GAPのサンプル

GAPModelSample.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "GAPModelSample.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "lnieGyUccbs-",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        },
        "outputId": "2c3f054b-3ab6-433a-c3e6-25e011f77854"
      },
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "from torchsummary import summary\n",
        "\n",
        "class Net(nn.Module):\n",
        "    def __init__(self):\n",
        "        super(Net, self).__init__()\n",
        "        self.conv1 = nn.Conv2d(3, 16, 5)\n",
        "        self.conv2 = nn.Conv2d(16, 32, 5)\n",
        "        self.conv3 = nn.Conv2d(32, 64, 5)\n",
        "        self.conv4 = nn.Conv2d(64, 10, 5)\n",
        "        self.avgpool = torch.nn.AdaptiveAvgPool2d((1,1))\n",
        "    def forward(self, x):\n",
        "        x = F.relu(self.conv1(x))\n",
        "        x = F.relu(self.conv2(x))\n",
        "        x = F.relu(self.conv3(x))\n",
        "        x = F.relu(self.conv4(x))\n",
        "        #GAP\n",
        "        x = self.avgpool(x)\n",
        "        x = torch.flatten(x, 1)\n",
        "        return x\n",
        "\n",
        "summary(Net(), (3, 32, 32))"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "----------------------------------------------------------------\n",
            "        Layer (type)               Output Shape         Param #\n",
            "================================================================\n",
            "            Conv2d-1           [-1, 16, 28, 28]           1,216\n",
            "            Conv2d-2           [-1, 32, 24, 24]          12,832\n",
            "            Conv2d-3           [-1, 64, 20, 20]          51,264\n",
            "            Conv2d-4           [-1, 10, 16, 16]          16,010\n",
            " AdaptiveAvgPool2d-5             [-1, 10, 1, 1]               0\n",
            "================================================================\n",
            "Total params: 81,322\n",
            "Trainable params: 81,322\n",
            "Non-trainable params: 0\n",
            "----------------------------------------------------------------\n",
            "Input size (MB): 0.01\n",
            "Forward/backward pass size (MB): 0.45\n",
            "Params size (MB): 0.31\n",
            "Estimated Total Size (MB): 0.77\n",
            "----------------------------------------------------------------\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}