036_Image_Augment.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "036.Image_Augment.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm",
      "authorship_tag": "ABX9TyMZp4bwDmM1grTQyN+mSifn",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "bbf7fafe684a4667b7c8a0807cc0de08": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_6bf05e38c0e4494e9b94a765b85d6cd6",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_3e071b02ba4e40fdbe0b32f9465eb50d",
              "IPY_MODEL_0e263f760b834d83a5bdf5ad830cc426",
              "IPY_MODEL_4e1aad8d11aa445d95df61c204c93c83"
            ]
          }
        },
        "6bf05e38c0e4494e9b94a765b85d6cd6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "3e071b02ba4e40fdbe0b32f9465eb50d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_4d2dc739c35d4e2b90b1daf39b4b8602",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_044442ec296a4b41acba171c5d5c1289"
          }
        },
        "0e263f760b834d83a5bdf5ad830cc426": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_82bd9e3f4eae4f5f98edc3ef1373f443",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 170498071,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 170498071,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_228c44d043134ab1bb18e0de39b033e8"
          }
        },
        "4e1aad8d11aa445d95df61c204c93c83": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_309a58e0aee74e05b54090fe0cc8f43d",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 170499072/? [00:10&lt;00:00, 17216678.22it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_7739b383791b433f994f13e08d4c940a"
          }
        },
        "4d2dc739c35d4e2b90b1daf39b4b8602": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "044442ec296a4b41acba171c5d5c1289": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "82bd9e3f4eae4f5f98edc3ef1373f443": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "228c44d043134ab1bb18e0de39b033e8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "309a58e0aee74e05b54090fe0cc8f43d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "7739b383791b433f994f13e08d4c940a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/PANDASANG1231/Deeplearning_byHand/blob/main/036_Image_Augment.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xnFXuqV2lQSx",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f8a176cc-3bd5-4699-9103-fb6e259c31d1"
      },
      "source": [
        "import sys\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive', force_remount=True)\n",
        "sys.path.append('/content/drive/MyDrive/Colab Notebooks/deeplearning_note')\n",
        "from tool import *"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "! nvidia-smi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5SemXnflYY1X",
        "outputId": "80a26bbc-4a05-4b96-899c-d98b28742373"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mon Jan 24 04:48:50 2022       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 495.46       Driver Version: 460.32.03    CUDA Version: 11.2     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla P100-PCIE...  Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   33C    P0    25W / 250W |      0MiB / 16280MiB |      0%      Default |\n",
            "|                               |                      |                  N/A |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "w-EGHWAqUwN3"
      },
      "source": [
        "import torch\n",
        "import torchvision"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6ARWWkvhVRF_"
      },
      "source": [
        "## Data Augment"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z9PHA8ckMYtS",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 104,
          "referenced_widgets": [
            "bbf7fafe684a4667b7c8a0807cc0de08",
            "6bf05e38c0e4494e9b94a765b85d6cd6",
            "3e071b02ba4e40fdbe0b32f9465eb50d",
            "0e263f760b834d83a5bdf5ad830cc426",
            "4e1aad8d11aa445d95df61c204c93c83",
            "4d2dc739c35d4e2b90b1daf39b4b8602",
            "044442ec296a4b41acba171c5d5c1289",
            "82bd9e3f4eae4f5f98edc3ef1373f443",
            "228c44d043134ab1bb18e0de39b033e8",
            "309a58e0aee74e05b54090fe0cc8f43d",
            "7739b383791b433f994f13e08d4c940a"
          ]
        },
        "outputId": "076906dc-73e3-4b48-96d4-b13035b35f0f"
      },
      "source": [
        "train_dataset = torchvision.datasets.CIFAR10(root=\"./\",train=True,download=True)\n",
        "test_dataset = torchvision.datasets.CIFAR10(root=\"./\",train=False,download=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./cifar-10-python.tar.gz\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "bbf7fafe684a4667b7c8a0807cc0de08",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "  0%|          | 0/170498071 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./cifar-10-python.tar.gz to ./\n",
            "Files already downloaded and verified\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "29Y7ZkTSYrCj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "357eaef9-3dc6-4c8e-b99a-ca92dd56885d"
      },
      "source": [
        "img_list = [train_dataset[i] for i in range(8)]\n",
        "lab_list = [x[1] for x in img_list]\n",
        "img_list = [x[0] for x in img_list]\n",
        "img_list, lab_list"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "([<PIL.Image.Image image mode=RGB size=32x32 at 0x7F6946979DD0>,\n",
              "  <PIL.Image.Image image mode=RGB size=32x32 at 0x7F6946979F50>,\n",
              "  <PIL.Image.Image image mode=RGB size=32x32 at 0x7F6946979FD0>,\n",
              "  <PIL.Image.Image image mode=RGB size=32x32 at 0x7F694645D090>,\n",
              "  <PIL.Image.Image image mode=RGB size=32x32 at 0x7F694645D110>,\n",
              "  <PIL.Image.Image image mode=RGB size=32x32 at 0x7F694645D1D0>,\n",
              "  <PIL.Image.Image image mode=RGB size=32x32 at 0x7F694645D250>,\n",
              "  <PIL.Image.Image image mode=RGB size=32x32 at 0x7F694645D2D0>],\n",
              " [6, 9, 9, 4, 1, 1, 2, 7])"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5XhYBD0vzt94",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 375
        },
        "outputId": "a0139176-9f75-4b89-b004-09402b05099d"
      },
      "source": [
        "transform_list = [\n",
        "    torchvision.transforms.CenterCrop(18),\n",
        "    torchvision.transforms.GaussianBlur(5),\n",
        "    torchvision.transforms.RandomAdjustSharpness(0),\n",
        "    torchvision.transforms.RandomAdjustSharpness(2),\n",
        "    torchvision.transforms.RandomVerticalFlip(1),\n",
        "    torchvision.transforms.RandomHorizontalFlip(1),\n",
        "    torchvision.transforms.ColorJitter(0.1,0.1,0.1,0.1),\n",
        "    torchvision.transforms.RandomGrayscale(1),\n",
        "    torchvision.transforms.RandomSolarize(0.5, p=1)\n",
        "]\n",
        "\n",
        "fig, axs = plt.subplots(2, 5, figsize=(20, 6))\n",
        "\n",
        "axs[0][0].imshow(img_list[0])\n",
        "\n",
        "for aug, ax in zip(transform_list, axs.flatten()[1:]):\n",
        "    ax.imshow(aug(img_list[0]))\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x432 with 10 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Resnet + Data Augment\n",
        "\n",
        "\n",
        "Because of the extra transforms in training data, training process will be slower and training accuracy is lower. But the gap between training and testing is also smaller"
      ],
      "metadata": {
        "id": "MDarzT3O41nL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_aug = torchvision.transforms.Compose([torchvision.transforms.RandomVerticalFlip(0.5), torchvision.transforms.ToTensor()])\n",
        "test_aug = torchvision.transforms.Compose([torchvision.transforms.ToTensor()])"
      ],
      "metadata": {
        "id": "aLwIFWNGwJMk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### No augment"
      ],
      "metadata": {
        "id": "bazmo924fKWO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_dataset = torchvision.datasets.CIFAR10(root=\"./\", train=True, download=True, transform=test_aug)\n",
        "test_dataset = torchvision.datasets.CIFAR10(root=\"./\", train=False, download=True, transform=test_aug)\n",
        "\n",
        "train_dataloader = torch.utils.data.DataLoader(dataset=train_dataset, batch_size=64, shuffle=True)\n",
        "test_dataloader = torch.utils.data.DataLoader(dataset=test_dataset, batch_size=64, shuffle=False)\n",
        "\n",
        "resnet = torchvision.models.resnet18(pretrained=False)\n",
        "\n",
        "device = torch.device('cuda')\n",
        "\n",
        "train_p2(num_epochs=10, \n",
        "         net=resnet,  \n",
        "         lr=0.03, \n",
        "         train_iter=train_dataloader,\n",
        "         test_iter=test_dataloader,\n",
        "         device=device,\n",
        "         optim_type=\"Adam\")"
      ],
      "metadata": {
        "id": "7cpwPQ9z8Tl3",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ba7bc86f-f1eb-4d3d-c788-d1f4a58fc87a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Files already downloaded and verified\n",
            "Files already downloaded and verified\n",
            "training on cuda\n",
            "loss 1.829, train acc 0.354, test acc 0.448\n",
            "loss 1.319, train acc 0.526, test acc 0.467\n",
            "loss 1.091, train acc 0.614, test acc 0.562\n",
            "loss 0.954, train acc 0.666, test acc 0.642\n",
            "loss 0.850, train acc 0.705, test acc 0.654\n",
            "loss 0.773, train acc 0.734, test acc 0.575\n",
            "loss 0.680, train acc 0.766, test acc 0.697\n",
            "loss 0.614, train acc 0.790, test acc 0.544\n",
            "loss 0.560, train acc 0.806, test acc 0.733\n",
            "loss 0.490, train acc 0.834, test acc 0.705\n",
            "3558.5 examples/sec on cuda\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### With Augment"
      ],
      "metadata": {
        "id": "dk7BdjXVfesr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_dataset = torchvision.datasets.CIFAR10(root=\"./\", train=True, download=True, transform=train_aug)\n",
        "test_dataset = torchvision.datasets.CIFAR10(root=\"./\", train=False, download=True, transform=test_aug)\n",
        "\n",
        "train_dataloader = torch.utils.data.DataLoader(dataset=train_dataset, batch_size=64, shuffle=True)\n",
        "test_dataloader = torch.utils.data.DataLoader(dataset=test_dataset, batch_size=64, shuffle=False)\n",
        "\n",
        "resnet = torchvision.models.resnet18(pretrained=False)\n",
        "\n",
        "device = torch.device('cuda')\n",
        "\n",
        "train_p2(num_epochs=10, \n",
        "         net=resnet,  \n",
        "         lr=0.03, \n",
        "         train_iter=train_dataloader,\n",
        "         test_iter=test_dataloader,\n",
        "         device=device,\n",
        "         optim_type=\"Adam\")"
      ],
      "metadata": {
        "id": "cbXetqFu8VkT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2cf2eb09-b755-4d14-e9e1-e5591452024c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Files already downloaded and verified\n",
            "Files already downloaded and verified\n",
            "training on cuda\n",
            "loss 1.932, train acc 0.310, test acc 0.244\n",
            "loss 1.494, train acc 0.449, test acc 0.373\n",
            "loss 1.303, train acc 0.527, test acc 0.503\n",
            "loss 1.161, train acc 0.584, test acc 0.480\n",
            "loss 1.077, train acc 0.619, test acc 0.558\n",
            "loss 0.998, train acc 0.647, test acc 0.549\n",
            "loss 0.942, train acc 0.668, test acc 0.618\n",
            "loss 0.883, train acc 0.692, test acc 0.658\n",
            "loss 0.834, train acc 0.708, test acc 0.625\n",
            "loss 0.784, train acc 0.728, test acc 0.667\n",
            "3555.2 examples/sec on cuda\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "-smKmcwsfnEu"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}