cnn_example.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOpo7H4Q4ZT8TjqG8irT8wl",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jsgovea/PyTorch-learning/blob/main/cnn_example.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "SB6rLS2zebqM"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "\n",
        "from torch.utils.data import DataLoader, sampler\n",
        "import torchvision.datasets as datasets\n",
        "import torchvision.transforms as T\n",
        "import matplotlib.pyplot as plt\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "DATA_PATH = '/content/cifar-10'\n",
        "NUM_TRAIN = 50000\n",
        "NUM_VAL = 5000\n",
        "NUM_TEST = 5000\n",
        "MINBATCH_SIZE = 64\n",
        "\n",
        "transform_cifar = T.Compose([\n",
        "    T.ToTensor(),\n",
        "    T.Normalize([0.491, 0.482, 0.447], [0.247, 0.243, 0.261])\n",
        "])"
      ],
      "metadata": {
        "id": "CfDGveWKfuIi"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train dataset\n",
        "\n",
        "# Download the dataset\n",
        "cifar10_train = datasets.CIFAR10(DATA_PATH, train=True, download=True,\n",
        "                                 transform=transform_cifar)\n",
        "\n",
        "train_loader = DataLoader(cifar10_train, batch_size=MINBATCH_SIZE,\n",
        "                          sampler=sampler.SubsetRandomSampler(range(NUM_TRAIN)))\n",
        "\n",
        "cifar10_val = datasets.CIFAR10(DATA_PATH, train=False, download=True,\n",
        "                              transform=transform_cifar)\n",
        "\n",
        "val_loader = DataLoader(cifar10_val, batch_size=MINBATCH_SIZE,\n",
        "                        sampler=sampler.SubsetRandomSampler(range(NUM_VAL)))\n",
        "\n",
        "cifar10_test = datasets.CIFAR10(DATA_PATH, train=False, download=True,\n",
        "                                transform=transform_cifar)\n",
        "\n",
        "test_loader = DataLoader(cifar10_test, batch_size=MINBATCH_SIZE,\n",
        "                         sampler=sampler.SubsetRandomSampler(range(NUM_VAL, len(cifar10_test))))\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5DJGX0PDgrJA",
        "outputId": "c430d708-e212-4b01-aa53-df35d464aedb"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to /content/cifar-10/cifar-10-python.tar.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 170498071/170498071 [00:02<00:00, 67732191.14it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting /content/cifar-10/cifar-10-python.tar.gz to /content/cifar-10\n",
            "Files already downloaded and verified\n",
            "Files already downloaded and verified\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_loader.dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oRxFhamwgrGb",
        "outputId": "3287a7b3-966d-40b8-a2ab-de37e860d606"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Dataset CIFAR10\n",
              "    Number of datapoints: 10000\n",
              "    Root location: /content/cifar-10\n",
              "    Split: Test\n",
              "    StandardTransform\n",
              "Transform: Compose(\n",
              "               ToTensor()\n",
              "               Normalize(mean=[0.491, 0.482, 0.447], std=[0.247, 0.243, 0.261])\n",
              "           )"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Check the GPUs"
      ],
      "metadata": {
        "id": "mX1NkQ9T0Nc7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "if torch.cuda.is_available():\n",
        "  device = torch.device('cuda')\n",
        "else:\n",
        "  device = torch.device('cpu')\n",
        "print(device)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LcpAYX6ezRLA",
        "outputId": "983c7440-b7a5-4df4-9466-4fa2b8141081"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "cpu\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Show images"
      ],
      "metadata": {
        "id": "sueW9v9M0TFo"
      }
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "2KHL7xPaKV9p"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "classes = test_loader.dataset.classes\n",
        "def plot_figure(image):\n",
        "  plt.imshow(np.transpose(image, (1, 2, 0)))\n",
        "  plt.axis('off')\n",
        "  plt.show()\n",
        "\n",
        "rnd_sample_idx = np.random.randint(len(test_loader))\n",
        "print(f\"The current image represents a: {classes[test_loader.dataset[rnd_sample_idx][1]]}\")\n",
        "image = test_loader.dataset[rnd_sample_idx][0]\n",
        "image = (image - image.min()) / (image.max() - image.min())\n",
        "plot_figure(image)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "lAh0VMk-0FZF",
        "outputId": "041b1d00-243b-43b7-a7ad-354b99145d13"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The current image represents a: horse\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_cifar10_grid():\n",
        "  classes = test_loader.dataset.classes\n",
        "  total_samples = 8\n",
        "  plt.figure(figsize=(15, 15))\n",
        "  for label, sample in enumerate(classes):\n",
        "    class_idxs = np.flatnonzero(label == np.array(test_loader.dataset.targets))\n",
        "    sample_idxs = np.random.choice(class_idxs, total_samples, replace=False)\n",
        "    for i, idx in enumerate(sample_idxs):\n",
        "      plt_idx = i * len(classes) * label + 1\n",
        "      plt.subplot(total_samples, len(classes), plt_idx)\n",
        "      plt.imshow(test_loader.dataset.data[idx])\n",
        "      plt.axis('off')\n",
        "\n",
        "      if i == 0: plt.title(sample)\n",
        "  plt.show()\n",
        "plot_cifar10_grid\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DtC26wDDcgpE",
        "outputId": "3df66366-6d72-4484-d54e-0e4840da8080"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<function __main__.plot_cifar10_grid()>"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Calculate accuracy"
      ],
      "metadata": {
        "id": "JaAFXO_P0V_j"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "HvYALS_k0W7d"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "fE-sb4Ac0Wll"
      },
      "execution_count": 8,
      "outputs": []
    }
  ]
}