DB_05_going_modular.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "LmeMFXFXvFCB",
        "n5I6TOc8yRMP",
        "O_H3OJOE0AxB",
        "e5UeslIJ1-Cg",
        "B8SDlCXAC_tl",
        "miiWraMQVgAf",
        "Drf4_OXRTsUU",
        "WB7PvOlnUt4s",
        "jiktJGZTVN_X",
        "OVtGBAqpVr-H",
        "KqZKDgBRWgl1",
        "vlQbm-6mWh5Z",
        "0J8fk6PEnzLA",
        "MuOLDxzlojue",
        "dEiGxlIWohNe",
        "KDscxtyZovu_",
        "B3ipEbd7oyJa",
        "Ex_00t_R2Dze",
        "gOjTCqhN2Lmq",
        "v0Qdu65G2NgC",
        "qAlRQVrT2PYW"
      ],
      "toc_visible": true,
      "authorship_tag": "ABX9TyNdlOPMHBSEkQl8AP0HpG7W",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "4393ed00091f49b7b22fdf45810d1c1c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_82c063b20adb44ef9b06611ff0e6dfb9",
              "IPY_MODEL_0ca0ab70204e427cb33abfe6e172902e",
              "IPY_MODEL_65d624d2bba748e1b6a6d9e70de6814f"
            ],
            "layout": "IPY_MODEL_ba37cdb09ac3426992d81b6f7b124dbe"
          }
        },
        "82c063b20adb44ef9b06611ff0e6dfb9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_67823386bf574e3ab02e88131c06abab",
            "placeholder": "​",
            "style": "IPY_MODEL_4883c7faaf644858be3cab8b382fdc0f",
            "value": "100%"
          }
        },
        "0ca0ab70204e427cb33abfe6e172902e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6e00ba70cbb84c5cb457c6d9012de825",
            "max": 5,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_34243e21a9914add948982058ef4f858",
            "value": 5
          }
        },
        "65d624d2bba748e1b6a6d9e70de6814f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0e04749af0074ad0b57acb061abd4ea7",
            "placeholder": "​",
            "style": "IPY_MODEL_9cd84067ed54486b97aa05606c6917a2",
            "value": " 5/5 [15:00&lt;00:00, 181.09s/it]"
          }
        },
        "ba37cdb09ac3426992d81b6f7b124dbe": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "67823386bf574e3ab02e88131c06abab": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4883c7faaf644858be3cab8b382fdc0f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6e00ba70cbb84c5cb457c6d9012de825": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "34243e21a9914add948982058ef4f858": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "0e04749af0074ad0b57acb061abd4ea7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9cd84067ed54486b97aa05606c6917a2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "bf2bd21e179e4b0d81e4f55f0e7e8008": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_f5be26fc2c124c1993fea94bd80fade4",
              "IPY_MODEL_c514326973974b8189f44dae73312c68",
              "IPY_MODEL_987697c9888c4d80a21180ec85dedfee"
            ],
            "layout": "IPY_MODEL_cddf712c80bf4d09a261f21bab1f99b9"
          }
        },
        "f5be26fc2c124c1993fea94bd80fade4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_99b1ff0f549d4891ad67c631be4d47cb",
            "placeholder": "​",
            "style": "IPY_MODEL_71457971f4ad472d8697ba2995186a24",
            "value": "100%"
          }
        },
        "c514326973974b8189f44dae73312c68": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6fbcb64f9ead4005bb1f94a4b3975118",
            "max": 5,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_f024d07b4f37479e86889b63dae0ddda",
            "value": 5
          }
        },
        "987697c9888c4d80a21180ec85dedfee": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_7f1218849bca4075ab95e1b1f78b4ba1",
            "placeholder": "​",
            "style": "IPY_MODEL_66fff017167546d9a5b08d7279f8b932",
            "value": " 5/5 [00:39&lt;00:00,  9.25s/it]"
          }
        },
        "cddf712c80bf4d09a261f21bab1f99b9": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "99b1ff0f549d4891ad67c631be4d47cb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "71457971f4ad472d8697ba2995186a24": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6fbcb64f9ead4005bb1f94a4b3975118": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f024d07b4f37479e86889b63dae0ddda": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "7f1218849bca4075ab95e1b1f78b4ba1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "66fff017167546d9a5b08d7279f8b932": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/h0806449f/PyTorch/blob/main/DB_05_going_modular.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **=== 0. 簡介 ===**\n",
        "1. Run from command line\n",
        "2. Final target structure\n",
        "3. Will create multiple python file for different functions and purpose and achieve target structure\n"
      ],
      "metadata": {
        "id": "LmeMFXFXvFCB"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 0.1 Example for run python from command line\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "n5I6TOc8yRMP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# python train.py --model MODEL_NAME --batch_size BATCH_SIZE --lr LEARNING_RATE --num_epochs NUM_EPOCHS"
      ],
      "metadata": {
        "id": "O-jo3Vnk01Xe"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# python train.py --model model_1 --batch_size 100 --lr 0.001 --num_epochs 50"
      ],
      "metadata": {
        "id": "wE0sYgZJ06Lz"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 0.2 Final target structure"
      ],
      "metadata": {
        "id": "O_H3OJOE0AxB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# going_modular/\n",
        "# ├── going_modular/\n",
        "# │   ├── data_setup.py\n",
        "# │   ├── engine.py\n",
        "# │   ├── model_builder.py\n",
        "# │   ├── train.py\n",
        "# │   └── utils.py\n",
        "# ├── models/\n",
        "# │   ├── 05_going_modular_cell_mode_tinyvgg_model.pth\n",
        "# │   └── 05_going_modular_script_mode_tinyvgg_model.pth\n",
        "# └── data/\n",
        "#     └── pizza_steak_sushi/\n",
        "#         ├── train/\n",
        "#         │   ├── pizza/\n",
        "#         │   │   ├── image01.jpeg\n",
        "#         │   │   └── ...\n",
        "#         │   ├── steak/\n",
        "#         │   └── sushi/\n",
        "#         └── test/\n",
        "#             ├── pizza/\n",
        "#             ├── steak/\n",
        "#             └── sushi/"
      ],
      "metadata": {
        "id": "rr4evJWTz-Gw"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **=== 1. Get data (download data) ===**\n",
        "Download the data and set into specific data dtructure"
      ],
      "metadata": {
        "id": "e5UeslIJ1-Cg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Import\n",
        "import requests             # HTTP request\n",
        "import zipfile              # zip & unzip\n",
        "import os                   #\n",
        "from pathlib import Path    #\n",
        "\n",
        "# 設置 data path 變數\n",
        "data_folder = Path(\"Data/\")\n",
        "image_folder = data_folder / \"Pizze_Steak_Sushi\"\n",
        "\n",
        "# 判斷是否已有資料夾 -> 無: 創造 / 有: 可重複接受\n",
        "if image_folder.is_dir():\n",
        "    print(f\"{image_folder} 資料夾已存在\")\n",
        "else:\n",
        "    print(f\"{image_folder} 未發現資料夾, 創造 ...\")\n",
        "    image_folder.mkdir(parents = True, exist_ok = True) # 無上層資料夾, 也會創造 image_folder 資料夾 / 已有資料夾, 不返回error\n",
        "\n",
        "# 下載\n",
        "with open(image_folder / \"pizza_steak_sushi.zip\", \"wb\") as f: # 名稱需一致 (pizza_steak_sushi.zip)\n",
        "    request = requests.get(\"https://github.com/mrdbourke/pytorch-deep-learning/raw/main/data/pizza_steak_sushi.zip\")\n",
        "    print(\"downloading ...\")\n",
        "    f.write(request.content)\n",
        "\n",
        "# 解壓縮\n",
        "with zipfile.ZipFile (image_folder / \"pizza_steak_sushi.zip\", \"r\") as zip_f:\n",
        "    print(\"unziping ...\")\n",
        "    zip_f.extractall(image_folder)\n",
        "\n",
        "# 移除 壓縮檔\n",
        "os.remove(image_folder / \"pizza_steak_sushi.zip\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PCFXGWiR0Ixg",
        "outputId": "17b6e9c5-8982-40b1-e450-03192e2fb0cf"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Data/Pizze_Steak_Sushi 未發現資料夾, 創造 ...\n",
            "downloading ...\n",
            "unziping ...\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **=== 2. Data_setup.py ===**\n",
        "1. Wrtie -> functions in Data_setup.py\n",
        "2. Test\n",
        "2. Import"
      ],
      "metadata": {
        "id": "B8SDlCXAC_tl"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 2.1 Func: create_dataloader() in **Data_setup.py**"
      ],
      "metadata": {
        "id": "miiWraMQVgAf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# writefile 將以下內容, 寫入指定檔案中 (須去除前置# writefile 才會正常運作)\n",
        "# %%writefile going_modular/Data_setup.py\n",
        "\"\"\"將創造 Pytorch DataLoader 用於影像分類\"\"\"\n",
        "\n",
        "# Import\n",
        "import os\n",
        "from torchvision import datasets, transforms\n",
        "from torch.utils.data import DataLoader\n",
        "\n",
        "# Create function - create_dataloader()\n",
        "# 預先設置參數的資料型態: 日後易維護, IDE可能偵錯\n",
        "\n",
        "NUM_WORKERS = os.cpu_count()\n",
        "\n",
        "def create_dataloader(\n",
        "    train_dir: str,\n",
        "    test_dir: str,\n",
        "    transform: transforms.Compose,\n",
        "    batch_size: int,\n",
        "    num_workers: int=None):\n",
        "    # 解說\n",
        "    \"\"\"\n",
        "    創造:\n",
        "        train_dataloader, test_dataloader\n",
        "\n",
        "    變數:\n",
        "        train_dir: 訓練資料路徑\n",
        "        test_dir: 測試資料路徑\n",
        "        transform: Pytorch torchvision transforms\n",
        "        batch_size: 每批次資料數量\n",
        "        num_workers: workers per DataLoader\n",
        "\n",
        "    返還:\n",
        "        tuple(train_dataloader, test_dataloader, class_names)\n",
        "        class_names 為list 包含訓練資料的 classes\n",
        "\n",
        "    範例:\n",
        "        train_dataloader, test_dataloader, class_names = create_dataloaders(train_dir=path/to/train_dir,\n",
        "                                                                            test_dir=path/to/test_dir,\n",
        "                                                                            transform=some_transform,\n",
        "                                                                            batch_size=32,\n",
        "                                                                            num_workers=4)\n",
        "    \"\"\"\n",
        "\n",
        "    # NUM_WORKERS\n",
        "    if num_workers is None:\n",
        "        num_workers = NUM_WORKERS\n",
        "\n",
        "    # ImageFolder -> datasset\n",
        "    train_dataset = datasets.ImageFolder(train_dir, transform=transform)\n",
        "    test_dataset = datasets.ImageFolder(test_dir, transform=transform)\n",
        "\n",
        "    # Get train_dataset class\n",
        "    class_name = train_dataset.classes\n",
        "\n",
        "    # Dataset -> dataloader\n",
        "    train_dataloader = DataLoader(train_dataset,\n",
        "                                  batch_size=batch_size,\n",
        "                                  shuffle=True,\n",
        "                                  num_workers=num_workers,\n",
        "                                  pin_memory=True) # pin_memory 使用GPU時 可加快處理速度\n",
        "\n",
        "    test_dataloader = DataLoader(test_dataset,\n",
        "                                 batch_size=batch_size,\n",
        "                                 shuffle=False,\n",
        "                                 num_workers=num_workers,\n",
        "                                 pin_memory=True)\n",
        "\n",
        "    return train_dataloader, test_dataloader, class_name"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mNYg_o-LDHK7",
        "outputId": "5700c371-4df3-4225-fb5f-0f73dbddb212"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Writing going_modular/Data_setup.py\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 2.2 Test"
      ],
      "metadata": {
        "id": "Drf4_OXRTsUU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "from torchvision import transforms\n",
        "from torchvision.transforms import ToTensor\n",
        "\n",
        "train_path = \"/content/Data/Pizze_Steak_Sushi/train\"\n",
        "test_path = \"/content/Data/Pizze_Steak_Sushi/test\"\n",
        "\n",
        "transform = transforms.Compose([\n",
        "    ToTensor()\n",
        "])\n",
        "\n",
        "create_dataloader(train_dir=train_path,\n",
        "                  test_dir=test_path,\n",
        "                  transform=transform,\n",
        "                  batch_size=100)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2qSZz1VXP28s",
        "outputId": "7cee3fbf-50ff-48cf-9f41-430ba24b4f1a"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<torch.utils.data.dataloader.DataLoader at 0x7f1708dc70d0>,\n",
              " <torch.utils.data.dataloader.DataLoader at 0x7f1641f752d0>,\n",
              " ['pizza', 'steak', 'sushi'])"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 2.3 Import"
      ],
      "metadata": {
        "id": "WB7PvOlnUt4s"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from going_modular import Data_setup\n",
        "\n",
        "train_dataloader, test_dataloader, class_names = create_dataloader(train_dir=train_path,\n",
        "                                                                   test_dir=test_path,\n",
        "                                                                   transform=transform,\n",
        "                                                                   batch_size=100)"
      ],
      "metadata": {
        "id": "O4pqu8yzUx3Z"
      },
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **=== 3. Models.py ===**\n",
        "1. Write -> functions in Model.py\n",
        "2. Test\n",
        "3. Import"
      ],
      "metadata": {
        "id": "jiktJGZTVN_X"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 3.1 Multiple models in **Models.py**"
      ],
      "metadata": {
        "id": "OVtGBAqpVr-H"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# writefile 將以下內容, 寫入指定檔案中 (須去除前置# writefile 才會正常運作)\n",
        "# %%writefile going_modular/Models.py\n",
        "\"\"\"將創造 tinyVGG 模型\"\"\"\n",
        "\n",
        "# Import\n",
        "import torch\n",
        "from torch import nn\n",
        "\n",
        "# Class\n",
        "class TinyVGG(nn.Module):\n",
        "    \"\"\"\n",
        "    原網址: https://poloclub.github.io/cnn-explainer/\n",
        "    原網址輸入圖片皆為 64*64*3\n",
        "\n",
        "    Args:\n",
        "        input_shape: number of input channels\n",
        "        hidden_units: number of hidden units\n",
        "        output_shape: number of output channels\n",
        "    \"\"\"\n",
        "\n",
        "    def __init__(self, input_shape: int, hidden_units: int, output_shape: int)->None:\n",
        "        super().__init__()\n",
        "\n",
        "        self.block1 = nn.Sequential(\n",
        "            nn.Conv2d(in_channels=input_shape,\n",
        "                      out_channels=hidden_units,\n",
        "                      kernel_size=3,\n",
        "                      stride=1,\n",
        "                      padding=1),\n",
        "            nn.ReLU(),\n",
        "            nn.Conv2d(in_channels=hidden_units,\n",
        "                      out_channels=hidden_units,\n",
        "                      kernel_size=3,\n",
        "                      stride=1,\n",
        "                      padding=1),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2,\n",
        "                         stride=1))\n",
        "\n",
        "        self.block2 = nn.Sequential(\n",
        "            nn.Conv2d(hidden_units, hidden_units, kernel_size=3, stride=1, padding=1),\n",
        "            nn.ReLU(),\n",
        "            nn.Conv2d(hidden_units, hidden_units, kernel_size=3, stride=1, padding=1),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2, stride=1))\n",
        "\n",
        "        self.classifier = nn.Sequential(\n",
        "            nn.Flatten(),\n",
        "            nn.Linear(38440, 1024),\n",
        "            nn.ReLU(),\n",
        "            nn.Dropout(p=0.3),\n",
        "            nn.Linear(1024, 128),\n",
        "            nn.ReLU(),\n",
        "            nn.Dropout(p=0.3),\n",
        "            nn.Linear(128, output_shape))\n",
        "\n",
        "    def forward(self, x: torch.Tensor):\n",
        "        x = self.block1(x)\n",
        "        x = self.block2(x)\n",
        "        x = self.classifier(x)\n",
        "        return x"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QMn9RQohW1iu",
        "outputId": "50649717-c553-46cb-c632-a102c898f16a"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Writing going_modular/Models.py\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 3.2 Test"
      ],
      "metadata": {
        "id": "KqZKDgBRWgl1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
        "\n",
        "\n",
        "model_1 = TinyVGG(64,10,len(class_names)).to(device)\n",
        "print(model_1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wx_aUpkejvnZ",
        "outputId": "e77ba7c6-7e3d-4628-e166-06cf997d6136"
      },
      "execution_count": 80,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "TinyVGG(\n",
            "  (block1): Sequential(\n",
            "    (0): Conv2d(64, 10, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
            "    (1): ReLU()\n",
            "    (2): Conv2d(10, 10, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
            "    (3): ReLU()\n",
            "    (4): MaxPool2d(kernel_size=2, stride=1, padding=0, dilation=1, ceil_mode=False)\n",
            "  )\n",
            "  (block2): Sequential(\n",
            "    (0): Conv2d(10, 10, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
            "    (1): ReLU()\n",
            "    (2): Conv2d(10, 10, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
            "    (3): ReLU()\n",
            "    (4): MaxPool2d(kernel_size=2, stride=1, padding=0, dilation=1, ceil_mode=False)\n",
            "  )\n",
            "  (classifier): Sequential(\n",
            "    (0): Flatten(start_dim=1, end_dim=-1)\n",
            "    (1): Linear(in_features=38440, out_features=1024, bias=True)\n",
            "    (2): ReLU()\n",
            "    (3): Dropout(p=0.3, inplace=False)\n",
            "    (4): Linear(in_features=1024, out_features=128, bias=True)\n",
            "    (5): ReLU()\n",
            "    (6): Dropout(p=0.3, inplace=False)\n",
            "    (7): Linear(in_features=128, out_features=3, bias=True)\n",
            "  )\n",
            ")\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 3.3 Import"
      ],
      "metadata": {
        "id": "vlQbm-6mWh5Z"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Import\n",
        "from going_modular import Models\n",
        "import torch\n",
        "\n",
        "# Device\n",
        "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
        "\n",
        "# Manual_seed\n",
        "torch.manual_seed(42)\n",
        "\n",
        "# 指定\n",
        "model_1 = TinyVGG(input_shape=3,\n",
        "                  hidden_units=10,\n",
        "                  output_shape=3).to(device)\n",
        "print(model_1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_vuxMQ-pku6W",
        "outputId": "3df5d5aa-0b07-4ba7-d101-a545a9def43a"
      },
      "execution_count": 87,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "TinyVGG(\n",
            "  (block1): Sequential(\n",
            "    (0): Conv2d(3, 10, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
            "    (1): ReLU()\n",
            "    (2): Conv2d(10, 10, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
            "    (3): ReLU()\n",
            "    (4): MaxPool2d(kernel_size=2, stride=1, padding=0, dilation=1, ceil_mode=False)\n",
            "  )\n",
            "  (block2): Sequential(\n",
            "    (0): Conv2d(10, 10, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
            "    (1): ReLU()\n",
            "    (2): Conv2d(10, 10, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
            "    (3): ReLU()\n",
            "    (4): MaxPool2d(kernel_size=2, stride=1, padding=0, dilation=1, ceil_mode=False)\n",
            "  )\n",
            "  (classifier): Sequential(\n",
            "    (0): Flatten(start_dim=1, end_dim=-1)\n",
            "    (1): Linear(in_features=38440, out_features=1024, bias=True)\n",
            "    (2): ReLU()\n",
            "    (3): Dropout(p=0.3, inplace=False)\n",
            "    (4): Linear(in_features=1024, out_features=128, bias=True)\n",
            "    (5): ReLU()\n",
            "    (6): Dropout(p=0.3, inplace=False)\n",
            "    (7): Linear(in_features=128, out_features=3, bias=True)\n",
            "  )\n",
            ")\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **=== 4. Engine.py ===**\n",
        "1. Write -> 3 functions in Engine.py\n",
        "2. Functions: train, test, train_test_loop\n",
        "3. Test\n",
        "4. Import"
      ],
      "metadata": {
        "id": "0J8fk6PEnzLA"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 4.1 Func: train, test, train_test_loop in **Engine.py**"
      ],
      "metadata": {
        "id": "MuOLDxzlojue"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# %%writefile going_modular/Engine.py\n",
        "\"\"\"將創造 train(), test(), training_loop()\"\"\"\n",
        "\n",
        "# Import\n",
        "import torch\n",
        "from tqdm.auto import tqdm\n",
        "from typing import Dict, List, Tuple\n",
        "\n",
        "# Train\n",
        "def train(model: torch.nn.Module,\n",
        "          dataloader: torch.utils.data.DataLoader,\n",
        "          loss_fn: torch.nn.Module,\n",
        "          optimizer: torch.optim.Optimizer,\n",
        "          device: torch.device) -> Tuple[float, float]:\n",
        "    \"\"\"\n",
        "    Train 1 epoch\n",
        "\n",
        "    Args:\n",
        "        model: 欲訓練模型\n",
        "        dataloader: 該模型的 dataloader\n",
        "        loss_fn: 計算損失函數\n",
        "        optimizer: 優化損失函數\n",
        "        device: 運作在何種裝置上\n",
        "\n",
        "    Returns:\n",
        "        Tuple (train_loss, train_accuracy)\n",
        "    \"\"\"\n",
        "    model.to(device)\n",
        "    model.train()\n",
        "\n",
        "    train_loss, train_acc = 0, 0\n",
        "\n",
        "    for batch, (X_train, y_train) in enumerate(dataloader):\n",
        "        # To device\n",
        "        X_train, y_train = X_train.to(device), y_train.to(device)\n",
        "\n",
        "        # Forward pass\n",
        "        train_preds = model(X_train)\n",
        "\n",
        "        # Calculate & Accumulate the loss\n",
        "        loss = loss_fn(train_preds, y_train)\n",
        "        train_loss = train_loss + loss.item()\n",
        "\n",
        "        # Optimizer zero grad\n",
        "        optimizer.zero_grad()\n",
        "\n",
        "        # Loss back ward\n",
        "        loss.backward()\n",
        "\n",
        "        # Optimizer step\n",
        "        optimizer.step()\n",
        "\n",
        "        # Calculate the accuracy (y_preds is logits -> probs that sum=1 -> argmax for label)\n",
        "        train_preds_class = torch.argmax(torch.softmax(train_preds, dim=1), dim=1)\n",
        "        train_acc = train_acc + (train_preds_class == y_train).sum().item() / len(train_preds)\n",
        "\n",
        "    # Calculate average loss and accuracy for each batch\n",
        "    train_loss = train_loss / len(dataloader)\n",
        "    train_acc = train_acc / len(dataloader)\n",
        "\n",
        "    return train_loss, train_acc\n",
        "\n",
        "\n",
        "# Test\n",
        "def test(model: torch.nn.Module,\n",
        "         dataloader: torch.utils.data.DataLoader,\n",
        "         loss_fn: torch.nn.Module,\n",
        "         device: torch.device) -> Tuple[float, float]:\n",
        "    \"\"\"\n",
        "    Test 1 epoch\n",
        "\n",
        "    Args:\n",
        "        model: 欲訓練模型\n",
        "        dataloader: 該模型的 dataloader\n",
        "        loss_fn: 計算損失函數\n",
        "        device: 運作在何種裝置上\n",
        "\n",
        "    Returns:\n",
        "        Tuple (tesr_loss, test_accuracy)\n",
        "    \"\"\"\n",
        "    model.to(device)\n",
        "    model.eval()\n",
        "\n",
        "    test_loss, test_acc = 0, 0\n",
        "\n",
        "    with torch.inference_mode():\n",
        "        for batch, (X_test, y_test) in enumerate(dataloader):\n",
        "            # To device\n",
        "            X_test, y_test = X_test.to(device), y_test.to(device)\n",
        "\n",
        "            # Forward pass\n",
        "            test_preds = model(X_test)\n",
        "\n",
        "            # Calculate & accumulate the loss\n",
        "            loss = loss_fn(test_preds, y_test)\n",
        "            test_loss = test_loss + loss\n",
        "\n",
        "            # Calculate & accumulate the accuracy\n",
        "            test_preds_labels = torch.argmax(torch.softmax(test_preds, dim=1), dim=1)\n",
        "            test_acc = test_acc + (test_preds_labels == y_test).sum().item() / len(test_preds)\n",
        "\n",
        "        test_loss = test_loss / len(dataloader)\n",
        "        test_acc = test_acc / len(dataloader)\n",
        "\n",
        "        return test_loss, test_acc\n",
        "\n",
        "\n",
        "# Train_test_loop\n",
        "def train_test_loop(model: torch.nn.Module,\n",
        "                    train_dataloader: torch.utils.data.DataLoader,\n",
        "                    test_dataloader: torch.utils.data.DataLoader,\n",
        "                    loss_fn: torch.nn.Module,\n",
        "                    optimizer: torch.optim.Optimizer,\n",
        "                    epochs: int,\n",
        "                    device: torch.device) -> Dict[str, List]:\n",
        "    \"\"\"\n",
        "    Train & test for specific epochs\n",
        "\n",
        "    Args:\n",
        "        model: 欲訓練模型\n",
        "        dataloader: 該模型的 dataloader\n",
        "        loss_fn: 計算損失函數\n",
        "        optimizer: 優化損失函數\n",
        "        device: 運作在何種裝置上\n",
        "        epochs: 欲訓練次數\n",
        "\n",
        "    Returns:\n",
        "        Dict{train_loss: [ ... ],\n",
        "             train_acc: [ ... ],\n",
        "             test_loss: [ ... ],\n",
        "             test_acc: [ ... ]}\n",
        "    \"\"\"\n",
        "    # 空字典, 儲存結果\n",
        "    results = {\"train_loss\": [], \"train_acc\": [], \"test_loss\": [], \"test_acc\": []}\n",
        "\n",
        "    for epoch in tqdm(range(epochs)):\n",
        "        # Train\n",
        "        train_loss, train_acc = train(model=model,\n",
        "                                      dataloader=train_dataloader,\n",
        "                                      loss_fn=loss_fn,\n",
        "                                      optimizer=optimizer,\n",
        "                                      device=device)\n",
        "\n",
        "        # Test\n",
        "        test_loss, test_acc = test(model=model,\n",
        "                                   dataloader=test_dataloader,\n",
        "                                   loss_fn=loss_fn,\n",
        "                                   device=device)\n",
        "\n",
        "        # Print process\n",
        "        print(f\"Epochs: {epoch}\\n\"\n",
        "              f\"Train loss: {train_loss} | Train acc: {train_acc}\\n\"\n",
        "              f\"Test loss: {test_loss} | Test acc: {test_acc}\")\n",
        "\n",
        "        # Update result Dict\n",
        "        results[\"train_loss\"].append(train_loss)\n",
        "        results[\"train_acc\"].append(train_acc)\n",
        "        results[\"test_loss\"].append(test_loss)\n",
        "        results[\"test_acc\"].append(test_acc)\n",
        "\n",
        "    return results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ay84rTuyny_C",
        "outputId": "0abb4365-653b-444a-821f-de6927f1f6a7"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Writing going_modular/Engine.py\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 4.2 Test"
      ],
      "metadata": {
        "id": "dEiGxlIWohNe"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Data path\n",
        "train_path = \"/content/Data/Pizze_Steak_Sushi/train\"\n",
        "test_path = \"/content/Data/Pizze_Steak_Sushi/test\"\n",
        "\n",
        "# Transform\n",
        "transform = transforms.Compose([\n",
        "    transforms.Resize((64, 64)),\n",
        "    transforms.ToTensor()\n",
        "])\n",
        "\n",
        "# Dataloader\n",
        "BATCH_SIZE = 1\n",
        "train_dataloader, test_dataloader, class_names = create_dataloader(train_dir=train_path,\n",
        "                                                                   test_dir=test_path,\n",
        "                                                                   transform=transform,\n",
        "                                                                   batch_size=BATCH_SIZE)\n",
        "\n",
        "#\n",
        "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
        "loss_fn = nn.CrossEntropyLoss()\n",
        "optimizer = torch.optim.Adam(model_1.parameters(),\n",
        "                             lr=0.001)\n",
        "\n",
        "results_1 = train_test_loop(model=model_1,\n",
        "                            train_dataloader = train_dataloader,\n",
        "                            test_dataloader = test_dataloader,\n",
        "                            loss_fn = loss_fn,\n",
        "                            optimizer = optimizer,\n",
        "                            epochs=5,\n",
        "                            device = device)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 309,
          "referenced_widgets": [
            "4393ed00091f49b7b22fdf45810d1c1c",
            "82c063b20adb44ef9b06611ff0e6dfb9",
            "0ca0ab70204e427cb33abfe6e172902e",
            "65d624d2bba748e1b6a6d9e70de6814f",
            "ba37cdb09ac3426992d81b6f7b124dbe",
            "67823386bf574e3ab02e88131c06abab",
            "4883c7faaf644858be3cab8b382fdc0f",
            "6e00ba70cbb84c5cb457c6d9012de825",
            "34243e21a9914add948982058ef4f858",
            "0e04749af0074ad0b57acb061abd4ea7",
            "9cd84067ed54486b97aa05606c6917a2"
          ]
        },
        "id": "fyYluXEkCWie",
        "outputId": "3b626da5-9067-446b-9e8d-f2d41a431dc0"
      },
      "execution_count": 114,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/5 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "4393ed00091f49b7b22fdf45810d1c1c"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 0\n",
            "Train loss: 1.0614273049960805 | Train acc: 0.47555555555555556\n",
            "Test loss: 1.0874457359313965 | Test acc: 0.38666666666666666\n",
            "Epochs: 1\n",
            "Train loss: 0.9243123758469503 | Train acc: 0.5688888888888889\n",
            "Test loss: 1.0457584857940674 | Test acc: 0.4666666666666667\n",
            "Epochs: 2\n",
            "Train loss: 0.9977163959723798 | Train acc: 0.5555555555555556\n",
            "Test loss: 1.0302966833114624 | Test acc: 0.4\n",
            "Epochs: 3\n",
            "Train loss: 0.931575952624755 | Train acc: 0.6222222222222222\n",
            "Test loss: 1.0164368152618408 | Test acc: 0.48\n",
            "Epochs: 4\n",
            "Train loss: 0.7529415562789573 | Train acc: 0.6622222222222223\n",
            "Test loss: 1.197981595993042 | Test acc: 0.5066666666666667\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 4.3 Import"
      ],
      "metadata": {
        "id": "KDscxtyZovu_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from going_modular import Engine\n",
        "import inspect\n",
        "\n",
        "functions = inspect.getmembers(Engine, inspect.isfunction)\n",
        "for name, info in functions:\n",
        "    print(f\"Function name: {name} -> info: {info}\\n\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XHlffFqrN9Xd",
        "outputId": "ee91e2fa-8a5e-45b2-b620-e8ee33b6fb01"
      },
      "execution_count": 122,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Function name: test -> info: <function test at 0x7f16400ed510>\n",
            "\n",
            "Function name: train -> info: <function train at 0x7f16400ee7a0>\n",
            "\n",
            "Function name: train_test_loop -> info: <function train_test_loop at 0x7f1631f1e3b0>\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **=== 5. Utils.py ===**\n",
        "1. Func: save_model() in **Utils.py**\n",
        "2. Test\n",
        "4. Import"
      ],
      "metadata": {
        "id": "B3ipEbd7oyJa"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 5.1 Func: save_model() in **Utils.py**"
      ],
      "metadata": {
        "id": "Ex_00t_R2Dze"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# %%writefile going_modular/Utils.py\n",
        "\"\"\"目前有: 儲存功能\"\"\"\n",
        "import torch\n",
        "from pathlib import Path\n",
        "\n",
        "def save_model(model: torch.nn.Module,\n",
        "               target_dir: str,\n",
        "               model_name: str):\n",
        "    \"\"\"\n",
        "    儲存模型至指定路徑資料夾\n",
        "\n",
        "    Args:\n",
        "        model: 模型名稱,\n",
        "        target_dir: 模型儲存路徑,\n",
        "        model_name: 儲存後模型名稱, ex. model.pt / model.pth\n",
        "\n",
        "    Ex:\n",
        "        save_model(model = model_1,\n",
        "                   target_dir = Data/Models,\n",
        "                   model_name = \"Model_1_TinyVGG.pt\")\n",
        "    \"\"\"\n",
        "    # Create target dir\n",
        "    target_dir_path = Path(target_dir)\n",
        "    target_dir_path.mkdir(parents = True, exist_ok = True)\n",
        "\n",
        "    # 判斷模型是否以特定字串結尾, 否則返回指定訊息\n",
        "    # Create model saved path\n",
        "    assert model_name.endswith(\".pth\") or model_name.endswith(\".pt\"), \"model_name should ends with '.pt' or '.pth'.\"\n",
        "    model_saved_path = target_dir_path / model_name\n",
        "\n",
        "    # 儲存模型\n",
        "    print(f\"[INFO] Saving model to {model_saved_path}\")\n",
        "    torch.save(obj = model.state_dict(),\n",
        "               f = model_saved_path)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i4wu5hWa2LWy",
        "outputId": "74f700eb-88d0-4861-f05f-01b5fcdee123"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Writing going_modular/Utils.py\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 5.2 Test"
      ],
      "metadata": {
        "id": "gOjTCqhN2Lmq"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "vSIppPfa2NNy"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 5.3 import"
      ],
      "metadata": {
        "id": "v0Qdu65G2NgC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from going_modular import Utils\n",
        "import inspect\n",
        "\n",
        "functions = inspect.getmembers(Utils, inspect.isfunction)\n",
        "\n",
        "for name, info in functions:\n",
        "    print(f\"Function name: {name}\")\n",
        "    print(f\"Function info: {info}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Hlc_OcL72O7r",
        "outputId": "f67b5286-a62e-43ab-bfb0-2228b3da7f6b"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Function name: save_model\n",
            "Function info: <function save_model at 0x7fae271b5480>\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **=== 6. Train.py ===**\n",
        "1. 執行檔, 整合以上所有 functions\n",
        "2. 執行檔名稱通常為 train.py"
      ],
      "metadata": {
        "id": "qAlRQVrT2PYW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# %%writefile going_modular/Train.py\n",
        "\"\"\"Train a Pytorch image classification model\"\"\"\n",
        "# Import\n",
        "import os\n",
        "from going_modular import Data_setup, Engine, Models, Utils\n",
        "import torch\n",
        "import torchvision\n",
        "from torchvision import transforms\n",
        "\n",
        "# Hyperparameters\n",
        "NUM_EPOCHS = 5\n",
        "BATCH_SIZE = 32\n",
        "HIDDEN_UNITS = 10\n",
        "LEARNING_RATE = 0.001\n",
        "\n",
        "# Set train, test dir\n",
        "train_dir = \"/content/Data/Pizze_Steak_Sushi/train\"\n",
        "test_dir = \"/content/Data/Pizze_Steak_Sushi/test\"\n",
        "\n",
        "# Set transforms\n",
        "transform = transforms.Compose([\n",
        "    transforms.Resize((64, 64)),\n",
        "    transforms.ToTensor()\n",
        "])\n",
        "\n",
        "# Set dataloader\n",
        "# By calling func: create_dataloader() from Data_setup.py\n",
        "train_dataloader, test_dataloader, class_names = Data_setup.create_dataloader(train_dir = train_dir,\n",
        "                                                                              test_dir = test_dir,\n",
        "                                                                              transform = transform,\n",
        "                                                                              batch_size = BATCH_SIZE)\n",
        "\n",
        "# Set device\n",
        "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
        "\n",
        "# Set model\n",
        "# By calling Models.py\n",
        "model_2 = Models.TinyVGG(input_shape = 3,\n",
        "                         hidden_units = HIDDEN_UNITS,\n",
        "                         output_shape = len(class_names))\n",
        "\n",
        "# Set loss_fn & optimizer\n",
        "loss_fn = torch.nn.CrossEntropyLoss()\n",
        "optimizer = torch.optim.Adam(model_2.parameters(),\n",
        "                             lr = LEARNING_RATE)\n",
        "\n",
        "# !!! Start train !!!\n",
        "result_2 = Engine.train_test_loop(model = model_2,\n",
        "                                  train_dataloader = train_dataloader,\n",
        "                                  test_dataloader = test_dataloader,\n",
        "                                  loss_fn = loss_fn,\n",
        "                                  optimizer = optimizer,\n",
        "                                  device = device,\n",
        "                                  epochs = NUM_EPOCHS)\n",
        "\n",
        "# Save model_2\n",
        "# By calling save_model() from Utils.py\n",
        "Utils.save_model(model = model_2,\n",
        "                 target_dir = \"/content/going_modular\",\n",
        "                 model_name = \"Model_2_for_temptest.pt\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 327,
          "referenced_widgets": [
            "bf2bd21e179e4b0d81e4f55f0e7e8008",
            "f5be26fc2c124c1993fea94bd80fade4",
            "c514326973974b8189f44dae73312c68",
            "987697c9888c4d80a21180ec85dedfee",
            "cddf712c80bf4d09a261f21bab1f99b9",
            "99b1ff0f549d4891ad67c631be4d47cb",
            "71457971f4ad472d8697ba2995186a24",
            "6fbcb64f9ead4005bb1f94a4b3975118",
            "f024d07b4f37479e86889b63dae0ddda",
            "7f1218849bca4075ab95e1b1f78b4ba1",
            "66fff017167546d9a5b08d7279f8b932"
          ]
        },
        "id": "TqR-Cu659Sxx",
        "outputId": "ef7b5f78-28c4-4dc4-be64-ee580092351d"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/5 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "bf2bd21e179e4b0d81e4f55f0e7e8008"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 0\n",
            "Train loss: 1.1855529323220253 | Train acc: 0.41015625\n",
            "Test loss: 1.2640395164489746 | Test acc: 0.2604166666666667\n",
            "Epochs: 1\n",
            "Train loss: 1.1523059904575348 | Train acc: 0.37890625\n",
            "Test loss: 1.0996471643447876 | Test acc: 0.3125\n",
            "Epochs: 2\n",
            "Train loss: 1.0861448645591736 | Train acc: 0.40234375\n",
            "Test loss: 1.1283758878707886 | Test acc: 0.2604166666666667\n",
            "Epochs: 3\n",
            "Train loss: 1.1130668818950653 | Train acc: 0.3125\n",
            "Test loss: 1.09740149974823 | Test acc: 0.2604166666666667\n",
            "Epochs: 4\n",
            "Train loss: 1.0923727750778198 | Train acc: 0.4609375\n",
            "Test loss: 1.073072910308838 | Test acc: 0.47253787878787873\n",
            "[INFO] Saving model to /content/going_modular/Model_2_for_temptest.pt\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **=== Footnote ===**"
      ],
      "metadata": {
        "id": "kWqkYk3NF46P"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# python train.py\n",
        "# 可以於檔案內設定超參數後, 執行 py檔案"
      ],
      "metadata": {
        "id": "KHGpCaU2GE2R"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# python train.py --model MODEL_NAME --batch_size BATCH_SIZE --lr LEARNING_RATE --num_epochs NUM_EPOCHS\n",
        "# 也可以於 command line 直接設定超參數後, 執行 py檔案"
      ],
      "metadata": {
        "id": "tmXbARS7GH8e"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# 練習 Visualize\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.figure(figsize = (12, 4))\n",
        "\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(result_2[\"train_loss\"], label = \"Train Loss\")\n",
        "plt.plot(result_2[\"test_loss\"], label = \"Test Loss\")\n",
        "plt.title(\"Loss\")\n",
        "plt.legend()\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(result_2[\"train_acc\"], label = \"Train Accuracy\")\n",
        "plt.plot(result_2[\"test_acc\"], label = \"Test Accuracy\")\n",
        "plt.title(\"Accuracy\")\n",
        "plt.legend()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "id": "OeR9FlroGaVi",
        "outputId": "871634ec-b1d0-4dbd-c386-9e05750a5253"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fad5df78490>"
            ]
          },
          "metadata": {},
          "execution_count": 22
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}