Classification/ResNext(2017).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ResNext(2017).ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPUo0SooRJgsXfy35EY1PxH",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Seonghoon-Yu/Paper_Review_and_Implementation_in_PyTorch/blob/master/Classification/ResNext(2017).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TBX5KeCFwCXO",
        "outputId": "beeebf44-16ca-40cf-a39c-26a1b52c9d31"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('resnext')"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at resnext\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fGvj6DMJxORX"
      },
      "source": [
        "# import package\n",
        "\n",
        "# model\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "from torchsummary import summary\n",
        "from torch import optim\n",
        "\n",
        "# dataset and transformation\n",
        "from torchvision import datasets\n",
        "import torchvision.transforms as transforms\n",
        "from torch.utils.data import DataLoader\n",
        "import os\n",
        "\n",
        "# display images\n",
        "from torchvision import utils\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "# utils\n",
        "import numpy as np\n",
        "from torchsummary import summary\n",
        "import time\n",
        "import copy"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3F1VA7N4xj18"
      },
      "source": [
        "# 1. 데이터셋 불러오기\n",
        "- 데이터셋은 torchvision 패키지에서 제공하는 STL10 dataset을 이용하겠습니다. STL10 dataset은 10개의 label을 갖으며 train dataset 5000개, test dataset 8000개로 구성됩니다."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9zq5YHCuxjYZ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "972bd2d5-4c55-47b7-dd4b-c80ec62009ee"
      },
      "source": [
        "# specify path to data\n",
        "path2data = '/content/resnext/MyDrive/data'\n",
        "\n",
        "# if not exists the path, make the directory\n",
        "if not os.path.exists(path2data):\n",
        "    os.mkdir(path2data)\n",
        "\n",
        "# load dataset\n",
        "train_ds = datasets.STL10(path2data, split='train', download=True, transform=transforms.ToTensor())\n",
        "val_ds = datasets.STL10(path2data, split='test', download=True, transform=transforms.ToTensor())\n",
        "\n",
        "print(len(train_ds))\n",
        "print(len(val_ds))"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Files already downloaded and verified\n",
            "Files already downloaded and verified\n",
            "5000\n",
            "8000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0lfYwTbuk7Hs"
      },
      "source": [
        "# define transformation\n",
        "transformation = transforms.Compose([\n",
        "                    transforms.ToTensor(),\n",
        "                    transforms.Resize(224)\n",
        "])\n",
        "\n",
        "# apply transformation to dataset\n",
        "train_ds.transform = transformation\n",
        "val_ds.transform = transformation\n",
        "\n",
        "# make dataloade\n",
        "train_dl = DataLoader(train_ds, batch_size=32, shuffle=True)\n",
        "val_dl = DataLoader(val_ds, batch_size=32, shuffle=True)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "B9ts3Z3xlCQ_",
        "outputId": "13d678de-df4f-4e12-9524-681b0344cbd7"
      },
      "source": [
        "# check sample images\n",
        "def show(img, y=None):\n",
        "    npimg = img.numpy()\n",
        "    npimg_tr = np.transpose(npimg, (1, 2, 0))\n",
        "    plt.imshow(npimg_tr)\n",
        "\n",
        "    if y is not None:\n",
        "        plt.title('labels:' + str(y))\n",
        "\n",
        "np.random.seed(10)\n",
        "torch.manual_seed(0)\n",
        "\n",
        "grid_size=4\n",
        "rnd_ind = np.random.randint(0, len(train_ds), grid_size)\n",
        "\n",
        "x_grid = [train_ds[i][0] for i in rnd_ind]\n",
        "y_grid = [val_ds[i][1] for i in rnd_ind]\n",
        "\n",
        "x_grid = utils.make_grid(x_grid, nrow=grid_size, padding=2)\n",
        "plt.figure(figsize=(10,10))\n",
        "show(x_grid, y_grid)"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N8EnmfFg1DEc"
      },
      "source": [
        "# 2. 모델 구축하기\n",
        "- 코드는 https://github.com/weiaicunzai/pytorch-cifar100/blob/master/models/resnext.py 를 참고했습니다."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P6awtAm-1Eed"
      },
      "source": [
        "# 블락수, gropu수\n",
        "\n",
        "\n",
        "\n",
        "class BottleNeck(nn.Module):\n",
        "    expansion = 4\n",
        "    Cardinality = 32 # group 수\n",
        "    Basewidth = 64 # bottleneck 채널이 64이면 group convolution의 채널은 depth가 됩니다.\n",
        "    Depth = 4 # basewidth일 때, group convolution의 채널 수\n",
        "    def __init__(self, in_channels, out_channels, stride=1):\n",
        "        super().__init__()\n",
        "        C = BottleNeck.Cardinality\n",
        "        D = int(BottleNeck.Depth * out_channels / BottleNeck.Basewidth)\n",
        "\n",
        "        self.conv_residual = nn.Sequential(\n",
        "            nn.Conv2d(in_channels, C * D, 1, stride=1, padding=0, bias=False),\n",
        "            nn.BatchNorm2d(C*D),\n",
        "            nn.ReLU(),\n",
        "            nn.Conv2d(C*D, C*D, 3, stride=stride, padding=1, groups=BottleNeck.Cardinality, bias=False),\n",
        "            nn.BatchNorm2d(C*D),\n",
        "            nn.ReLU(),\n",
        "            nn.Conv2d(C*D, out_channels * BottleNeck.expansion, 1, stride=1, padding=0, bias=False),\n",
        "            nn.BatchNorm2d(out_channels * BottleNeck.expansion)\n",
        "        )\n",
        "\n",
        "        self.conv_shortcut = nn.Sequential()\n",
        "\n",
        "        if stride != 1 or in_channels != out_channels * BottleNeck.expansion:\n",
        "            self.conv_shortcut = nn.Conv2d(in_channels, out_channels * BottleNeck.expansion, 1, stride=stride, padding=0)\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.conv_residual(x) + self.conv_shortcut(x)\n",
        "        return x\n",
        "\n",
        "\n",
        "# ResNext\n",
        "class ResNext(nn.Module):\n",
        "    def __init__(self, nblocks, num_classes=10, init_weights=True):\n",
        "        super().__init__()\n",
        "        self.init_weights=init_weights\n",
        "        self.in_channels = 64\n",
        "\n",
        "        self.conv1 = nn.Sequential(\n",
        "            nn.Conv2d(3, 64, 7, stride=2, padding=2, bias=False),\n",
        "            nn.BatchNorm2d(64),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(3, stride=2, padding=1)\n",
        "        )\n",
        "\n",
        "        self.conv2 = self._make_res_block(nblocks[0], 64, 1)\n",
        "        self.conv3 = self._make_res_block(nblocks[1], 128, 2)\n",
        "        self.conv4 = self._make_res_block(nblocks[2], 256, 2)\n",
        "        self.conv5 = self._make_res_block(nblocks[3], 512, 2)\n",
        "\n",
        "        self.avg_pool = nn.AdaptiveAvgPool2d((1,1))\n",
        "        self.linear = nn.Linear(512 * BottleNeck.expansion, num_classes)\n",
        "\n",
        "        # weights initialization\n",
        "        if self.init_weights:\n",
        "            self._initialize_weights()\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.conv1(x)\n",
        "        x = self.conv2(x)\n",
        "        x = self.conv3(x)\n",
        "        x = self.conv4(x)\n",
        "        x = self.conv5(x)\n",
        "        x = self.avg_pool(x)\n",
        "        x = x.view(x.size(0), -1)\n",
        "        x = self.linear(x)\n",
        "        return x\n",
        "\n",
        "    def _make_res_block(self, nblock, out_channels, stride):\n",
        "        strides = [stride] + [1] * (nblock-1)\n",
        "        res_block = nn.Sequential()\n",
        "        for i, stride in enumerate(strides):\n",
        "            res_block.add_module('dens_layer_{}'.format(i), BottleNeck(self.in_channels, out_channels, stride))\n",
        "            self.in_channels = out_channels * BottleNeck.expansion\n",
        "        return res_block\n",
        "\n",
        "    # weights initialization function\n",
        "    def _initialize_weights(self):\n",
        "        for m in self.modules():\n",
        "            if isinstance(m, nn.Conv2d):\n",
        "                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')\n",
        "                if m.bias is not None:\n",
        "                    nn.init.constant_(m.bias, 0)\n",
        "            elif isinstance(m, nn.BatchNorm2d):\n",
        "                nn.init.constant_(m.weight, 1)\n",
        "                nn.init.constant_(m.bias, 0)\n",
        "            elif isinstance(m, nn.Linear):\n",
        "                nn.init.normal_(m.weight, 0, 0.01)\n",
        "                nn.init.constant_(m.bias, 0)\n",
        "\n",
        "def ResNext50():\n",
        "    return ResNext([3, 4, 6, 3])"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oYkna86giKlP",
        "outputId": "6387b1a8-b345-4284-d824-56c82f40984b"
      },
      "source": [
        "# check model\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "x = torch.randn((3, 3, 224, 224)).to(device)\n",
        "model = ResNext50().to(device)\n",
        "output = model(x)\n",
        "print('output size: ', output.size())"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "output size:  torch.Size([3, 10])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6LXYxWJnkASV",
        "outputId": "2446cc59-11af-4527-8b45-b000222ef792"
      },
      "source": [
        "# print summary\n",
        "summary(model, (3, 224, 224), device=device.type)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "----------------------------------------------------------------\n",
            "        Layer (type)               Output Shape         Param #\n",
            "================================================================\n",
            "            Conv2d-1         [-1, 64, 111, 111]           9,408\n",
            "       BatchNorm2d-2         [-1, 64, 111, 111]             128\n",
            "              ReLU-3         [-1, 64, 111, 111]               0\n",
            "         MaxPool2d-4           [-1, 64, 56, 56]               0\n",
            "            Conv2d-5          [-1, 128, 56, 56]           8,192\n",
            "       BatchNorm2d-6          [-1, 128, 56, 56]             256\n",
            "              ReLU-7          [-1, 128, 56, 56]               0\n",
            "            Conv2d-8          [-1, 128, 56, 56]           4,608\n",
            "       BatchNorm2d-9          [-1, 128, 56, 56]             256\n",
            "             ReLU-10          [-1, 128, 56, 56]               0\n",
            "           Conv2d-11          [-1, 256, 56, 56]          32,768\n",
            "      BatchNorm2d-12          [-1, 256, 56, 56]             512\n",
            "           Conv2d-13          [-1, 256, 56, 56]          16,640\n",
            "       BottleNeck-14          [-1, 256, 56, 56]               0\n",
            "           Conv2d-15          [-1, 128, 56, 56]          32,768\n",
            "      BatchNorm2d-16          [-1, 128, 56, 56]             256\n",
            "             ReLU-17          [-1, 128, 56, 56]               0\n",
            "           Conv2d-18          [-1, 128, 56, 56]           4,608\n",
            "      BatchNorm2d-19          [-1, 128, 56, 56]             256\n",
            "             ReLU-20          [-1, 128, 56, 56]               0\n",
            "           Conv2d-21          [-1, 256, 56, 56]          32,768\n",
            "      BatchNorm2d-22          [-1, 256, 56, 56]             512\n",
            "       BottleNeck-23          [-1, 256, 56, 56]               0\n",
            "           Conv2d-24          [-1, 128, 56, 56]          32,768\n",
            "      BatchNorm2d-25          [-1, 128, 56, 56]             256\n",
            "             ReLU-26          [-1, 128, 56, 56]               0\n",
            "           Conv2d-27          [-1, 128, 56, 56]           4,608\n",
            "      BatchNorm2d-28          [-1, 128, 56, 56]             256\n",
            "             ReLU-29          [-1, 128, 56, 56]               0\n",
            "           Conv2d-30          [-1, 256, 56, 56]          32,768\n",
            "      BatchNorm2d-31          [-1, 256, 56, 56]             512\n",
            "       BottleNeck-32          [-1, 256, 56, 56]               0\n",
            "           Conv2d-33          [-1, 256, 56, 56]          65,536\n",
            "      BatchNorm2d-34          [-1, 256, 56, 56]             512\n",
            "             ReLU-35          [-1, 256, 56, 56]               0\n",
            "           Conv2d-36          [-1, 256, 28, 28]          18,432\n",
            "      BatchNorm2d-37          [-1, 256, 28, 28]             512\n",
            "             ReLU-38          [-1, 256, 28, 28]               0\n",
            "           Conv2d-39          [-1, 512, 28, 28]         131,072\n",
            "      BatchNorm2d-40          [-1, 512, 28, 28]           1,024\n",
            "           Conv2d-41          [-1, 512, 28, 28]         131,584\n",
            "       BottleNeck-42          [-1, 512, 28, 28]               0\n",
            "           Conv2d-43          [-1, 256, 28, 28]         131,072\n",
            "      BatchNorm2d-44          [-1, 256, 28, 28]             512\n",
            "             ReLU-45          [-1, 256, 28, 28]               0\n",
            "           Conv2d-46          [-1, 256, 28, 28]          18,432\n",
            "      BatchNorm2d-47          [-1, 256, 28, 28]             512\n",
            "             ReLU-48          [-1, 256, 28, 28]               0\n",
            "           Conv2d-49          [-1, 512, 28, 28]         131,072\n",
            "      BatchNorm2d-50          [-1, 512, 28, 28]           1,024\n",
            "       BottleNeck-51          [-1, 512, 28, 28]               0\n",
            "           Conv2d-52          [-1, 256, 28, 28]         131,072\n",
            "      BatchNorm2d-53          [-1, 256, 28, 28]             512\n",
            "             ReLU-54          [-1, 256, 28, 28]               0\n",
            "           Conv2d-55          [-1, 256, 28, 28]          18,432\n",
            "      BatchNorm2d-56          [-1, 256, 28, 28]             512\n",
            "             ReLU-57          [-1, 256, 28, 28]               0\n",
            "           Conv2d-58          [-1, 512, 28, 28]         131,072\n",
            "      BatchNorm2d-59          [-1, 512, 28, 28]           1,024\n",
            "       BottleNeck-60          [-1, 512, 28, 28]               0\n",
            "           Conv2d-61          [-1, 256, 28, 28]         131,072\n",
            "      BatchNorm2d-62          [-1, 256, 28, 28]             512\n",
            "             ReLU-63          [-1, 256, 28, 28]               0\n",
            "           Conv2d-64          [-1, 256, 28, 28]          18,432\n",
            "      BatchNorm2d-65          [-1, 256, 28, 28]             512\n",
            "             ReLU-66          [-1, 256, 28, 28]               0\n",
            "           Conv2d-67          [-1, 512, 28, 28]         131,072\n",
            "      BatchNorm2d-68          [-1, 512, 28, 28]           1,024\n",
            "       BottleNeck-69          [-1, 512, 28, 28]               0\n",
            "           Conv2d-70          [-1, 512, 28, 28]         262,144\n",
            "      BatchNorm2d-71          [-1, 512, 28, 28]           1,024\n",
            "             ReLU-72          [-1, 512, 28, 28]               0\n",
            "           Conv2d-73          [-1, 512, 14, 14]          73,728\n",
            "      BatchNorm2d-74          [-1, 512, 14, 14]           1,024\n",
            "             ReLU-75          [-1, 512, 14, 14]               0\n",
            "           Conv2d-76         [-1, 1024, 14, 14]         524,288\n",
            "      BatchNorm2d-77         [-1, 1024, 14, 14]           2,048\n",
            "           Conv2d-78         [-1, 1024, 14, 14]         525,312\n",
            "       BottleNeck-79         [-1, 1024, 14, 14]               0\n",
            "           Conv2d-80          [-1, 512, 14, 14]         524,288\n",
            "      BatchNorm2d-81          [-1, 512, 14, 14]           1,024\n",
            "             ReLU-82          [-1, 512, 14, 14]               0\n",
            "           Conv2d-83          [-1, 512, 14, 14]          73,728\n",
            "      BatchNorm2d-84          [-1, 512, 14, 14]           1,024\n",
            "             ReLU-85          [-1, 512, 14, 14]               0\n",
            "           Conv2d-86         [-1, 1024, 14, 14]         524,288\n",
            "      BatchNorm2d-87         [-1, 1024, 14, 14]           2,048\n",
            "       BottleNeck-88         [-1, 1024, 14, 14]               0\n",
            "           Conv2d-89          [-1, 512, 14, 14]         524,288\n",
            "      BatchNorm2d-90          [-1, 512, 14, 14]           1,024\n",
            "             ReLU-91          [-1, 512, 14, 14]               0\n",
            "           Conv2d-92          [-1, 512, 14, 14]          73,728\n",
            "      BatchNorm2d-93          [-1, 512, 14, 14]           1,024\n",
            "             ReLU-94          [-1, 512, 14, 14]               0\n",
            "           Conv2d-95         [-1, 1024, 14, 14]         524,288\n",
            "      BatchNorm2d-96         [-1, 1024, 14, 14]           2,048\n",
            "       BottleNeck-97         [-1, 1024, 14, 14]               0\n",
            "           Conv2d-98          [-1, 512, 14, 14]         524,288\n",
            "      BatchNorm2d-99          [-1, 512, 14, 14]           1,024\n",
            "            ReLU-100          [-1, 512, 14, 14]               0\n",
            "          Conv2d-101          [-1, 512, 14, 14]          73,728\n",
            "     BatchNorm2d-102          [-1, 512, 14, 14]           1,024\n",
            "            ReLU-103          [-1, 512, 14, 14]               0\n",
            "          Conv2d-104         [-1, 1024, 14, 14]         524,288\n",
            "     BatchNorm2d-105         [-1, 1024, 14, 14]           2,048\n",
            "      BottleNeck-106         [-1, 1024, 14, 14]               0\n",
            "          Conv2d-107          [-1, 512, 14, 14]         524,288\n",
            "     BatchNorm2d-108          [-1, 512, 14, 14]           1,024\n",
            "            ReLU-109          [-1, 512, 14, 14]               0\n",
            "          Conv2d-110          [-1, 512, 14, 14]          73,728\n",
            "     BatchNorm2d-111          [-1, 512, 14, 14]           1,024\n",
            "            ReLU-112          [-1, 512, 14, 14]               0\n",
            "          Conv2d-113         [-1, 1024, 14, 14]         524,288\n",
            "     BatchNorm2d-114         [-1, 1024, 14, 14]           2,048\n",
            "      BottleNeck-115         [-1, 1024, 14, 14]               0\n",
            "          Conv2d-116          [-1, 512, 14, 14]         524,288\n",
            "     BatchNorm2d-117          [-1, 512, 14, 14]           1,024\n",
            "            ReLU-118          [-1, 512, 14, 14]               0\n",
            "          Conv2d-119          [-1, 512, 14, 14]          73,728\n",
            "     BatchNorm2d-120          [-1, 512, 14, 14]           1,024\n",
            "            ReLU-121          [-1, 512, 14, 14]               0\n",
            "          Conv2d-122         [-1, 1024, 14, 14]         524,288\n",
            "     BatchNorm2d-123         [-1, 1024, 14, 14]           2,048\n",
            "      BottleNeck-124         [-1, 1024, 14, 14]               0\n",
            "          Conv2d-125         [-1, 1024, 14, 14]       1,048,576\n",
            "     BatchNorm2d-126         [-1, 1024, 14, 14]           2,048\n",
            "            ReLU-127         [-1, 1024, 14, 14]               0\n",
            "          Conv2d-128           [-1, 1024, 7, 7]         294,912\n",
            "     BatchNorm2d-129           [-1, 1024, 7, 7]           2,048\n",
            "            ReLU-130           [-1, 1024, 7, 7]               0\n",
            "          Conv2d-131           [-1, 2048, 7, 7]       2,097,152\n",
            "     BatchNorm2d-132           [-1, 2048, 7, 7]           4,096\n",
            "          Conv2d-133           [-1, 2048, 7, 7]       2,099,200\n",
            "      BottleNeck-134           [-1, 2048, 7, 7]               0\n",
            "          Conv2d-135           [-1, 1024, 7, 7]       2,097,152\n",
            "     BatchNorm2d-136           [-1, 1024, 7, 7]           2,048\n",
            "            ReLU-137           [-1, 1024, 7, 7]               0\n",
            "          Conv2d-138           [-1, 1024, 7, 7]         294,912\n",
            "     BatchNorm2d-139           [-1, 1024, 7, 7]           2,048\n",
            "            ReLU-140           [-1, 1024, 7, 7]               0\n",
            "          Conv2d-141           [-1, 2048, 7, 7]       2,097,152\n",
            "     BatchNorm2d-142           [-1, 2048, 7, 7]           4,096\n",
            "      BottleNeck-143           [-1, 2048, 7, 7]               0\n",
            "          Conv2d-144           [-1, 1024, 7, 7]       2,097,152\n",
            "     BatchNorm2d-145           [-1, 1024, 7, 7]           2,048\n",
            "            ReLU-146           [-1, 1024, 7, 7]               0\n",
            "          Conv2d-147           [-1, 1024, 7, 7]         294,912\n",
            "     BatchNorm2d-148           [-1, 1024, 7, 7]           2,048\n",
            "            ReLU-149           [-1, 1024, 7, 7]               0\n",
            "          Conv2d-150           [-1, 2048, 7, 7]       2,097,152\n",
            "     BatchNorm2d-151           [-1, 2048, 7, 7]           4,096\n",
            "      BottleNeck-152           [-1, 2048, 7, 7]               0\n",
            "AdaptiveAvgPool2d-153           [-1, 2048, 1, 1]               0\n",
            "          Linear-154                   [-1, 10]          20,490\n",
            "================================================================\n",
            "Total params: 22,996,554\n",
            "Trainable params: 22,996,554\n",
            "Non-trainable params: 0\n",
            "----------------------------------------------------------------\n",
            "Input size (MB): 0.57\n",
            "Forward/backward pass size (MB): 307.85\n",
            "Params size (MB): 87.72\n",
            "Estimated Total Size (MB): 396.15\n",
            "----------------------------------------------------------------\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SZC9sU4j1FXI"
      },
      "source": [
        "# 3. 학습하기"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wuX07Z3ZkxNG"
      },
      "source": [
        "# define loss function, optimizer, lr_scheduler\n",
        "loss_func = nn.CrossEntropyLoss(reduction='sum')\n",
        "opt = optim.Adam(model.parameters(), lr=0.01)\n",
        "\n",
        "from torch.optim.lr_scheduler import ReduceLROnPlateau\n",
        "lr_scheduler = ReduceLROnPlateau(opt, mode='min', factor=0.1, patience=10)\n",
        "\n",
        "\n",
        "# get current lr\n",
        "def get_lr(opt):\n",
        "    for param_group in opt.param_groups:\n",
        "        return param_group['lr']\n",
        "\n",
        "\n",
        "# calculate the metric per mini-batch\n",
        "def metric_batch(output, target):\n",
        "    pred = output.argmax(1, keepdim=True)\n",
        "    corrects = pred.eq(target.view_as(pred)).sum().item()\n",
        "    return corrects\n",
        "\n",
        "\n",
        "# calculate the loss per mini-batch\n",
        "def loss_batch(loss_func, output, target, opt=None):\n",
        "    loss_b = loss_func(output, target)\n",
        "    metric_b = metric_batch(output, target)\n",
        "\n",
        "    if opt is not None:\n",
        "        opt.zero_grad()\n",
        "        loss_b.backward()\n",
        "        opt.step()\n",
        "    \n",
        "    return loss_b.item(), metric_b\n",
        "\n",
        "\n",
        "# calculate the loss per epochs\n",
        "def loss_epoch(model, loss_func, dataset_dl, sanity_check=False, opt=None):\n",
        "    running_loss = 0.0\n",
        "    running_metric = 0.0\n",
        "    len_data = len(dataset_dl.dataset)\n",
        "\n",
        "    for xb, yb in dataset_dl:\n",
        "        xb = xb.to(device)\n",
        "        yb = yb.to(device)\n",
        "        output = model(xb)\n",
        "\n",
        "        loss_b, metric_b = loss_batch(loss_func, output, yb, opt)\n",
        "\n",
        "        running_loss += loss_b\n",
        "        \n",
        "        if metric_b is not None:\n",
        "            running_metric += metric_b\n",
        "\n",
        "        if sanity_check is True:\n",
        "            break\n",
        "\n",
        "    loss = running_loss / len_data\n",
        "    metric = running_metric / len_data\n",
        "    return loss, metric\n",
        "\n",
        "\n",
        "# function to start training\n",
        "def train_val(model, params):\n",
        "    num_epochs=params['num_epochs']\n",
        "    loss_func=params['loss_func']\n",
        "    opt=params['optimizer']\n",
        "    train_dl=params['train_dl']\n",
        "    val_dl=params['val_dl']\n",
        "    sanity_check=params['sanity_check']\n",
        "    lr_scheduler=params['lr_scheduler']\n",
        "    path2weights=params['path2weights']\n",
        "\n",
        "    loss_history = {'train': [], 'val': []}\n",
        "    metric_history = {'train': [], 'val': []}\n",
        "\n",
        "    best_loss = float('inf')\n",
        "    best_model_wts = copy.deepcopy(model.state_dict())\n",
        "    start_time = time.time()\n",
        "\n",
        "    for epoch in range(num_epochs):\n",
        "        current_lr = get_lr(opt)\n",
        "        print('Epoch {}/{}, current lr= {}'.format(epoch, num_epochs-1, current_lr))\n",
        "\n",
        "        model.train()\n",
        "        train_loss, train_metric = loss_epoch(model, loss_func, train_dl, sanity_check, opt)\n",
        "        loss_history['train'].append(train_loss)\n",
        "        metric_history['train'].append(train_metric)\n",
        "\n",
        "        model.eval()\n",
        "        with torch.no_grad():\n",
        "            val_loss, val_metric = loss_epoch(model, loss_func, val_dl, sanity_check)\n",
        "        loss_history['val'].append(val_loss)\n",
        "        metric_history['val'].append(val_metric)\n",
        "\n",
        "        if val_loss < best_loss:\n",
        "            best_loss = val_loss\n",
        "            best_model_wts = copy.deepcopy(model.state_dict())\n",
        "            torch.save(model.state_dict(), path2weights)\n",
        "            print('Copied best model weights!')\n",
        "\n",
        "        lr_scheduler.step(val_loss)\n",
        "        if current_lr != get_lr(opt):\n",
        "            print('Loading best model weights!')\n",
        "            model.load_state_dict(best_model_wts)\n",
        "\n",
        "        print('train loss: %.6f, val loss: %.6f, accuracy: %.2f, time: %.4f min' %(train_loss, val_loss, 100*val_metric, (time.time()-start_time)/60))\n",
        "        print('-'*10)\n",
        "\n",
        "    model.load_state_dict(best_model_wts)\n",
        "    return model, loss_history, metric_history"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ShUIfqzzlHc7"
      },
      "source": [
        "# define the training parameters\n",
        "params_train = {\n",
        "    'num_epochs':30,\n",
        "    'optimizer':opt,\n",
        "    'loss_func':loss_func,\n",
        "    'train_dl':train_dl,\n",
        "    'val_dl':val_dl,\n",
        "    'sanity_check':False,\n",
        "    'lr_scheduler':lr_scheduler,\n",
        "    'path2weights':'./models/weights.pt',\n",
        "}\n",
        "\n",
        "# check the directory to save weights.pt\n",
        "def createFolder(directory):\n",
        "    try:\n",
        "        if not os.path.exists(directory):\n",
        "            os.makedirs(directory)\n",
        "    except OSerror:\n",
        "        print('Error')\n",
        "createFolder('./models')"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3kwG7sJBlJHE",
        "outputId": "4fae7107-150f-46e0-b902-da37619987f8"
      },
      "source": [
        "model, loss_hist, metric_hist = train_val(model, params_train)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 0/29, current lr= 0.01\n",
            "Copied best model weights!\n",
            "train loss: 48.434045, val loss: 1.964803, accuracy: 28.78, time: 1.9040 min\n",
            "----------\n",
            "Epoch 1/29, current lr= 0.01\n",
            "Copied best model weights!\n",
            "train loss: 1.852148, val loss: 1.817145, accuracy: 31.01, time: 3.8396 min\n",
            "----------\n",
            "Epoch 2/29, current lr= 0.01\n",
            "Copied best model weights!\n",
            "train loss: 1.747134, val loss: 1.813179, accuracy: 33.17, time: 5.7742 min\n",
            "----------\n",
            "Epoch 3/29, current lr= 0.01\n",
            "Copied best model weights!\n",
            "train loss: 1.717768, val loss: 1.663561, accuracy: 36.41, time: 7.7070 min\n",
            "----------\n",
            "Epoch 4/29, current lr= 0.01\n",
            "train loss: 1.639641, val loss: 1.739531, accuracy: 37.51, time: 9.6359 min\n",
            "----------\n",
            "Epoch 5/29, current lr= 0.01\n",
            "train loss: 1.659204, val loss: 1.730320, accuracy: 36.65, time: 11.5660 min\n",
            "----------\n",
            "Epoch 6/29, current lr= 0.01\n",
            "Copied best model weights!\n",
            "train loss: 1.524757, val loss: 1.567042, accuracy: 41.56, time: 13.4991 min\n",
            "----------\n",
            "Epoch 7/29, current lr= 0.01\n",
            "train loss: 1.484428, val loss: 1.674726, accuracy: 40.25, time: 15.4286 min\n",
            "----------\n",
            "Epoch 8/29, current lr= 0.01\n",
            "train loss: 1.459734, val loss: 2.067420, accuracy: 35.09, time: 17.3608 min\n",
            "----------\n",
            "Epoch 9/29, current lr= 0.01\n",
            "train loss: 1.460582, val loss: 1.580220, accuracy: 44.70, time: 19.2949 min\n",
            "----------\n",
            "Epoch 10/29, current lr= 0.01\n",
            "train loss: 1.419178, val loss: 1.708644, accuracy: 42.38, time: 21.2277 min\n",
            "----------\n",
            "Epoch 11/29, current lr= 0.01\n",
            "train loss: 1.449635, val loss: 1.645457, accuracy: 43.54, time: 23.1614 min\n",
            "----------\n",
            "Epoch 12/29, current lr= 0.01\n",
            "train loss: 1.459456, val loss: 1.814912, accuracy: 42.60, time: 25.0987 min\n",
            "----------\n",
            "Epoch 13/29, current lr= 0.01\n",
            "train loss: 1.372052, val loss: 1.890350, accuracy: 40.95, time: 27.0301 min\n",
            "----------\n",
            "Epoch 14/29, current lr= 0.01\n",
            "Copied best model weights!\n",
            "train loss: 1.431225, val loss: 1.504633, accuracy: 47.62, time: 28.9603 min\n",
            "----------\n",
            "Epoch 15/29, current lr= 0.01\n",
            "train loss: 1.339471, val loss: 2.197080, accuracy: 36.50, time: 30.8943 min\n",
            "----------\n",
            "Epoch 16/29, current lr= 0.01\n",
            "train loss: 1.507124, val loss: 2.461717, accuracy: 31.99, time: 32.8255 min\n",
            "----------\n",
            "Epoch 17/29, current lr= 0.01\n",
            "train loss: 1.687618, val loss: 1.987158, accuracy: 39.15, time: 34.7577 min\n",
            "----------\n",
            "Epoch 18/29, current lr= 0.01\n",
            "train loss: 1.451438, val loss: 1.757191, accuracy: 43.51, time: 36.6888 min\n",
            "----------\n",
            "Epoch 19/29, current lr= 0.01\n",
            "train loss: 1.603508, val loss: 1.911977, accuracy: 44.44, time: 38.6183 min\n",
            "----------\n",
            "Epoch 20/29, current lr= 0.01\n",
            "train loss: 1.582411, val loss: 2.303870, accuracy: 39.17, time: 40.5500 min\n",
            "----------\n",
            "Epoch 21/29, current lr= 0.01\n",
            "train loss: 1.574459, val loss: 2.126603, accuracy: 42.35, time: 42.4840 min\n",
            "----------\n",
            "Epoch 22/29, current lr= 0.01\n",
            "train loss: 1.665131, val loss: 2.896331, accuracy: 31.51, time: 44.4189 min\n",
            "----------\n",
            "Epoch 23/29, current lr= 0.01\n",
            "train loss: 10361.830338, val loss: 497621.238500, accuracy: 9.75, time: 46.3475 min\n",
            "----------\n",
            "Epoch 24/29, current lr= 0.01\n",
            "train loss: 54733.165933, val loss: 401.387620, accuracy: 13.66, time: 48.2378 min\n",
            "----------\n",
            "Epoch 25/29, current lr= 0.01\n",
            "Loading best model weights!\n",
            "train loss: 264.527253, val loss: 98.857951, accuracy: 15.20, time: 50.1317 min\n",
            "----------\n",
            "Epoch 26/29, current lr= 0.001\n",
            "Copied best model weights!\n",
            "train loss: 1.023910, val loss: 1.288899, accuracy: 53.49, time: 52.0717 min\n",
            "----------\n",
            "Epoch 27/29, current lr= 0.001\n",
            "Copied best model weights!\n",
            "train loss: 0.861088, val loss: 1.233332, accuracy: 55.43, time: 54.0089 min\n",
            "----------\n",
            "Epoch 28/29, current lr= 0.001\n",
            "Copied best model weights!\n",
            "train loss: 0.789872, val loss: 1.216383, accuracy: 56.69, time: 55.9409 min\n",
            "----------\n",
            "Epoch 29/29, current lr= 0.001\n",
            "Copied best model weights!\n",
            "train loss: 0.734605, val loss: 1.196059, accuracy: 57.59, time: 57.8748 min\n",
            "----------\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "id": "LX2qm7kGlKgn",
        "outputId": "cc90526c-102d-4eb6-e715-1e1439ec2c69"
      },
      "source": [
        "# train-val progress\n",
        "num_epochs = params_train['num_epochs']\n",
        "\n",
        "# plot loss progress\n",
        "plt.title('Train-Val Loss')\n",
        "plt.plot(range(1, num_epochs+1), loss_hist['train'], label='train')\n",
        "plt.plot(range(1, num_epochs+1), loss_hist['val'], label='val')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Training Epochs')\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "# plot accuracy progress\n",
        "plt.title('Train-Val Accuracy')\n",
        "plt.plot(range(1, num_epochs+1), metric_hist['train'], label='train')\n",
        "plt.plot(range(1, num_epochs+1), metric_hist['val'], label='val')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Training Epochs')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JObmI2hllLrT"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}