sign_language_pytorch.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "view-in-github"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Arshad221b/Sign-Language-Recognition/blob/master/sign_language_pytorch.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 185,
      "metadata": {
        "id": "p_WhBfDb_2y5"
      },
      "outputs": [],
      "source": [
        "import torch \n",
        "import torch.nn as nn \n",
        "import pandas as pd \n",
        "import numpy as np \n",
        "import matplotlib.pyplot as plt \n",
        "from sklearn.model_selection import train_test_split\n",
        "from torch.utils.data import Dataset, DataLoader\n",
        "from torchvision.transforms import transforms\n",
        "from PIL import Image\n",
        "from tqdm import tqdm\n",
        "from torch.utils.data import Dataset\n",
        "import torch\n",
        "import numpy as np\n",
        "from PIL import Image\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "metadata": {
        "id": "cmFIz-UoLgtn"
      },
      "outputs": [],
      "source": [
        "device = 'cuda' if torch.cuda.is_available() else 'cpu'"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MUdQPpI0MKLp"
      },
      "source": [
        "# Import Data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v7Ulzw6uBRNH",
        "outputId": "133633a4-4695-4f29-e582-a5579d51a238"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "mkdir: cannot create directory ‘/root/.kaggle’: File exists\n",
            "Warning: Your Kaggle API key is readable by other users on this system! To fix this, you can run 'chmod 600 /root/.kaggle/kaggle.json'\n",
            "Downloading sign-language-mnist.zip to /content\n",
            " 94% 59.0M/62.6M [00:02<00:00, 41.6MB/s]\n",
            "100% 62.6M/62.6M [00:02<00:00, 30.0MB/s]\n"
          ]
        }
      ],
      "source": [
        "! mkdir ~/.kaggle\n",
        "! cp kaggle.json ~/.kaggle/\n",
        "\n",
        "! kaggle datasets download -d datamunge/sign-language-mnist"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2VyrYNNBB5Z9",
        "outputId": "b2b79aa3-a622-4ca6-c643-97c82f57d2ef"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Archive:  /content/sign-language-mnist.zip\n",
            "  inflating: amer_sign2.png          \n",
            "  inflating: amer_sign3.png          \n",
            "  inflating: american_sign_language.PNG  \n",
            "  inflating: sign_mnist_test.csv     \n",
            "  inflating: sign_mnist_test/sign_mnist_test.csv  \n",
            "  inflating: sign_mnist_train.csv    \n",
            "  inflating: sign_mnist_train/sign_mnist_train.csv  \n"
          ]
        }
      ],
      "source": [
        "! unzip /content/sign-language-mnist.zip"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "nVGeqCWwDH6R"
      },
      "outputs": [],
      "source": [
        "train = pd.read_csv('kidus_b/contents/sign_mnist_train/sign_mnist_train.csv')\n",
        "test = pd.read_csv('kidus_b/contents/sign_mnist_test/sign_mnist_test.csv')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 233
        },
        "id": "8vwKEQObDVQJ",
        "outputId": "3be037a7-1859-4aec-d6b3-7fff844f5973"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-31bb0d3d-4eee-419d-b5e8-74c1df0dd397\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>label</th>\n",
              "      <th>pixel1</th>\n",
              "      <th>pixel2</th>\n",
              "      <th>pixel3</th>\n",
              "      <th>pixel4</th>\n",
              "      <th>pixel5</th>\n",
              "      <th>pixel6</th>\n",
              "      <th>pixel7</th>\n",
              "      <th>pixel8</th>\n",
              "      <th>pixel9</th>\n",
              "      <th>...</th>\n",
              "      <th>pixel775</th>\n",
              "      <th>pixel776</th>\n",
              "      <th>pixel777</th>\n",
              "      <th>pixel778</th>\n",
              "      <th>pixel779</th>\n",
              "      <th>pixel780</th>\n",
              "      <th>pixel781</th>\n",
              "      <th>pixel782</th>\n",
              "      <th>pixel783</th>\n",
              "      <th>pixel784</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3</td>\n",
              "      <td>107</td>\n",
              "      <td>118</td>\n",
              "      <td>127</td>\n",
              "      <td>134</td>\n",
              "      <td>139</td>\n",
              "      <td>143</td>\n",
              "      <td>146</td>\n",
              "      <td>150</td>\n",
              "      <td>153</td>\n",
              "      <td>...</td>\n",
              "      <td>207</td>\n",
              "      <td>207</td>\n",
              "      <td>207</td>\n",
              "      <td>207</td>\n",
              "      <td>206</td>\n",
              "      <td>206</td>\n",
              "      <td>206</td>\n",
              "      <td>204</td>\n",
              "      <td>203</td>\n",
              "      <td>202</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>6</td>\n",
              "      <td>155</td>\n",
              "      <td>157</td>\n",
              "      <td>156</td>\n",
              "      <td>156</td>\n",
              "      <td>156</td>\n",
              "      <td>157</td>\n",
              "      <td>156</td>\n",
              "      <td>158</td>\n",
              "      <td>158</td>\n",
              "      <td>...</td>\n",
              "      <td>69</td>\n",
              "      <td>149</td>\n",
              "      <td>128</td>\n",
              "      <td>87</td>\n",
              "      <td>94</td>\n",
              "      <td>163</td>\n",
              "      <td>175</td>\n",
              "      <td>103</td>\n",
              "      <td>135</td>\n",
              "      <td>149</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>187</td>\n",
              "      <td>188</td>\n",
              "      <td>188</td>\n",
              "      <td>187</td>\n",
              "      <td>187</td>\n",
              "      <td>186</td>\n",
              "      <td>187</td>\n",
              "      <td>188</td>\n",
              "      <td>187</td>\n",
              "      <td>...</td>\n",
              "      <td>202</td>\n",
              "      <td>201</td>\n",
              "      <td>200</td>\n",
              "      <td>199</td>\n",
              "      <td>198</td>\n",
              "      <td>199</td>\n",
              "      <td>198</td>\n",
              "      <td>195</td>\n",
              "      <td>194</td>\n",
              "      <td>195</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2</td>\n",
              "      <td>211</td>\n",
              "      <td>211</td>\n",
              "      <td>212</td>\n",
              "      <td>212</td>\n",
              "      <td>211</td>\n",
              "      <td>210</td>\n",
              "      <td>211</td>\n",
              "      <td>210</td>\n",
              "      <td>210</td>\n",
              "      <td>...</td>\n",
              "      <td>235</td>\n",
              "      <td>234</td>\n",
              "      <td>233</td>\n",
              "      <td>231</td>\n",
              "      <td>230</td>\n",
              "      <td>226</td>\n",
              "      <td>225</td>\n",
              "      <td>222</td>\n",
              "      <td>229</td>\n",
              "      <td>163</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>13</td>\n",
              "      <td>164</td>\n",
              "      <td>167</td>\n",
              "      <td>170</td>\n",
              "      <td>172</td>\n",
              "      <td>176</td>\n",
              "      <td>179</td>\n",
              "      <td>180</td>\n",
              "      <td>184</td>\n",
              "      <td>185</td>\n",
              "      <td>...</td>\n",
              "      <td>92</td>\n",
              "      <td>105</td>\n",
              "      <td>105</td>\n",
              "      <td>108</td>\n",
              "      <td>133</td>\n",
              "      <td>163</td>\n",
              "      <td>157</td>\n",
              "      <td>163</td>\n",
              "      <td>164</td>\n",
              "      <td>179</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 785 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-31bb0d3d-4eee-419d-b5e8-74c1df0dd397')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-31bb0d3d-4eee-419d-b5e8-74c1df0dd397 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-31bb0d3d-4eee-419d-b5e8-74c1df0dd397');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   label  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  \\\n",
              "0      3     107     118     127     134     139     143     146     150   \n",
              "1      6     155     157     156     156     156     157     156     158   \n",
              "2      2     187     188     188     187     187     186     187     188   \n",
              "3      2     211     211     212     212     211     210     211     210   \n",
              "4     13     164     167     170     172     176     179     180     184   \n",
              "\n",
              "   pixel9  ...  pixel775  pixel776  pixel777  pixel778  pixel779  pixel780  \\\n",
              "0     153  ...       207       207       207       207       206       206   \n",
              "1     158  ...        69       149       128        87        94       163   \n",
              "2     187  ...       202       201       200       199       198       199   \n",
              "3     210  ...       235       234       233       231       230       226   \n",
              "4     185  ...        92       105       105       108       133       163   \n",
              "\n",
              "   pixel781  pixel782  pixel783  pixel784  \n",
              "0       206       204       203       202  \n",
              "1       175       103       135       149  \n",
              "2       198       195       194       195  \n",
              "3       225       222       229       163  \n",
              "4       157       163       164       179  \n",
              "\n",
              "[5 rows x 785 columns]"
            ]
          },
          "execution_count": 10,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "train.head(5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "9091-cLADWXx"
      },
      "outputs": [],
      "source": [
        "X = train.drop(['label'], axis = 1)\n",
        "y = train['label']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 168,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S_9BYi6pTil-",
        "outputId": "ed917c06-8def-4868-ef17-b622ce3f2901"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "((27455, 784), (27455,))"
            ]
          },
          "execution_count": 168,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "X.shape, y.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "PygT6Ut4DhUI"
      },
      "outputs": [],
      "source": [
        "X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Wz1ufFTBMNS5"
      },
      "source": [
        "# Hyperparameters"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 169,
      "metadata": {
        "id": "MhLEi1OqMQ42"
      },
      "outputs": [],
      "source": [
        "\n",
        "#class SignLanguageDataset(Dataset):\n",
        "LEARNING_RATE = 0.001\n",
        "EPOCHS = 50\n",
        "OPTIMIZER = 'Adam' \n",
        "BATCH_SIZE = 32\n",
        "IMAGE_SIZE = 28\n",
        "NUM_CLASS = y.nunique() + 1  \n",
        "LOSS_FUNCTION = 'categorical_crossentropy' \n",
        "DROPOUT_RATE = 0.5  \n",
        "MOMENTUM = 0.9\n",
        "WEIGHT_INIT = 'He_normal'  \n",
        "ACTIVATION_FUNCTION = 'ReLU'  \n",
        "REGULARIZATION = 'L2'  \n",
        "LR_DECAY = 1e-6  \n",
        "BATCH_NORM = True  \n",
        "NUM_LAYERS = 3  \n",
        "UNITS_PER_LAYER = [64, 128, 64]  \n",
        "GRADIENT_CLIPPING = 5.0  \n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hh9y4TBVH7Z-"
      },
      "source": [
        "# Preprocessing and Data Loader"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 170,
      "metadata": {
        "id": "1flAi02GH-Z5"
      },
      "outputs": [],
      "source": [
        "random_transforms = transforms.Compose([\n",
        "    transforms.RandomRotation(30),  \n",
        "])\n",
        "\n",
        "\n",
        "fixed_transforms = transforms.Compose([\n",
        "    transforms.ToTensor(),\n",
        "    transforms.Normalize(mean=[0.5], std=[0.5])\n",
        "])\n",
        "\n",
        "\n",
        "transform = transforms.Compose([\n",
        "    transforms.RandomApply([random_transforms], p=0.5), \n",
        "    fixed_transforms\n",
        "])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 178,
      "metadata": {
        "id": "g3FNAGmTDsk4"
      },
      "outputs": [],
      "source": [
        "\n",
        "class SignDataSet(Dataset):\n",
        "    def __init__(self, image_df, label_df, transform, split=None):\n",
        "        self.image_df = image_df\n",
        "        self.label_df = torch.nn.functional.one_hot(torch.tensor(label_df.astype(np.int64))).float()\n",
        "        self.transform = transform\n",
        "        self.split = split\n",
        "\n",
        "    def __len__(self):\n",
        "        return len(self.image_df)\n",
        "\n",
        "    def __getitem__(self, index):\n",
        "        image = self.image_df.iloc[index].values.reshape(28, 28).astype(np.uint8)\n",
        "        image = Image.fromarray(image)\n",
        "        label = self.label_df[index]\n",
        "\n",
        "        if self.split == 'train':\n",
        "            image = self.transform(image)\n",
        "        \n",
        "        return image, label\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 179,
      "metadata": {
        "id": "CrHfPlQTEsva"
      },
      "outputs": [],
      "source": [
        "\n",
        "train_data = SignDataSet(X_train, y_train, transform, 'train')\n",
        "test_data = SignDataSet(X_val, y_val, transform)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 180,
      "metadata": {
        "id": "xhBN15fDEtnt"
      },
      "outputs": [],
      "source": [
        "train_loader = DataLoader(train_data, batch_size = BATCH_SIZE, drop_last = True)\n",
        "test_loader = DataLoader(test_data, batch_size = BATCH_SIZE, drop_last = True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 181,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "bUukorhQJMGt",
        "outputId": "781b0e59-5b5b-468c-cfbe-bf0eef59a075"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "for i, j in train_data: \n",
        "  temp_img = i \n",
        "\n",
        "  plt.imshow(temp_img.permute(1,2,0))\n",
        "  break"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EiVtrOj8R1RD"
      },
      "source": [
        "# Model Building"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 182,
      "metadata": {
        "id": "scS5lSmRR0_W"
      },
      "outputs": [],
      "source": [
        "class SignLabelModel(nn.Module):\n",
        "    def __init__(self, num_classes):\n",
        "        super(SignLabelModel, self).__init__()\n",
        "        self.features = nn.Sequential(\n",
        "            nn.Conv2d(1, 16, kernel_size=3, stride=1, padding=1),\n",
        "            nn.ReLU(inplace=True),\n",
        "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
        "            nn.Conv2d(16, 32, kernel_size=3, stride=1, padding=1),\n",
        "            nn.ReLU(inplace=True),\n",
        "            nn.MaxPool2d(kernel_size=2, stride=2)\n",
        "        )\n",
        "        self.classifier = nn.Sequential(\n",
        "            nn.Linear(32 * 7 * 7, 128),\n",
        "            nn.ReLU(inplace=True),\n",
        "            nn.Linear(128, num_classes)\n",
        "        )\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.features(x)\n",
        "        x = x.view(x.size(0), -1)\n",
        "        x = self.classifier(x)\n",
        "        return x\n",
        "\n",
        "\n",
        "model = SignLabelModel(NUM_CLASS).to(device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 187,
      "metadata": {
        "id": "Bx0_V_AFSTmd"
      },
      "outputs": [],
      "source": [
        "num_epochs = 20\n",
        "criterion  = nn.CrossEntropyLoss()\n",
        "optimizer = torch.optim.Adam(model.parameters(), lr=0.0001)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 188,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AlGYqVvQK2b1",
        "outputId": "f4bac700-940b-47db-b58c-127ac9a93c65"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r  0%|          | 0/20 [00:00<?, ?it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [1/20], Loss: 0.0658\n",
            "Epoch [1/20], Loss: 0.2724\n",
            "Epoch [1/20], Loss: 0.2633\n",
            "Epoch [1/20], Loss: 0.1275\n",
            "Epoch [1/20], Loss: 0.1453\n",
            "Epoch [1/20], Loss: 0.1104\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r  5%|▌         | 1/20 [00:12<03:50, 12.11s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [2/20], Loss: 0.1661\n",
            "Epoch [2/20], Loss: 0.1278\n",
            "Epoch [2/20], Loss: 0.1764\n",
            "Epoch [2/20], Loss: 0.2534\n",
            "Epoch [2/20], Loss: 0.2307\n",
            "Epoch [2/20], Loss: 0.0386\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 10%|█         | 2/20 [00:23<03:34, 11.94s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [3/20], Loss: 0.0690\n",
            "Epoch [3/20], Loss: 0.2340\n",
            "Epoch [3/20], Loss: 0.1753\n",
            "Epoch [3/20], Loss: 0.0758\n",
            "Epoch [3/20], Loss: 0.1124\n",
            "Epoch [3/20], Loss: 0.0213\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 15%|█▌        | 3/20 [00:35<03:22, 11.92s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [4/20], Loss: 0.0999\n",
            "Epoch [4/20], Loss: 0.3457\n",
            "Epoch [4/20], Loss: 0.2807\n",
            "Epoch [4/20], Loss: 0.1919\n",
            "Epoch [4/20], Loss: 0.1315\n",
            "Epoch [4/20], Loss: 0.0663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 20%|██        | 4/20 [00:47<03:11, 11.99s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [5/20], Loss: 0.0740\n",
            "Epoch [5/20], Loss: 0.0681\n",
            "Epoch [5/20], Loss: 0.0753\n",
            "Epoch [5/20], Loss: 0.1335\n",
            "Epoch [5/20], Loss: 0.0359\n",
            "Epoch [5/20], Loss: 0.0640\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 25%|██▌       | 5/20 [01:00<03:01, 12.08s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [6/20], Loss: 0.0824\n",
            "Epoch [6/20], Loss: 0.3208\n",
            "Epoch [6/20], Loss: 0.1746\n",
            "Epoch [6/20], Loss: 0.1462\n",
            "Epoch [6/20], Loss: 0.0325\n",
            "Epoch [6/20], Loss: 0.0166\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 30%|███       | 6/20 [01:12<02:49, 12.11s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [7/20], Loss: 0.0266\n",
            "Epoch [7/20], Loss: 0.0698\n",
            "Epoch [7/20], Loss: 0.1176\n",
            "Epoch [7/20], Loss: 0.0971\n",
            "Epoch [7/20], Loss: 0.0727\n",
            "Epoch [7/20], Loss: 0.0162\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 35%|███▌      | 7/20 [01:24<02:36, 12.07s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [8/20], Loss: 0.1202\n",
            "Epoch [8/20], Loss: 0.0308\n",
            "Epoch [8/20], Loss: 0.2305\n",
            "Epoch [8/20], Loss: 0.0595\n",
            "Epoch [8/20], Loss: 0.0532\n",
            "Epoch [8/20], Loss: 0.0617\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 40%|████      | 8/20 [01:36<02:24, 12.07s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [9/20], Loss: 0.0548\n",
            "Epoch [9/20], Loss: 0.2174\n",
            "Epoch [9/20], Loss: 0.0857\n",
            "Epoch [9/20], Loss: 0.0317\n",
            "Epoch [9/20], Loss: 0.0402\n",
            "Epoch [9/20], Loss: 0.0114\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 45%|████▌     | 9/20 [01:48<02:12, 12.04s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [10/20], Loss: 0.1095\n",
            "Epoch [10/20], Loss: 0.0566\n",
            "Epoch [10/20], Loss: 0.0696\n",
            "Epoch [10/20], Loss: 0.0521\n",
            "Epoch [10/20], Loss: 0.0177\n",
            "Epoch [10/20], Loss: 0.0366\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 50%|█████     | 10/20 [02:00<02:00, 12.05s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [11/20], Loss: 0.1117\n",
            "Epoch [11/20], Loss: 0.0173\n",
            "Epoch [11/20], Loss: 0.0519\n",
            "Epoch [11/20], Loss: 0.0278\n",
            "Epoch [11/20], Loss: 0.1070\n",
            "Epoch [11/20], Loss: 0.0231\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 55%|█████▌    | 11/20 [02:12<01:48, 12.03s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [12/20], Loss: 0.0667\n",
            "Epoch [12/20], Loss: 0.1846\n",
            "Epoch [12/20], Loss: 0.0576\n",
            "Epoch [12/20], Loss: 0.0531\n",
            "Epoch [12/20], Loss: 0.0215\n",
            "Epoch [12/20], Loss: 0.0326\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 60%|██████    | 12/20 [02:24<01:35, 11.98s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [13/20], Loss: 0.0168\n",
            "Epoch [13/20], Loss: 0.0464\n",
            "Epoch [13/20], Loss: 0.0739\n",
            "Epoch [13/20], Loss: 0.0704\n",
            "Epoch [13/20], Loss: 0.0531\n",
            "Epoch [13/20], Loss: 0.0398\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 65%|██████▌   | 13/20 [02:36<01:23, 11.95s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [14/20], Loss: 0.0316\n",
            "Epoch [14/20], Loss: 0.0414\n",
            "Epoch [14/20], Loss: 0.0685\n",
            "Epoch [14/20], Loss: 0.1667\n",
            "Epoch [14/20], Loss: 0.0165\n",
            "Epoch [14/20], Loss: 0.0066\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 70%|███████   | 14/20 [02:48<01:11, 11.95s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [15/20], Loss: 0.0273\n",
            "Epoch [15/20], Loss: 0.0591\n",
            "Epoch [15/20], Loss: 0.0187\n",
            "Epoch [15/20], Loss: 0.0399\n",
            "Epoch [15/20], Loss: 0.0506\n",
            "Epoch [15/20], Loss: 0.0397\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 75%|███████▌  | 15/20 [03:00<00:59, 11.96s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [16/20], Loss: 0.0701\n",
            "Epoch [16/20], Loss: 0.0747\n",
            "Epoch [16/20], Loss: 0.0265\n",
            "Epoch [16/20], Loss: 0.0180\n",
            "Epoch [16/20], Loss: 0.0547\n",
            "Epoch [16/20], Loss: 0.0278\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 80%|████████  | 16/20 [03:11<00:47, 11.94s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [17/20], Loss: 0.0774\n",
            "Epoch [17/20], Loss: 0.1419\n",
            "Epoch [17/20], Loss: 0.0338\n",
            "Epoch [17/20], Loss: 0.0303\n",
            "Epoch [17/20], Loss: 0.0251\n",
            "Epoch [17/20], Loss: 0.0426\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 85%|████████▌ | 17/20 [03:24<00:35, 11.99s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [18/20], Loss: 0.1617\n",
            "Epoch [18/20], Loss: 0.0936\n",
            "Epoch [18/20], Loss: 0.0481\n",
            "Epoch [18/20], Loss: 0.0486\n",
            "Epoch [18/20], Loss: 0.0931\n",
            "Epoch [18/20], Loss: 0.0111\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 90%|█████████ | 18/20 [03:36<00:23, 11.97s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [19/20], Loss: 0.1165\n",
            "Epoch [19/20], Loss: 0.1108\n",
            "Epoch [19/20], Loss: 0.0284\n",
            "Epoch [19/20], Loss: 0.0708\n",
            "Epoch [19/20], Loss: 0.0253\n",
            "Epoch [19/20], Loss: 0.0037\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\r 95%|█████████▌| 19/20 [03:48<00:11, 11.98s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch [20/20], Loss: 0.0214\n",
            "Epoch [20/20], Loss: 0.0849\n",
            "Epoch [20/20], Loss: 0.0236\n",
            "Epoch [20/20], Loss: 0.0095\n",
            "Epoch [20/20], Loss: 0.0024\n",
            "Epoch [20/20], Loss: 0.0082\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 20/20 [03:59<00:00, 12.00s/it]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Training finished.\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        }
      ],
      "source": [
        "for epoch in tqdm(range(num_epochs)):\n",
        "    for i, (images, labels) in enumerate(train_loader):\n",
        "        # Move images and labels to the device\n",
        "        images = images.to(device)\n",
        "        labels = labels.to(device)\n",
        "\n",
        "        # Forward pass\n",
        "        outputs = model(images.to(device))\n",
        "        loss = criterion(outputs, labels)\n",
        "\n",
        "        # Backward and optimize\n",
        "        optimizer.zero_grad()\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "\n",
        "        # Print training progress\n",
        "        if (i+1) % 100 == 0:\n",
        "            print(f\"Epoch [{epoch+1}/{num_epochs}], Loss: {loss.item():.4f}\")\n",
        "\n",
        "print(\"Training finished.\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "uz1aEUDCPq_x"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "authorship_tag": "ABX9TyPQXDa+HvS3VuC0eoyCgQ3s",
      "gpuType": "T4",
      "include_colab_link": true,
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}