\n",
    "\n",
    "xmin, ymin, xmax, ymax = bbox\n",
    "\n",
    "pt1 = (int(xmin), int(ymin))\n",
    "pt2 = (int(xmax), int(ymax))\n",
    "\n",
    "bnd_img = cv2.rectangle(img.permute(1, 2, 0).numpy(),pt1, pt2, (255,0,0), 2)\n",
    "plt.imshow(bnd_img)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9GeiS58d6C8x"
   },
   "source": [
    "# *Load dataset:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "1AALOj5h4jp-"
   },
   "outputs": [],
   "source": [
    "trainloader = torch.utils.data.DataLoader(trainset, batch_size = BATCH_SIZE, shuffle = True)\n",
    "validloader = torch.utils.data.DataLoader(validset, batch_size = BATCH_SIZE, shuffle = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "MmTSFxDf5doO"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total no. batches in trainloader : 10\n",
      "Total no. batches in validloader : 3\n"
     ]
    }
   ],
   "source": [
    "print(\"Total no. batches in trainloader : {}\".format(len(trainloader)))\n",
    "print(\"Total no. batches in validloader : {}\".format(len(validloader)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "X8QEUCAM5gMx"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of one batch images : torch.Size([16, 3, 140, 140])\n",
      "Shape of one batch bboxes : torch.Size([16, 4])\n"
     ]
    }
   ],
   "source": [
    "images, bboxes = next(iter(trainloader))\n",
    "\n",
    "print(\"Shape of one batch images : {}\".format(images.shape))\n",
    "print(\"Shape of one batch bboxes : {}\".format(bboxes.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gw47Me0s6NXo"
   },
   "source": [
    "# *Create Model:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "QXZW4mdk6Mpw"
   },
   "outputs": [],
   "source": [
    "class ObjLocModel(nn.Module):\n",
    "    \n",
    "    def __init__(self):\n",
    "        super(ObjLocModel, self).__init__()\n",
    "        \n",
    "        self.backbone = timm.create_model(MODEL_NAME, pretrained = True, num_classes = 4)\n",
    "        \n",
    "    def forward(self, images, gt_bboxes = None):\n",
    "        bboxes = self.backbone(images)\n",
    "        \n",
    "        if gt_bboxes != None:\n",
    "            loss = nn.MSELoss()(bboxes, gt_bboxes)\n",
    "            return bboxes, loss\n",
    "       \n",
    "        return bboxes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "1GM8RDN05tca"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ObjLocModel(\n",
       "  (backbone): EfficientNet(\n",
       "    (conv_stem): Conv2d(3, 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "    (bn1): BatchNormAct2d(\n",
       "      32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "      (drop): Identity()\n",
       "      (act): SiLU(inplace=True)\n",
       "    )\n",
       "    (blocks): Sequential(\n",
       "      (0): Sequential(\n",
       "        (0): DepthwiseSeparableConv(\n",
       "          (conv_dw): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(32, 8, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(8, 32, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pw): Conv2d(32, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "      )\n",
       "      (1): Sequential(\n",
       "        (0): InvertedResidual(\n",
       "          (conv_pw): Conv2d(16, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(96, 96, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=96, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(96, 4, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(4, 96, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (1): InvertedResidual(\n",
       "          (conv_pw): Conv2d(24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(144, 144, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=144, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(144, 6, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(6, 144, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(144, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "      )\n",
       "      (2): Sequential(\n",
       "        (0): InvertedResidual(\n",
       "          (conv_pw): Conv2d(24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(144, 144, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), groups=144, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(144, 6, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(6, 144, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(144, 40, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            40, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (1): InvertedResidual(\n",
       "          (conv_pw): Conv2d(40, 240, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            240, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(240, 240, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=240, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            240, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(240, 10, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(10, 240, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(240, 40, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            40, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "      )\n",
       "      (3): Sequential(\n",
       "        (0): InvertedResidual(\n",
       "          (conv_pw): Conv2d(40, 240, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            240, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(240, 240, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=240, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            240, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(240, 10, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(10, 240, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(240, 80, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (1): InvertedResidual(\n",
       "          (conv_pw): Conv2d(80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(480, 480, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=480, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(480, 20, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(20, 480, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(480, 80, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (2): InvertedResidual(\n",
       "          (conv_pw): Conv2d(80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(480, 480, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=480, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(480, 20, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(20, 480, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(480, 80, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "      )\n",
       "      (4): Sequential(\n",
       "        (0): InvertedResidual(\n",
       "          (conv_pw): Conv2d(80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(480, 480, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=480, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(480, 20, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(20, 480, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(480, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (1): InvertedResidual(\n",
       "          (conv_pw): Conv2d(112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(672, 672, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=672, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(672, 28, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(28, 672, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(672, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (2): InvertedResidual(\n",
       "          (conv_pw): Conv2d(112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(672, 672, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=672, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(672, 28, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(28, 672, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(672, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "      )\n",
       "      (5): Sequential(\n",
       "        (0): InvertedResidual(\n",
       "          (conv_pw): Conv2d(112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(672, 672, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), groups=672, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(672, 28, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(28, 672, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(672, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (1): InvertedResidual(\n",
       "          (conv_pw): Conv2d(192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(1152, 1152, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1152, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(1152, 48, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(48, 1152, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(1152, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (2): InvertedResidual(\n",
       "          (conv_pw): Conv2d(192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(1152, 1152, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1152, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(1152, 48, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(48, 1152, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(1152, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "        (3): InvertedResidual(\n",
       "          (conv_pw): Conv2d(192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(1152, 1152, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1152, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(1152, 48, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(48, 1152, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(1152, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "      )\n",
       "      (6): Sequential(\n",
       "        (0): InvertedResidual(\n",
       "          (conv_pw): Conv2d(192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn1): BatchNormAct2d(\n",
       "            1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (conv_dw): Conv2d(1152, 1152, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1152, bias=False)\n",
       "          (bn2): BatchNormAct2d(\n",
       "            1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): SiLU(inplace=True)\n",
       "          )\n",
       "          (se): SqueezeExcite(\n",
       "            (conv_reduce): Conv2d(1152, 48, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (act1): SiLU(inplace=True)\n",
       "            (conv_expand): Conv2d(48, 1152, kernel_size=(1, 1), stride=(1, 1))\n",
       "            (gate): Sigmoid()\n",
       "          )\n",
       "          (conv_pwl): Conv2d(1152, 320, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (bn3): BatchNormAct2d(\n",
       "            320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "            (drop): Identity()\n",
       "            (act): Identity()\n",
       "          )\n",
       "          (drop_path): Identity()\n",
       "        )\n",
       "      )\n",
       "    )\n",
       "    (conv_head): Conv2d(320, 1280, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "    (bn2): BatchNormAct2d(\n",
       "      1280, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "      (drop): Identity()\n",
       "      (act): SiLU(inplace=True)\n",
       "    )\n",
       "    (global_pool): SelectAdaptivePool2d (pool_type=avg, flatten=Flatten(start_dim=1, end_dim=-1))\n",
       "    (classifier): Linear(in_features=1280, out_features=4, bias=True)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = ObjLocModel()\n",
    "model.to(DEVICE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 4])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "random_image = torch.rand(1, 3, 140, 140).to(DEVICE)\n",
    "model(random_image).shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WKf-x-o06juW"
   },
   "source": [
    "# *Train and Validation Function:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "W31-rSfG6jUs"
   },
   "outputs": [],
   "source": [
    "def train_model(model, dataloader, optimizer):\n",
    "    total_loss = 0.0\n",
    "    model.train()\n",
    "    \n",
    "    for data in tqdm(dataloader):\n",
    "        \n",
    "        images, gt_bboxes = data\n",
    "        images, gt_bboxes = images.to(DEVICE), gt_bboxes.to(DEVICE)\n",
    "        \n",
    "        bboxes, loss = model(images, gt_bboxes)\n",
    "        \n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "        total_loss += loss.item()\n",
    "        \n",
    "    return total_loss / len(dataloader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "hnBMNmgk6gsq"
   },
   "outputs": [],
   "source": [
    "def eval_model(model, dataloader):\n",
    "    total_loss = 0.0\n",
    "    model.eval()\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        for data in tqdm(dataloader):\n",
    "\n",
    "            images, gt_bboxes = data\n",
    "            images, gt_bboxes = images.to(DEVICE), gt_bboxes.to(DEVICE)\n",
    "\n",
    "            bboxes, loss = model(images, gt_bboxes)\n",
    "\n",
    "            total_loss += loss.item()\n",
    "\n",
    "        return total_loss / len(dataloader)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LjiHeyg1Oh7k"
   },
   "source": [
    "# *Training Loop:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "FzoosXHs7NtU"
   },
   "outputs": [],
   "source": [
    "optimizer = torch.optim.Adam(model.parameters(), lr = LR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "iW_huEVT7kP3"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a2ca7f5530b644e785564df8c1a31983",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1f8a7932b8444a6cb79e68eb8cfb3ef6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 1 Train Loss : 5285.190258789063 Valid Loss : 4030.8392740885415\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6203e695a8cd4713913c5991c5e4b7ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d57175f1b3b42cf8d67e9c99c903f76",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 2 Train Loss : 2041.4535766601562 Valid Loss : 2853.0396321614585\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7df93bbb537f46d5b63b00e1fa01e602",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6621a66502cc49e2b61ae9164aa1ccbe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 3 Train Loss : 1192.5076110839843 Valid Loss : 394.2691650390625\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "18a25d8a3a874f44a2619abf1aed11bd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "189d23727b92493cba85d075bf75d60d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 4 Train Loss : 813.9419921875 Valid Loss : 292.7824198404948\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "50cf0a91f26d432a991b9573c04da81e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "05f84774f4684bf9a61f4d3fd047ec6a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 5 Train Loss : 574.1286865234375 Valid Loss : 134.78126780192056\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8615b2ee51e64b01ab09f65cc46dafe8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "49ca63c6e61440048ff5c28cea694473",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 6 Train Loss : 437.4954559326172 Valid Loss : 167.10302225748697\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "78be655e9cec4ceba4635fc14a9dce7d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a864b729cb3423dbb06bc18f0064c33",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 7 Train Loss : 322.2231246948242 Valid Loss : 194.02776082356772\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "375046fc5ba048d3b8ad87850fb3ebc2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d43981532c6847f492f9144af05b54da",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 8 Train Loss : 257.96254272460936 Valid Loss : 124.42020416259766\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e93d618f11f440d094c14632dca5e7ad",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5cff32612be5490eac4c4a9339ba9212",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 9 Train Loss : 195.49150390625 Valid Loss : 110.5224380493164\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "354e87a7f7c448d597b2c04456d89cb8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0849210536d348e780fc1a58a93ac495",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 10 Train Loss : 173.37640533447265 Valid Loss : 125.033203125\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "be9606318a3a419d93ccaae4d662bec1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a102e320adc44640b91e7a1869526d30",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 11 Train Loss : 129.1849807739258 Valid Loss : 125.92116292317708\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ae2a2435d1ee4231b5a9e9418492bdaa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "324cf6937754486fb20141a6b8f4e769",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 12 Train Loss : 120.32753219604493 Valid Loss : 117.29336039225261\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f630522e12f43cfbdc054d72239fb06",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ceb872e80b8c40f8b7ec82752a2a693b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 13 Train Loss : 122.8256217956543 Valid Loss : 108.17993672688802\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d49d96d4c1b545dc892743466cfb7ba5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1c4c5c8eee654bb1af9d59a9113c40aa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 14 Train Loss : 105.03789939880372 Valid Loss : 108.33396911621094\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cc5c0e717ecd46ff834abd31fa872d05",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "04d702da30ce450fa01e2a6672c2d72a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 15 Train Loss : 101.31521835327149 Valid Loss : 86.91665649414062\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "54503b9f6bed4d2793a49bad7515ee5d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0bd21929ba3544a5b5ae452cc2485dbd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 16 Train Loss : 97.59057159423828 Valid Loss : 102.38520050048828\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1026e3ce7e3e49e9acc9903159848338",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "395197c7487945d4a0ac4ce4df2c2b25",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 17 Train Loss : 83.40586585998535 Valid Loss : 87.70598602294922\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "33c0b9b855944ff1864ebe4af88643cf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4e61f4daaf8b4989bca901d91b978417",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 18 Train Loss : 98.4215633392334 Valid Loss : 83.62984720865886\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3c1fa28b86f0402ab04c7821eedb8a87",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7eb4558718ee466a8ed3a156d1086247",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 19 Train Loss : 86.42916946411133 Valid Loss : 78.3333231608073\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c8a3f51281b84a49b004262b51d720ba",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0399836003ce4c4f9ac5ce04891f675a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 20 Train Loss : 77.14911575317383 Valid Loss : 83.62183888753255\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1aa1ca51548c48259744a02942118b5a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1605948618424c22bf83f13b8a025f20",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 21 Train Loss : 59.93084144592285 Valid Loss : 66.6417948404948\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c759d9f455034f3ea2b765492f741bad",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ff133de737674feba6a857dbd92fce57",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 22 Train Loss : 65.73363647460937 Valid Loss : 64.1520004272461\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ff385255b7f54870a07ee1835e28e045",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6edba0459ab34c74b8ad0f1ddd5b27b6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 23 Train Loss : 50.21168231964111 Valid Loss : 56.36591593424479\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "92fba77f536b444d8a44ed060953f036",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c804d1655dd44a9096cdbac4973ea54a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 24 Train Loss : 58.5775203704834 Valid Loss : 57.760450998942055\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cdd0f0b83ce64c05a3a4bc86ffac3995",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0c41a40c3e954119a9f1b90b66c9deaf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 25 Train Loss : 64.84975643157959 Valid Loss : 45.58387056986491\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b1dae6e248eb4f33875249b9628a5473",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d7a3e05eb648434299fbc4c4c1131c0f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 26 Train Loss : 44.95496654510498 Valid Loss : 66.49660110473633\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a889577e528a44c5a9b66f491b505d26",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d336df93d6614e8b89bee0531b58e51d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 27 Train Loss : 44.360662651062015 Valid Loss : 60.341253916422524\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d90e2acf3141444a929688837c8381af",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "34cfc63201084c30b276d0cdc396dcc8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 28 Train Loss : 45.59176254272461 Valid Loss : 58.52057393391927\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00bb0eb5d24448feb250a0cb014aeb80",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1793f047568843acbc99c0b206b833af",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 29 Train Loss : 44.51098899841308 Valid Loss : 45.73501523335775\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3ecba282007d41e782e2c923fa6845bf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "012ab3daf53f444fb9570329c2c28364",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 30 Train Loss : 58.48039588928223 Valid Loss : 48.4793955485026\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cb6fe10c4140466e8002e75c43d66221",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7a7fe28cf2ce41beb4e1caee6b35d41e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 31 Train Loss : 47.06371250152588 Valid Loss : 64.61798731486003\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2012f3c443aa45079b34fb187ff89af4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "43a452e6fdb54d8dbc9f0126dd8ee976",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 32 Train Loss : 61.522269058227536 Valid Loss : 78.04480743408203\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "240508b6030444749d48a9ea39b8f33e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e07d273fd0d645799a39cae6c699e465",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL ARE SAVED\n",
      "Epoch : 33 Train Loss : 56.112228584289554 Valid Loss : 42.33534622192383\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9b1667aa6b2847e2a6d7091b21ec7897",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9d9effcfeb5e4dff9ef8dc7ab8d0660b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 34 Train Loss : 52.52180194854736 Valid Loss : 49.865946451822914\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b7fb60edf3cd4cbbb63b1dbc86974e83",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "26c590844782480db9386c8323cd4d84",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 35 Train Loss : 44.99024505615235 Valid Loss : 92.55803934733073\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "407b86916f6b409fbac49f0b32b2465e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7bf93c9373b54b54bebafa58ff3c4176",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 36 Train Loss : 55.0930570602417 Valid Loss : 60.5361696879069\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "09a3493f161b4b1e9ffed184bbf10737",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ec64e0dbcdf54410ad9c8a863ae604e1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 37 Train Loss : 39.35981740951538 Valid Loss : 51.42660649617513\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c15a59133e0449919d80bacb2ed8c4f7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6edca5c237d844c198b5cb1fd01bbf26",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 38 Train Loss : 42.75083770751953 Valid Loss : 45.22932815551758\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c3be70fc5e3247739db1f334e9c9d1f1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4ae0865b2af54e7c81b363d992095acc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 39 Train Loss : 48.31570930480957 Valid Loss : 82.61095937093098\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4bba1cc1c57a4be089532452fc510a22",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b65e5493e2f44f4383229a068869b92c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch : 40 Train Loss : 44.1829610824585 Valid Loss : 62.907318115234375\n"
     ]
    }
   ],
   "source": [
    "best_valid_loss = np.Inf\n",
    "\n",
    "for i in range(EPOCHS):\n",
    "    \n",
    "    train_loss = train_model(model, trainloader, optimizer)\n",
    "    valid_loss = eval_model(model, validloader)\n",
    "    \n",
    "    if valid_loss < best_valid_loss:\n",
    "        torch.save(model.state_dict(), 'best_model.pt')\n",
    "        print('MODEL ARE SAVED')\n",
    "        best_valid_loss = valid_loss\n",
    "        \n",
    "    print(f\"Epoch : {i+1} Train Loss : {train_loss} Valid Loss : {valid_loss}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "id": "EyNZhifNKGMQ"
   },
   "outputs": [],
   "source": [
    "def compare_plots(image, gt_bbox, out_bbox):\n",
    "\n",
    "    xmin, ymin, xmax, ymax = gt_bbox\n",
    "\n",
    "    pt1 = (int(xmin), int(ymin))\n",
    "    pt2 = (int(xmax), int(ymax))\n",
    "\n",
    "    out_xmin, out_ymin, out_xmax, out_ymax = out_bbox[0]\n",
    "\n",
    "    out_pt1 = (int(out_xmin), int(out_ymin))\n",
    "    out_pt2 = (int(out_xmax), int(out_ymax))\n",
    "\n",
    "    out_img = cv2.rectangle(image.squeeze().permute(1, 2, 0).cpu().numpy(),pt1, pt2,(0,255,0),2)\n",
    "    out_img = cv2.rectangle(out_img,out_pt1, out_pt2,(255,0,0),2)\n",
    "    plt.imshow(out_img)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "pBTq15Oz7oBq"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.load_state_dict(torch.load('best_model.pt'))\n",
    "model.eval()\n",
    "\n",
    "with torch.no_grad():\n",
    "    \n",
    "    image, gt_bbox = validset[28]\n",
    "    image = image.unsqueeze(0).to(DEVICE)\n",
    "    out_bbox = model(image)\n",
    "    \n",
    "    compare_plots(image, gt_bbox, out_bbox)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "authorship_tag": "ABX9TyO1JYgO4jQmM/EoY+feesnQ",
   "collapsed_sections": [],
   "name": "Deep Learning with PyTorch : Object Localization.ipynb",
   "provenance": []
  },
  "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}