TD6_CNNArchitectures.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "TD6_CNNArchitectures.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "pOukmG4j7Qyx",
        "ZTkfVe_TbECg"
      ],
      "toc_visible": true,
      "authorship_tag": "ABX9TyOtPB9ncnHAYhCCvxkSWNKo",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "a2432529ab734f4cae8aceb7a7a3129a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_deaed59505a04076b18c991182b8482d",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_ac12f8ff8a524269aafa86edf4160a4c",
              "IPY_MODEL_46b2d262143d48b9b27167bf519a4a2c"
            ]
          }
        },
        "deaed59505a04076b18c991182b8482d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "ac12f8ff8a524269aafa86edf4160a4c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_631501490b2f4a5f8e161f3e9eeba26b",
            "_dom_classes": [],
            "description": "",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_20265dcad32349a68e8d84287b08988a"
          }
        },
        "46b2d262143d48b9b27167bf519a4a2c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_ba1e656a9c00496081b4e3e8b4e3f784",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 170500096/? [00:08&lt;00:00, 20472350.46it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_080d0eb611354d5cae3c8a84d06f8892"
          }
        },
        "631501490b2f4a5f8e161f3e9eeba26b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "20265dcad32349a68e8d84287b08988a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "ba1e656a9c00496081b4e3e8b4e3f784": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "080d0eb611354d5cae3c8a84d06f8892": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "013996910f7f4d2697faacca35ee6728": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_66282fa8c7344afcafbb587b77461ef0",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_9c11ed381ae148e08bf81361e8c72451",
              "IPY_MODEL_89e7dccc6d4a4dc09f657b875031cb87"
            ]
          }
        },
        "66282fa8c7344afcafbb587b77461ef0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "9c11ed381ae148e08bf81361e8c72451": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_69293b6c1526455d8fec5c05297f310e",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 244418560,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 244418560,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_fa883ba7653d442ab56e97f874267c47"
          }
        },
        "89e7dccc6d4a4dc09f657b875031cb87": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_bb5e62dd36154e4c9c5330fc2dd82485",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 233M/233M [00:02&lt;00:00, 88.2MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_259023d69c3e40f39ff8c890bb47cbae"
          }
        },
        "69293b6c1526455d8fec5c05297f310e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "fa883ba7653d442ab56e97f874267c47": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "bb5e62dd36154e4c9c5330fc2dd82485": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "259023d69c3e40f39ff8c890bb47cbae": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "b02c7733cfbe46638df97d3df56a9306": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_f2f213c29b3e44549ba1072d707cf361",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_fb75d7e9be574831a4772810cc17c7e7",
              "IPY_MODEL_50e5fe6003a54a2fb85af9ad74b048cd"
            ]
          }
        },
        "f2f213c29b3e44549ba1072d707cf361": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "fb75d7e9be574831a4772810cc17c7e7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_51a8ab02847740e7a0308f7dde72f08f",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 531503671,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 531503671,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_aa63e2d9761f4137b8300144ab092686"
          }
        },
        "50e5fe6003a54a2fb85af9ad74b048cd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_1717f68be248410d9b27cc5ded0fe851",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 507M/507M [08:03&lt;00:00, 1.10MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_d841f87f62dc45caba71526224d383e7"
          }
        },
        "51a8ab02847740e7a0308f7dde72f08f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "aa63e2d9761f4137b8300144ab092686": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "1717f68be248410d9b27cc5ded0fe851": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "d841f87f62dc45caba71526224d383e7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "f07a6b9698d841698c38abcfc4017653": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_d50821987164451383836b569c077b0c",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_f4e8e515f4da4911ad72f9578323555f",
              "IPY_MODEL_65ab8ca766f44de9b855f287827dd1db"
            ]
          }
        },
        "d50821987164451383836b569c077b0c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "f4e8e515f4da4911ad72f9578323555f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_f7c2a6b6114f415f8e51649b7f7682c0",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 10,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 10,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_3433a9abb9fb4527890283c2e6fdac3c"
          }
        },
        "65ab8ca766f44de9b855f287827dd1db": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_9ada21cf2880481bba50cc424642e5f3",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 10/10 [05:27&lt;00:00, 32.77s/it]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_8520902a82dc42adabd81004f1f71df2"
          }
        },
        "f7c2a6b6114f415f8e51649b7f7682c0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "3433a9abb9fb4527890283c2e6fdac3c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "9ada21cf2880481bba50cc424642e5f3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "8520902a82dc42adabd81004f1f71df2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "46d7bbaccad0458ab52d5e44ad6aeac4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_3e36bcdc654d4b3db4ce02f57dda70d8",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_d026ff66c2a2494fa434e9fb2679d191",
              "IPY_MODEL_76b9c5fa459840a78ab2dd059c7612c7"
            ]
          }
        },
        "3e36bcdc654d4b3db4ce02f57dda70d8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "d026ff66c2a2494fa434e9fb2679d191": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_67df71fcc551486b9ef5642102d1b484",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 10,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 10,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_a05c20fe5e204169acb751f87ee64acd"
          }
        },
        "76b9c5fa459840a78ab2dd059c7612c7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_a281c4933fcc4bf8a01b3a5baf274c9f",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 10/10 [05:13&lt;00:00, 31.32s/it]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_82977c4a15fb4d148537a14d25c626cb"
          }
        },
        "67df71fcc551486b9ef5642102d1b484": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "a05c20fe5e204169acb751f87ee64acd": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "a281c4933fcc4bf8a01b3a5baf274c9f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "82977c4a15fb4d148537a14d25c626cb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "8dc6c54e8c5f4ad99f14d05144505a9c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_dbd533d84cb24e8994983e118c68167b",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_5bbecd4ebe214721b47fb9140d9e23b5",
              "IPY_MODEL_93c6f9402bc54c88a68994220bd6043f"
            ]
          }
        },
        "dbd533d84cb24e8994983e118c68167b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "5bbecd4ebe214721b47fb9140d9e23b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_1cc41ec45ae64da1b1db581373fd88e1",
            "_dom_classes": [],
            "description": "",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_e915be758aa8423dbe138cf6f781b8e3"
          }
        },
        "93c6f9402bc54c88a68994220bd6043f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_27ed6d5eab2148b2b3030299ec99c76c",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 26427392/? [00:04&lt;00:00, 6339398.87it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_50202f20508545e79d13fca8d3606001"
          }
        },
        "1cc41ec45ae64da1b1db581373fd88e1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "e915be758aa8423dbe138cf6f781b8e3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "27ed6d5eab2148b2b3030299ec99c76c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "50202f20508545e79d13fca8d3606001": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "e433bd147d4c443f998e61fa389551cc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_cd901762a5784799b4e55d1d44c871a8",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_25631a6c8663440782fa6934fcc469a8",
              "IPY_MODEL_d52cb9d568eb4f699bd3914e1f7fcb9b"
            ]
          }
        },
        "cd901762a5784799b4e55d1d44c871a8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "25631a6c8663440782fa6934fcc469a8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_35c150ba05d74ec39e8e37522062c46c",
            "_dom_classes": [],
            "description": "",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_8e3fe029227d4400b4b08d97ea743a56"
          }
        },
        "d52cb9d568eb4f699bd3914e1f7fcb9b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_7ed5c59c9d7d4911865ae879985f408e",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 32768/? [00:01&lt;00:00, 24212.81it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_af59ac319e324358a2d956064deeb5c8"
          }
        },
        "35c150ba05d74ec39e8e37522062c46c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "8e3fe029227d4400b4b08d97ea743a56": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "7ed5c59c9d7d4911865ae879985f408e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "af59ac319e324358a2d956064deeb5c8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "b53601df547f458cbb2f85be3a7bf973": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_788802b31c7c4725885c8bf74af760e6",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_bdbe8136586d476c895307ef2bdd1229",
              "IPY_MODEL_df908da626ea4fb1a5d27b775b49b850"
            ]
          }
        },
        "788802b31c7c4725885c8bf74af760e6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "bdbe8136586d476c895307ef2bdd1229": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_4ce3ab4979604731bdf1a167d34df045",
            "_dom_classes": [],
            "description": "",
            "_model_name": "IntProgressModel",
            "bar_style": "info",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_f5b8780f065747c5930550119832f3d4"
          }
        },
        "df908da626ea4fb1a5d27b775b49b850": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_7e6e1c9320524f83960cd598cce01e9e",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 4423680/? [00:16&lt;00:00, 805394.10it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_87b4dd731fc745e689fe2317e9e7bc4b"
          }
        },
        "4ce3ab4979604731bdf1a167d34df045": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "f5b8780f065747c5930550119832f3d4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "7e6e1c9320524f83960cd598cce01e9e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "87b4dd731fc745e689fe2317e9e7bc4b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "6c658ea12ad1459dbe1692093ae7373f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_9e0156f13c314988b65e9209d9dba79a",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_de2f3b7035744258a9cc11b96bdd6c3c",
              "IPY_MODEL_d494d3201e1e42b7ae4fdb26460396a8"
            ]
          }
        },
        "9e0156f13c314988b65e9209d9dba79a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "de2f3b7035744258a9cc11b96bdd6c3c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_ffda2f4c8da1416f8e3b43d00f0a7906",
            "_dom_classes": [],
            "description": "",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_0e844fca3d594e43804e776d7e08ca5d"
          }
        },
        "d494d3201e1e42b7ae4fdb26460396a8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_8185daed45054680bca8dd9fc9781730",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 8192/? [00:00&lt;00:00, 17300.44it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_202a3bd512564a1e80095467812c60eb"
          }
        },
        "ffda2f4c8da1416f8e3b43d00f0a7906": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "0e844fca3d594e43804e776d7e08ca5d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "8185daed45054680bca8dd9fc9781730": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "202a3bd512564a1e80095467812c60eb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "7a38788e9405460dbfb9a81ddfdafdd6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_131f96954d23497f9705235384f32714",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_376dd3f3010b4c24b8b864df14c7fa77",
              "IPY_MODEL_b8af58d01d9748c2bfa2cde80e6748c4"
            ]
          }
        },
        "131f96954d23497f9705235384f32714": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "376dd3f3010b4c24b8b864df14c7fa77": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_d251bfd320284028afcdf07d63f1f87d",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 10,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 10,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_155bdd082cea403f8e04db76c83c0a0c"
          }
        },
        "b8af58d01d9748c2bfa2cde80e6748c4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_324d740bd6414f0d84fe1029a9b29240",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 10/10 [10:56&lt;00:00, 65.69s/it]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_56d3a8457f4b4b7f8723f2ae58e43847"
          }
        },
        "d251bfd320284028afcdf07d63f1f87d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "155bdd082cea403f8e04db76c83c0a0c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "324d740bd6414f0d84fe1029a9b29240": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "56d3a8457f4b4b7f8723f2ae58e43847": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "1b61b9178bd1405197ab6e852903e97f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_7f1840aa27864dea8df473889e4fa437",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_bdb74454631b454bb33163b34928c80d",
              "IPY_MODEL_29ecf9fd4aa141569e940c23961e46fd"
            ]
          }
        },
        "7f1840aa27864dea8df473889e4fa437": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "bdb74454631b454bb33163b34928c80d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_806b3a1ce73147e3a4c47b8e291a2244",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 10,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 10,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_6b8d03fb7d43417eb8a9b9e0cd277f53"
          }
        },
        "29ecf9fd4aa141569e940c23961e46fd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_cf17fcd276124a8e9cb4b919ccd75b82",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 10/10 [06:10&lt;00:00, 37.09s/it]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_9f05380c02fa4717817d3953b3f7c435"
          }
        },
        "806b3a1ce73147e3a4c47b8e291a2244": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "6b8d03fb7d43417eb8a9b9e0cd277f53": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "cf17fcd276124a8e9cb4b919ccd75b82": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "9f05380c02fa4717817d3953b3f7c435": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "a0a9bca1b29d4ee3bffe47f4fe4dc491": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_d585bad3453c47048ac625a2079b262b",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_fdf1f6209ad940168c39f33cb621b4a2",
              "IPY_MODEL_9693f7501b524c9ab1580df0aa7feab6"
            ]
          }
        },
        "d585bad3453c47048ac625a2079b262b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "fdf1f6209ad940168c39f33cb621b4a2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_3e25ec94c7fd4117a732ae99b4adeae7",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 10,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 10,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_a47912d8d6914632a747921179a64909"
          }
        },
        "9693f7501b524c9ab1580df0aa7feab6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_507457deba8c462885fa1871defc50dc",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 10/10 [12:13&lt;00:00, 73.33s/it]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_2159a93e0a034b6ab25a3ea3a878bcca"
          }
        },
        "3e25ec94c7fd4117a732ae99b4adeae7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "a47912d8d6914632a747921179a64909": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "507457deba8c462885fa1871defc50dc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "2159a93e0a034b6ab25a3ea3a878bcca": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "01ba15a12a6d4de08f6c57c23b66de21": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_e82ec39c040b4543945243b12f0ad2c2",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_e7be80fc0ca9443c917e0cb6a99c89db",
              "IPY_MODEL_6f78ca37d82947898f8036228de320c9"
            ]
          }
        },
        "e82ec39c040b4543945243b12f0ad2c2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "e7be80fc0ca9443c917e0cb6a99c89db": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_75bdfa6f78b444f9abc0651dce0edc6b",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 10,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 10,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_d076df6dcec64b68947a944d1bb86760"
          }
        },
        "6f78ca37d82947898f8036228de320c9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_d9a91775898b4a5cb3206c257746e696",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 10/10 [12:09&lt;00:00, 72.98s/it]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_5f07b8bc85c5420c831f6d82363644a4"
          }
        },
        "75bdfa6f78b444f9abc0651dce0edc6b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_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"
          }
        },
        "d076df6dcec64b68947a944d1bb86760": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "d9a91775898b4a5cb3206c257746e696": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "5f07b8bc85c5420c831f6d82363644a4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/thinhngo-x/deep-learning/blob/master/TD6_CNNArchitectures.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pOukmG4j7Qyx",
        "colab_type": "text"
      },
      "source": [
        "# 0. Setup"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fVjZRSBX7VkU",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import torch.nn as nn\n",
        "import torch.utils.model_zoo as model_zoo\n",
        "from tqdm.notebook import tqdm"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NT4dEo7ZKDsc",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import torch\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "from torch.utils.data.sampler import SubsetRandomSampler"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2ldsOZGniW3O",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def imshow(img):\n",
        "    npimg = img.numpy()\n",
        "    plt.imshow(np.transpose(npimg, (1, 2, 0)))\n",
        "    plt.show()\n",
        "\n",
        "def show_first_layer(model):\n",
        "    filter = model.features[0].weight.cpu().data\n",
        "    filter = (1/(2*filter.max()))*filter + 0.5 #Normalizing the values to [0,1]\n",
        "    print(filter.shape)\n",
        "    imshow(torchvision.utils.make_grid(filter))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "x7Qj_9nmOCMa",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def init(m):\n",
        "    if isinstance(m, nn.Conv2d):\n",
        "        nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')\n",
        "        if m.bias is not None:\n",
        "            nn.init.constant_(m.bias, 0)\n",
        "    elif isinstance(m, nn.BatchNorm2d):\n",
        "        nn.init.constant_(m.weight, 1)\n",
        "        nn.init.constant_(m.bias, 0)\n",
        "    elif isinstance(m, nn.Linear):\n",
        "        nn.init.normal_(m.weight, 0, 0.01)\n",
        "        nn.init.constant_(m.bias, 0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uoo2oHec7YiA",
        "colab_type": "text"
      },
      "source": [
        "# 1. AlexNet"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o-Wy1SR2Jwno",
        "colab_type": "code",
        "outputId": "49191a97-5460-49c4-9888-322cf3a4708c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369,
          "referenced_widgets": [
            "a2432529ab734f4cae8aceb7a7a3129a",
            "deaed59505a04076b18c991182b8482d",
            "ac12f8ff8a524269aafa86edf4160a4c",
            "46b2d262143d48b9b27167bf519a4a2c",
            "631501490b2f4a5f8e161f3e9eeba26b",
            "20265dcad32349a68e8d84287b08988a",
            "ba1e656a9c00496081b4e3e8b4e3f784",
            "080d0eb611354d5cae3c8a84d06f8892"
          ]
        }
      },
      "source": [
        "torchvision.transforms.functional.resize\n",
        "transform = transforms.Compose(\n",
        "    [\n",
        "     transforms.Resize(size=(224, 224)),\n",
        "     transforms.ToTensor(),\n",
        "     transforms.Normalize((0.5,), (0.5,)),\n",
        "])\n",
        "     \n",
        "\n",
        "batch_size = 64\n",
        "\n",
        "idx_train = np.arange(50000)\n",
        "np.random.shuffle(idx_train)\n",
        "idx_train = idx_train[:1000]\n",
        "\n",
        "trainset = torchvision.datasets.CIFAR10(root=\"./data\", train=True, transform=transform, download=True)\n",
        "trainloader = torch.utils.data.DataLoader(trainset,batch_size=batch_size,shuffle=False,num_workers=2,\n",
        "                                         sampler=SubsetRandomSampler(idx_train)\n",
        "                                         )\n",
        "\n",
        "idx_test = np.arange(10000)\n",
        "np.random.shuffle(idx_test)\n",
        "idx_test = idx_train[:1000]\n",
        "\n",
        "testset = torchvision.datasets.CIFAR10(root=\"./data\", train=False, transform=transform, download=True)\n",
        "testloader = torch.utils.data.DataLoader(trainset,batch_size=batch_size,shuffle=False,num_workers=2)\n",
        "\n",
        "\n",
        "def imshow_(img):\n",
        "    img = img / 2 + 0.5     # unnormalize\n",
        "    npimg = img.numpy()\n",
        "    plt.imshow(np.transpose(npimg, (1, 2, 0)))\n",
        "    plt.show()\n",
        "\n",
        "\n",
        "# get some random training images\n",
        "dataiter = iter(trainloader)\n",
        "images, labels = dataiter.next()\n",
        "\n",
        "# show images\n",
        "imshow_(torchvision.utils.make_grid(images))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "a2432529ab734f4cae8aceb7a7a3129a",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Extracting ./data/cifar-10-python.tar.gz to ./data\n",
            "Files already downloaded and verified\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fxbTptDALrlR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "criterion = nn.CrossEntropyLoss()\n",
        "\n",
        "def accuracy(net, test_loader, cuda=True):\n",
        "  net.eval()\n",
        "  correct = 0\n",
        "  total = 0\n",
        "  loss = 0\n",
        "  with torch.no_grad():\n",
        "      for data in test_loader:\n",
        "          images, labels = data\n",
        "          if cuda:\n",
        "            images = images.type(torch.cuda.FloatTensor)\n",
        "            labels = labels.type(torch.cuda.LongTensor)\n",
        "          outputs = net(images)\n",
        "          # loss+= criterion(outputs, labels).item()\n",
        "          _, predicted = torch.max(outputs.data, 1)\n",
        "          total += labels.size(0)\n",
        "          correct += (predicted == labels).sum().item()\n",
        "          # if total > 100:\n",
        "            # break\n",
        "  net.train()\n",
        "  print('Accuracy of the network on the test images: %d %%' % (\n",
        "      100 * correct / total))\n",
        "  # return (100.0 * correct / total, loss/total)\n",
        "  return 100.0 * correct\n",
        "\n",
        "def train(net, optimizer, train_loader, test_loader, loss,  n_epoch = 5,\n",
        "          train_acc_period = 100,\n",
        "          test_acc_period = 5,\n",
        "          cuda=True):\n",
        "  loss_train = []\n",
        "  loss_test = []\n",
        "  total = 0\n",
        "  for epoch in range(n_epoch):  # loop over the dataset multiple times\n",
        "      running_loss = 0.0\n",
        "      running_acc = 0.0\n",
        "      for i, data in enumerate(train_loader, 0):\n",
        "          # get the inputs\n",
        "          inputs, labels = data\n",
        "          if cuda:\n",
        "            inputs = inputs.type(torch.cuda.FloatTensor)\n",
        "            labels = labels.type(torch.cuda.LongTensor)\n",
        "          # print(inputs.shape)\n",
        "          # zero the parameter gradients\n",
        "          optimizer.zero_grad()\n",
        "\n",
        "          # forward + backward + optimize\n",
        "          outputs = net(inputs)\n",
        "          \n",
        "          loss = criterion(outputs, labels)\n",
        "          loss.backward()\n",
        "          optimizer.step()\n",
        "          total += labels.size(0)\n",
        "          # print statistics\n",
        "          running_loss = 0.33*loss.item()/labels.size(0) + 0.66*running_loss\n",
        "          _, predicted = torch.max(outputs.data, 1)\n",
        "          correct = (predicted == labels).sum().item()/labels.size(0)\n",
        "          running_acc = 0.3*correct + 0.66*running_acc\n",
        "          if i % train_acc_period == train_acc_period-1:\n",
        "            print('[%d, %5d] loss: %.3f' %(epoch + 1, i + 1, running_loss))\n",
        "            print('[%d, %5d] acc: %.3f' %(epoch + 1, i + 1, running_acc))\n",
        "            running_loss = 0.0\n",
        "            total = 0\n",
        "            # break\n",
        "      if epoch % test_acc_period == test_acc_period-1:\n",
        "          cur_acc, cur_loss = accuracy(net, test_loader, cuda=cuda)\n",
        "          print('[%d] loss: %.3f' %(epoch + 1, cur_loss))\n",
        "          print('[%d] acc: %.3f' %(epoch + 1, cur_acc))\n",
        "      \n",
        "  print('Finished Training')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LMJo3qKe7ie6",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class AlexNet(nn.Module):\n",
        "\n",
        "    def __init__(self, num_classes=1000):\n",
        "        super(AlexNet, self).__init__()\n",
        "        self.features = nn.Sequential(#224\n",
        "            nn.Conv2d(3, 64, 11, stride=4, padding=2), # 55\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(2, stride=2), # 27\n",
        "            nn.Conv2d(64, 192, 5, padding=2), # 27\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(2, stride=2), # 13\n",
        "            nn.Conv2d(192, 384, 3, padding=1), # 13\n",
        "            nn.ReLU(),\n",
        "            nn.Conv2d(384, 256, 3, padding=1), # 13\n",
        "            nn.ReLU(),\n",
        "            nn.Conv2d(256, 256, 3, padding=1), # 13\n",
        "            nn.ReLU(),\n",
        "\t\t)\n",
        "        self.avgpool = nn.AdaptiveAvgPool2d((6, 6))\n",
        "        self.classifier = nn.Sequential(\n",
        "            nn.Dropout(),\n",
        "            nn.Linear(256 * 6 * 6, 4096),\n",
        "            nn.ReLU(inplace=True),\n",
        "            nn.Dropout(),\n",
        "            nn.Linear(4096, 4096),\n",
        "            nn.ReLU(inplace=True),\n",
        "            nn.Linear(4096, num_classes),\n",
        "        )\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.features(x)\n",
        "        x = self.avgpool(x)\n",
        "        x = x.view(x.size(0), 256 * 6 * 6)\n",
        "        x = self.classifier(x)\n",
        "        return x"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Md998KORJLoI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_urls = {\n",
        "'alexnet': 'https://download.pytorch.org/models/alexnet-owt-4df8aa71.pth',\n",
        "}\n",
        "\n",
        "def alexnet_classifier(num_classes):\n",
        "    classifier = nn.Sequential(\n",
        "        nn.Dropout(),\n",
        "        nn.Linear(256 * 6 * 6, 128),\n",
        "        nn.ReLU(inplace=True),\n",
        "        nn.Dropout(),\n",
        "        nn.Linear(128, 64),\n",
        "        nn.ReLU(inplace=True),\n",
        "        nn.Linear(64, num_classes),\n",
        "    )\n",
        "    return classifier\n",
        "\n",
        "def alexnet(num_classes, pretrained=False, **kwargs):\n",
        "    \"\"\"AlexNet model architecture from the \"One weird trick...\" paper.\n",
        "    Args:\n",
        "    pretrained (bool): If True, returns a model pre-trained on ImageNet\n",
        "    \"\"\"\n",
        "    model = AlexNet(**kwargs)\n",
        "    if pretrained:\n",
        "        model.load_state_dict(model_zoo.load_url(model_urls['alexnet']))\n",
        "        for p in model.features.parameters():\n",
        "            p.requires_grad=False\n",
        "    classifier = alexnet_classifier(num_classes)\n",
        "    model.classifier = classifier\n",
        "    return model"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "x_3U5kgaNMDN",
        "colab_type": "code",
        "outputId": "6ce1c128-7564-4527-a1af-0d5bef811ae9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "net_no_pretrained = alexnet(num_classes=10, pretrained=False)\n",
        "\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    net_no_pretrained.cuda()\n",
        "learning_rate = 1e-3\n",
        "optimizer = torch.optim.Adam(net_no_pretrained.parameters(),lr=learning_rate)\n",
        "train(net_no_pretrained, optimizer, trainloader, testloader, criterion,  n_epoch = 50,\n",
        "      train_acc_period = 10,\n",
        "      test_acc_period = 1000)\n",
        "accuracy(net_no_pretrained, testloader, cuda=use_cuda)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "using cuda\n",
            "[1,    10] loss: 0.034\n",
            "[1,    10] acc: 0.152\n",
            "[2,    10] loss: 0.034\n",
            "[2,    10] acc: 0.143\n",
            "[3,    10] loss: 0.033\n",
            "[3,    10] acc: 0.150\n",
            "[4,    10] loss: 0.033\n",
            "[4,    10] acc: 0.191\n",
            "[5,    10] loss: 0.030\n",
            "[5,    10] acc: 0.208\n",
            "[6,    10] loss: 0.032\n",
            "[6,    10] acc: 0.173\n",
            "[7,    10] loss: 0.032\n",
            "[7,    10] acc: 0.183\n",
            "[8,    10] loss: 0.031\n",
            "[8,    10] acc: 0.194\n",
            "[9,    10] loss: 0.030\n",
            "[9,    10] acc: 0.208\n",
            "[10,    10] loss: 0.029\n",
            "[10,    10] acc: 0.243\n",
            "[11,    10] loss: 0.029\n",
            "[11,    10] acc: 0.243\n",
            "[12,    10] loss: 0.028\n",
            "[12,    10] acc: 0.281\n",
            "[13,    10] loss: 0.029\n",
            "[13,    10] acc: 0.258\n",
            "[14,    10] loss: 0.028\n",
            "[14,    10] acc: 0.305\n",
            "[15,    10] loss: 0.027\n",
            "[15,    10] acc: 0.297\n",
            "[16,    10] loss: 0.027\n",
            "[16,    10] acc: 0.280\n",
            "[17,    10] loss: 0.025\n",
            "[17,    10] acc: 0.317\n",
            "[18,    10] loss: 0.025\n",
            "[18,    10] acc: 0.310\n",
            "[19,    10] loss: 0.025\n",
            "[19,    10] acc: 0.321\n",
            "[20,    10] loss: 0.024\n",
            "[20,    10] acc: 0.317\n",
            "[21,    10] loss: 0.023\n",
            "[21,    10] acc: 0.352\n",
            "[22,    10] loss: 0.024\n",
            "[22,    10] acc: 0.339\n",
            "[23,    10] loss: 0.022\n",
            "[23,    10] acc: 0.404\n",
            "[24,    10] loss: 0.022\n",
            "[24,    10] acc: 0.411\n",
            "[25,    10] loss: 0.020\n",
            "[25,    10] acc: 0.412\n",
            "[26,    10] loss: 0.019\n",
            "[26,    10] acc: 0.450\n",
            "[27,    10] loss: 0.021\n",
            "[27,    10] acc: 0.407\n",
            "[28,    10] loss: 0.021\n",
            "[28,    10] acc: 0.402\n",
            "[29,    10] loss: 0.019\n",
            "[29,    10] acc: 0.455\n",
            "[30,    10] loss: 0.017\n",
            "[30,    10] acc: 0.491\n",
            "[31,    10] loss: 0.020\n",
            "[31,    10] acc: 0.466\n",
            "[32,    10] loss: 0.017\n",
            "[32,    10] acc: 0.502\n",
            "[33,    10] loss: 0.016\n",
            "[33,    10] acc: 0.505\n",
            "[34,    10] loss: 0.015\n",
            "[34,    10] acc: 0.515\n",
            "[35,    10] loss: 0.015\n",
            "[35,    10] acc: 0.500\n",
            "[36,    10] loss: 0.015\n",
            "[36,    10] acc: 0.538\n",
            "[37,    10] loss: 0.017\n",
            "[37,    10] acc: 0.510\n",
            "[38,    10] loss: 0.015\n",
            "[38,    10] acc: 0.555\n",
            "[39,    10] loss: 0.014\n",
            "[39,    10] acc: 0.593\n",
            "[40,    10] loss: 0.014\n",
            "[40,    10] acc: 0.564\n",
            "[41,    10] loss: 0.012\n",
            "[41,    10] acc: 0.589\n",
            "[42,    10] loss: 0.011\n",
            "[42,    10] acc: 0.637\n",
            "[43,    10] loss: 0.010\n",
            "[43,    10] acc: 0.660\n",
            "[44,    10] loss: 0.011\n",
            "[44,    10] acc: 0.621\n",
            "[45,    10] loss: 0.011\n",
            "[45,    10] acc: 0.610\n",
            "[46,    10] loss: 0.009\n",
            "[46,    10] acc: 0.653\n",
            "[47,    10] loss: 0.009\n",
            "[47,    10] acc: 0.682\n",
            "[48,    10] loss: 0.009\n",
            "[48,    10] acc: 0.667\n",
            "[49,    10] loss: 0.009\n",
            "[49,    10] acc: 0.663\n",
            "[50,    10] loss: 0.009\n",
            "[50,    10] acc: 0.694\n",
            "Finished Training\n",
            "Accuracy of the network on the test images: 41 %\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2070600.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s8urxWH9NeFF",
        "colab_type": "code",
        "outputId": "12279a7f-840c-4c60-938e-ac5d68cf2996",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000,
          "referenced_widgets": [
            "013996910f7f4d2697faacca35ee6728",
            "66282fa8c7344afcafbb587b77461ef0",
            "9c11ed381ae148e08bf81361e8c72451",
            "89e7dccc6d4a4dc09f657b875031cb87",
            "69293b6c1526455d8fec5c05297f310e",
            "fa883ba7653d442ab56e97f874267c47",
            "bb5e62dd36154e4c9c5330fc2dd82485",
            "259023d69c3e40f39ff8c890bb47cbae"
          ]
        }
      },
      "source": [
        "net = alexnet(10, pretrained=True)\n",
        "\n",
        "# net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    net.cuda()\n",
        "learning_rate = 1e-3\n",
        "optimizer = torch.optim.Adam(net.parameters(),lr=learning_rate)\n",
        "train(net, optimizer, trainloader, testloader, criterion,  n_epoch = 50,\n",
        "      train_acc_period = 10,\n",
        "      test_acc_period = 1000)\n",
        "accuracy(net, testloader, cuda=use_cuda)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading: \"https://download.pytorch.org/models/alexnet-owt-4df8aa71.pth\" to /root/.cache/torch/checkpoints/alexnet-owt-4df8aa71.pth\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "013996910f7f4d2697faacca35ee6728",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=244418560), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "using cuda\n",
            "[1,    10] loss: 0.033\n",
            "[1,    10] acc: 0.146\n",
            "[2,    10] loss: 0.028\n",
            "[2,    10] acc: 0.314\n",
            "[3,    10] loss: 0.022\n",
            "[3,    10] acc: 0.439\n",
            "[4,    10] loss: 0.018\n",
            "[4,    10] acc: 0.499\n",
            "[5,    10] loss: 0.016\n",
            "[5,    10] acc: 0.533\n",
            "[6,    10] loss: 0.012\n",
            "[6,    10] acc: 0.623\n",
            "[7,    10] loss: 0.012\n",
            "[7,    10] acc: 0.627\n",
            "[8,    10] loss: 0.012\n",
            "[8,    10] acc: 0.639\n",
            "[9,    10] loss: 0.010\n",
            "[9,    10] acc: 0.680\n",
            "[10,    10] loss: 0.008\n",
            "[10,    10] acc: 0.687\n",
            "[11,    10] loss: 0.007\n",
            "[11,    10] acc: 0.744\n",
            "[12,    10] loss: 0.006\n",
            "[12,    10] acc: 0.750\n",
            "[13,    10] loss: 0.006\n",
            "[13,    10] acc: 0.747\n",
            "[14,    10] loss: 0.005\n",
            "[14,    10] acc: 0.784\n",
            "[15,    10] loss: 0.005\n",
            "[15,    10] acc: 0.787\n",
            "[16,    10] loss: 0.004\n",
            "[16,    10] acc: 0.792\n",
            "[17,    10] loss: 0.004\n",
            "[17,    10] acc: 0.795\n",
            "[18,    10] loss: 0.003\n",
            "[18,    10] acc: 0.807\n",
            "[19,    10] loss: 0.003\n",
            "[19,    10] acc: 0.816\n",
            "[20,    10] loss: 0.003\n",
            "[20,    10] acc: 0.820\n",
            "[21,    10] loss: 0.003\n",
            "[21,    10] acc: 0.825\n",
            "[22,    10] loss: 0.002\n",
            "[22,    10] acc: 0.822\n",
            "[23,    10] loss: 0.003\n",
            "[23,    10] acc: 0.826\n",
            "[24,    10] loss: 0.002\n",
            "[24,    10] acc: 0.814\n",
            "[25,    10] loss: 0.002\n",
            "[25,    10] acc: 0.829\n",
            "[26,    10] loss: 0.002\n",
            "[26,    10] acc: 0.833\n",
            "[27,    10] loss: 0.002\n",
            "[27,    10] acc: 0.839\n",
            "[28,    10] loss: 0.002\n",
            "[28,    10] acc: 0.824\n",
            "[29,    10] loss: 0.002\n",
            "[29,    10] acc: 0.839\n",
            "[30,    10] loss: 0.002\n",
            "[30,    10] acc: 0.840\n",
            "[31,    10] loss: 0.002\n",
            "[31,    10] acc: 0.840\n",
            "[32,    10] loss: 0.001\n",
            "[32,    10] acc: 0.842\n",
            "[33,    10] loss: 0.002\n",
            "[33,    10] acc: 0.833\n",
            "[34,    10] loss: 0.002\n",
            "[34,    10] acc: 0.846\n",
            "[35,    10] loss: 0.001\n",
            "[35,    10] acc: 0.853\n",
            "[36,    10] loss: 0.001\n",
            "[36,    10] acc: 0.839\n",
            "[37,    10] loss: 0.001\n",
            "[37,    10] acc: 0.847\n",
            "[38,    10] loss: 0.001\n",
            "[38,    10] acc: 0.844\n",
            "[39,    10] loss: 0.001\n",
            "[39,    10] acc: 0.839\n",
            "[40,    10] loss: 0.001\n",
            "[40,    10] acc: 0.837\n",
            "[41,    10] loss: 0.002\n",
            "[41,    10] acc: 0.839\n",
            "[42,    10] loss: 0.001\n",
            "[42,    10] acc: 0.859\n",
            "[43,    10] loss: 0.001\n",
            "[43,    10] acc: 0.837\n",
            "[44,    10] loss: 0.001\n",
            "[44,    10] acc: 0.853\n",
            "[45,    10] loss: 0.002\n",
            "[45,    10] acc: 0.818\n",
            "[46,    10] loss: 0.001\n",
            "[46,    10] acc: 0.859\n",
            "[47,    10] loss: 0.001\n",
            "[47,    10] acc: 0.851\n",
            "[48,    10] loss: 0.001\n",
            "[48,    10] acc: 0.848\n",
            "[49,    10] loss: 0.001\n",
            "[49,    10] acc: 0.849\n",
            "[50,    10] loss: 0.001\n",
            "[50,    10] acc: 0.854\n",
            "Finished Training\n",
            "Accuracy of the network on the test images: 67 %\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "3388000.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aplmsob5QGk9",
        "colab_type": "code",
        "outputId": "7f955d81-e6bf-4752-d457-954c946062ca",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 302
        }
      },
      "source": [
        "show_first_layer(net_no_pretrained)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "torch.Size([64, 3, 11, 11])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FJ8KOW5vawLB",
        "colab_type": "code",
        "outputId": "31aa7fd4-d22c-44a3-f34b-85e1c4cc3400",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        }
      },
      "source": [
        "show_first_layer(net)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "torch.Size([64, 3, 11, 11])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZTkfVe_TbECg",
        "colab_type": "text"
      },
      "source": [
        "# 2. VGG"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rWI4c977bHRm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import torch.nn as nn\n",
        "import torch.utils.model_zoo as model_zoo\n",
        "\n",
        "\n",
        "model_urls = {\n",
        "    'vgg11_bn': 'https://download.pytorch.org/models/vgg11_bn-6002323d.pth'\n",
        "}\n",
        "\n",
        "\n",
        "class VGG(nn.Module):\n",
        "\n",
        "    def __init__(self, features, num_classes=1000, init_weights=True):\n",
        "        super(VGG, self).__init__()\n",
        "        self.features = features\n",
        "        self.avgpool = nn.AdaptiveAvgPool2d((7, 7))\n",
        "        self.classifier = nn.Sequential(\n",
        "            nn.Linear(512 * 7 * 7, 4096),\n",
        "            nn.ReLU(True),\n",
        "            nn.Dropout(),\n",
        "            nn.Linear(4096, 4096),\n",
        "            nn.ReLU(True),\n",
        "            nn.Dropout(),\n",
        "            nn.Linear(4096, num_classes)\n",
        "        )\n",
        "        if init_weights:\n",
        "            self._initialize_weights()\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.features(x)\n",
        "        x = self.avgpool(x)\n",
        "        x = x.view(x.size(0), -1)\n",
        "        x = self.classifier(x)\n",
        "        return x\n",
        "\n",
        "    def _initialize_weights(self):\n",
        "        for m in self.modules():\n",
        "            if isinstance(m, nn.Conv2d):\n",
        "                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')\n",
        "                if m.bias is not None:\n",
        "                    nn.init.constant_(m.bias, 0)\n",
        "            elif isinstance(m, nn.BatchNorm2d):\n",
        "                nn.init.constant_(m.weight, 1)\n",
        "                nn.init.constant_(m.bias, 0)\n",
        "            elif isinstance(m, nn.Linear):\n",
        "                nn.init.normal_(m.weight, 0, 0.01)\n",
        "                nn.init.constant_(m.bias, 0)\n",
        "\n",
        "\n",
        "def make_layers(cfg, batch_norm=False):\n",
        "    layers = []\n",
        "    in_chan = 3\n",
        "    for name in cfg :\n",
        "        if type(name) == int:\n",
        "            layers.append(nn.Conv2d(in_chan, name, 3, padding=1))\n",
        "            in_chan = name\n",
        "            if batch_norm :\n",
        "                layers.append(nn.BatchNorm2d(name))\n",
        "            layers.append(nn.ReLU())\n",
        "        elif name == 'M':\n",
        "            layers.append(nn.MaxPool2d(2, stride=2))\n",
        "    return nn.Sequential(*layers)\n",
        "\n",
        "cfg = {\n",
        "    'A': [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M']\n",
        "}"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hPTHyRbTexvE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def vgg_11_classifier(num_classes):\n",
        "  classifier = nn.Sequential(\n",
        "            nn.Dropout(),\n",
        "            nn.Linear(512 * 7 * 7, 128),\n",
        "            nn.ReLU(inplace=True),\n",
        "            nn.Dropout(),\n",
        "            nn.Linear(128, 64),\n",
        "            nn.ReLU(inplace=True),\n",
        "            nn.Linear(64, num_classes),\n",
        "        )\n",
        "  return classifier\n",
        "  \n",
        "def vgg11(num_classes, pretrained=False, **kwargs):\n",
        "    \"\"\"VGG 11-layer model (configuration \"A\") with batch normalization\n",
        "    Args:\n",
        "        pretrained (bool): If True, returns a model pre-trained on ImageNet\n",
        "    \"\"\"\n",
        "    if pretrained:\n",
        "        kwargs['init_weights'] = False\n",
        "    model = VGG(make_layers(cfg['A'], batch_norm=True), **kwargs)\n",
        "    if pretrained:\n",
        "        model.load_state_dict(model_zoo.load_url(model_urls['vgg11_bn']))\n",
        "        for p in model.features.parameters():\n",
        "            p.requires_grad=False\n",
        "    model.classifier = vgg_11_classifier(num_classes)\n",
        "    return model"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WJTSE7c8g8fx",
        "colab_type": "code",
        "outputId": "634a8638-538d-4a90-faeb-b367ddd549af",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "vgg_no_pretrained = vgg11(num_classes=10, pretrained=False)\n",
        "\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    vgg_no_pretrained.cuda()\n",
        "learning_rate = 1e-3\n",
        "optimizer = torch.optim.Adam(vgg_no_pretrained.parameters(),lr=learning_rate)\n",
        "train(vgg_no_pretrained, optimizer, trainloader, testloader, criterion,  n_epoch = 50,\n",
        "      train_acc_period = 10,\n",
        "      test_acc_period = 1000)\n",
        "accuracy(vgg_no_pretrained, testloader, cuda=use_cuda)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "using cuda\n",
            "[1,    10] loss: 0.035\n",
            "[1,    10] acc: 0.129\n",
            "[2,    10] loss: 0.032\n",
            "[2,    10] acc: 0.162\n",
            "[3,    10] loss: 0.031\n",
            "[3,    10] acc: 0.174\n",
            "[4,    10] loss: 0.031\n",
            "[4,    10] acc: 0.196\n",
            "[5,    10] loss: 0.030\n",
            "[5,    10] acc: 0.206\n",
            "[6,    10] loss: 0.030\n",
            "[6,    10] acc: 0.239\n",
            "[7,    10] loss: 0.030\n",
            "[7,    10] acc: 0.239\n",
            "[8,    10] loss: 0.029\n",
            "[8,    10] acc: 0.238\n",
            "[9,    10] loss: 0.028\n",
            "[9,    10] acc: 0.244\n",
            "[10,    10] loss: 0.029\n",
            "[10,    10] acc: 0.224\n",
            "[11,    10] loss: 0.028\n",
            "[11,    10] acc: 0.241\n",
            "[12,    10] loss: 0.027\n",
            "[12,    10] acc: 0.268\n",
            "[13,    10] loss: 0.028\n",
            "[13,    10] acc: 0.279\n",
            "[14,    10] loss: 0.026\n",
            "[14,    10] acc: 0.294\n",
            "[15,    10] loss: 0.027\n",
            "[15,    10] acc: 0.275\n",
            "[16,    10] loss: 0.026\n",
            "[16,    10] acc: 0.287\n",
            "[17,    10] loss: 0.027\n",
            "[17,    10] acc: 0.276\n",
            "[18,    10] loss: 0.026\n",
            "[18,    10] acc: 0.293\n",
            "[19,    10] loss: 0.024\n",
            "[19,    10] acc: 0.353\n",
            "[20,    10] loss: 0.026\n",
            "[20,    10] acc: 0.323\n",
            "[21,    10] loss: 0.024\n",
            "[21,    10] acc: 0.353\n",
            "[22,    10] loss: 0.026\n",
            "[22,    10] acc: 0.285\n",
            "[23,    10] loss: 0.024\n",
            "[23,    10] acc: 0.300\n",
            "[24,    10] loss: 0.024\n",
            "[24,    10] acc: 0.351\n",
            "[25,    10] loss: 0.023\n",
            "[25,    10] acc: 0.367\n",
            "[26,    10] loss: 0.025\n",
            "[26,    10] acc: 0.297\n",
            "[27,    10] loss: 0.024\n",
            "[27,    10] acc: 0.358\n",
            "[28,    10] loss: 0.023\n",
            "[28,    10] acc: 0.341\n",
            "[29,    10] loss: 0.022\n",
            "[29,    10] acc: 0.387\n",
            "[30,    10] loss: 0.022\n",
            "[30,    10] acc: 0.373\n",
            "[31,    10] loss: 0.022\n",
            "[31,    10] acc: 0.358\n",
            "[32,    10] loss: 0.021\n",
            "[32,    10] acc: 0.416\n",
            "[33,    10] loss: 0.022\n",
            "[33,    10] acc: 0.405\n",
            "[34,    10] loss: 0.021\n",
            "[34,    10] acc: 0.421\n",
            "[35,    10] loss: 0.020\n",
            "[35,    10] acc: 0.412\n",
            "[36,    10] loss: 0.019\n",
            "[36,    10] acc: 0.436\n",
            "[37,    10] loss: 0.019\n",
            "[37,    10] acc: 0.447\n",
            "[38,    10] loss: 0.020\n",
            "[38,    10] acc: 0.432\n",
            "[39,    10] loss: 0.019\n",
            "[39,    10] acc: 0.421\n",
            "[40,    10] loss: 0.019\n",
            "[40,    10] acc: 0.418\n",
            "[41,    10] loss: 0.019\n",
            "[41,    10] acc: 0.431\n",
            "[42,    10] loss: 0.019\n",
            "[42,    10] acc: 0.425\n",
            "[43,    10] loss: 0.019\n",
            "[43,    10] acc: 0.440\n",
            "[44,    10] loss: 0.022\n",
            "[44,    10] acc: 0.389\n",
            "[45,    10] loss: 0.019\n",
            "[45,    10] acc: 0.410\n",
            "[46,    10] loss: 0.019\n",
            "[46,    10] acc: 0.433\n",
            "[47,    10] loss: 0.018\n",
            "[47,    10] acc: 0.450\n",
            "[48,    10] loss: 0.017\n",
            "[48,    10] acc: 0.462\n",
            "[49,    10] loss: 0.017\n",
            "[49,    10] acc: 0.483\n",
            "[50,    10] loss: 0.017\n",
            "[50,    10] acc: 0.497\n",
            "Finished Training\n",
            "Accuracy of the network on the test images: 40 %\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2010100.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RexrvJxlhUA6",
        "colab_type": "code",
        "outputId": "5288ee60-60b7-45d1-da36-ceff54beb034",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000,
          "referenced_widgets": [
            "b02c7733cfbe46638df97d3df56a9306",
            "f2f213c29b3e44549ba1072d707cf361",
            "fb75d7e9be574831a4772810cc17c7e7",
            "50e5fe6003a54a2fb85af9ad74b048cd",
            "51a8ab02847740e7a0308f7dde72f08f",
            "aa63e2d9761f4137b8300144ab092686",
            "1717f68be248410d9b27cc5ded0fe851",
            "d841f87f62dc45caba71526224d383e7"
          ]
        }
      },
      "source": [
        "vgg = vgg11(num_classes=10, pretrained=True)\n",
        "\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    vgg.cuda()\n",
        "learning_rate = 1e-3\n",
        "optimizer = torch.optim.Adam(vgg.parameters(),lr=learning_rate)\n",
        "train(vgg, optimizer, trainloader, testloader, criterion,  n_epoch = 50,\n",
        "      train_acc_period = 10,\n",
        "      test_acc_period = 1000)\n",
        "accuracy(vgg, testloader, cuda=use_cuda)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading: \"https://download.pytorch.org/models/vgg11_bn-6002323d.pth\" to /root/.cache/torch/checkpoints/vgg11_bn-6002323d.pth\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "b02c7733cfbe46638df97d3df56a9306",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=531503671), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "using cuda\n",
            "[1,    10] loss: 0.032\n",
            "[1,    10] acc: 0.216\n",
            "[2,    10] loss: 0.019\n",
            "[2,    10] acc: 0.530\n",
            "[3,    10] loss: 0.011\n",
            "[3,    10] acc: 0.669\n",
            "[4,    10] loss: 0.007\n",
            "[4,    10] acc: 0.753\n",
            "[5,    10] loss: 0.004\n",
            "[5,    10] acc: 0.800\n",
            "[6,    10] loss: 0.003\n",
            "[6,    10] acc: 0.843\n",
            "[7,    10] loss: 0.002\n",
            "[7,    10] acc: 0.837\n",
            "[8,    10] loss: 0.001\n",
            "[8,    10] acc: 0.853\n",
            "[9,    10] loss: 0.001\n",
            "[9,    10] acc: 0.862\n",
            "[10,    10] loss: 0.001\n",
            "[10,    10] acc: 0.856\n",
            "[11,    10] loss: 0.001\n",
            "[11,    10] acc: 0.858\n",
            "[12,    10] loss: 0.000\n",
            "[12,    10] acc: 0.866\n",
            "[13,    10] loss: 0.000\n",
            "[13,    10] acc: 0.865\n",
            "[14,    10] loss: 0.000\n",
            "[14,    10] acc: 0.866\n",
            "[15,    10] loss: 0.000\n",
            "[15,    10] acc: 0.864\n",
            "[16,    10] loss: 0.000\n",
            "[16,    10] acc: 0.867\n",
            "[17,    10] loss: 0.000\n",
            "[17,    10] acc: 0.863\n",
            "[18,    10] loss: 0.000\n",
            "[18,    10] acc: 0.865\n",
            "[19,    10] loss: 0.000\n",
            "[19,    10] acc: 0.865\n",
            "[20,    10] loss: 0.000\n",
            "[20,    10] acc: 0.866\n",
            "[21,    10] loss: 0.000\n",
            "[21,    10] acc: 0.865\n",
            "[22,    10] loss: 0.000\n",
            "[22,    10] acc: 0.864\n",
            "[23,    10] loss: 0.000\n",
            "[23,    10] acc: 0.868\n",
            "[24,    10] loss: 0.000\n",
            "[24,    10] acc: 0.868\n",
            "[25,    10] loss: 0.000\n",
            "[25,    10] acc: 0.867\n",
            "[26,    10] loss: 0.000\n",
            "[26,    10] acc: 0.869\n",
            "[27,    10] loss: 0.000\n",
            "[27,    10] acc: 0.868\n",
            "[28,    10] loss: 0.000\n",
            "[28,    10] acc: 0.864\n",
            "[29,    10] loss: 0.000\n",
            "[29,    10] acc: 0.868\n",
            "[30,    10] loss: 0.000\n",
            "[30,    10] acc: 0.868\n",
            "[31,    10] loss: 0.000\n",
            "[31,    10] acc: 0.869\n",
            "[32,    10] loss: 0.000\n",
            "[32,    10] acc: 0.869\n",
            "[33,    10] loss: 0.000\n",
            "[33,    10] acc: 0.866\n",
            "[34,    10] loss: 0.000\n",
            "[34,    10] acc: 0.866\n",
            "[35,    10] loss: 0.000\n",
            "[35,    10] acc: 0.869\n",
            "[36,    10] loss: 0.000\n",
            "[36,    10] acc: 0.868\n",
            "[37,    10] loss: 0.000\n",
            "[37,    10] acc: 0.863\n",
            "[38,    10] loss: 0.000\n",
            "[38,    10] acc: 0.869\n",
            "[39,    10] loss: 0.000\n",
            "[39,    10] acc: 0.869\n",
            "[40,    10] loss: 0.000\n",
            "[40,    10] acc: 0.869\n",
            "[41,    10] loss: 0.000\n",
            "[41,    10] acc: 0.869\n",
            "[42,    10] loss: 0.000\n",
            "[42,    10] acc: 0.869\n",
            "[43,    10] loss: 0.000\n",
            "[43,    10] acc: 0.868\n",
            "[44,    10] loss: 0.000\n",
            "[44,    10] acc: 0.869\n",
            "[45,    10] loss: 0.000\n",
            "[45,    10] acc: 0.865\n",
            "[46,    10] loss: 0.000\n",
            "[46,    10] acc: 0.868\n",
            "[47,    10] loss: 0.000\n",
            "[47,    10] acc: 0.866\n",
            "[48,    10] loss: 0.000\n",
            "[48,    10] acc: 0.864\n",
            "[49,    10] loss: 0.000\n",
            "[49,    10] acc: 0.867\n",
            "[50,    10] loss: 0.000\n",
            "[50,    10] acc: 0.866\n",
            "Finished Training\n",
            "Accuracy of the network on the test images: 72 %\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "3625700.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UDsT71vsw_gv",
        "colab_type": "code",
        "outputId": "0efd7090-9f25-45fc-ab56-12357a1b4737",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        }
      },
      "source": [
        "show_first_layer(vgg_no_pretrained)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "torch.Size([64, 3, 3, 3])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EnajamXuxMf-",
        "colab_type": "code",
        "outputId": "cabdf42f-0477-4706-ab5c-5b3aaf0f5c85",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        }
      },
      "source": [
        "show_first_layer(vgg)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "torch.Size([64, 3, 3, 3])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "n9rKSbCImw2d",
        "colab_type": "text"
      },
      "source": [
        "# 3. ResNet"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fXIz0oLkUMIt",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def trainer(train_loader, test_loader, model, use_cuda, loss_fn=nn.CrossEntropyLoss(), epoch=10, rate=1e-3):\n",
        "    loss_eval = []\n",
        "    err_eval = []\n",
        "    #   data = torch.utils.data.DataLoader(dataset, batch_size)\n",
        "    optimizer = torch.optim.Adam(model.parameters(),lr=rate)\n",
        "    j = 0\n",
        "    for i in tqdm(range(epoch)):\n",
        "        for imgs, labels in train_loader:\n",
        "            j += 1\n",
        "            if use_cuda and torch.cuda.is_available():\n",
        "                imgs = imgs.type(torch.cuda.FloatTensor)\n",
        "                labels = labels.type(torch.cuda.LongTensor)\n",
        "\n",
        "            optimizer.zero_grad()\n",
        "\n",
        "            output = model(imgs)\n",
        "            loss = loss_fn(output, labels)\n",
        "            if j % 500 == 1 :\n",
        "                loss_eval.append(loss.item())\n",
        "                err_eval.append(1 - accuracy(test_loader, model, use_cuda))\n",
        "            loss.backward()\n",
        "\n",
        "            optimizer.step()\n",
        "        \n",
        "    return loss_eval, err_eval\n",
        "\n",
        "def accuracy(dataloader, model, use_cuda):\n",
        "    model.eval()\n",
        "    #   test_loader = torch.utils.data.DataLoader(test_data, batch_size=len(test_data))\n",
        "    corr = 0\n",
        "    total = 0\n",
        "    for imgs, labels in dataloader:\n",
        "        if use_cuda and torch.cuda.is_available():\n",
        "            imgs = imgs.type(torch.cuda.FloatTensor)\n",
        "            labels = labels.type(torch.cuda.LongTensor)\n",
        "        with torch.no_grad():\n",
        "            output = model(imgs)\n",
        "            output = [torch.argmax(i) for i in output]\n",
        "            correct = [i for i,j in zip(output, labels) if i == j]\n",
        "            corr += len(correct)\n",
        "            total += len(labels)\n",
        "    model.train()\n",
        "    # print(len(correct))\n",
        "    return(corr / total)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7BqTSz6Bm1Aw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "\n",
        "\n",
        "class SimpleBlock(nn.Module):\n",
        "    def __init__(self, in_planes, planes, stride=1):\n",
        "        super(SimpleBlock, self).__init__()\n",
        "\n",
        "        assert((stride == 1) or (stride == 2 and planes == in_planes * 2))\n",
        "\n",
        "        padding = 0\n",
        "        self.conv1 = nn.Conv2d(in_planes, planes, 3, stride=stride, padding=1, bias=False)\n",
        "        self.bn1 = nn.BatchNorm2d(planes)\n",
        "        self.conv2 = nn.Conv2d(planes, planes, 3, stride=1, padding=1, bias=False)\n",
        "        self.bn2 = nn.BatchNorm2d(planes)\n",
        "\n",
        "    def forward(self, x):\n",
        "        out = F.relu(self.bn1(self.conv1(x)))\n",
        "        out = self.conv2(out)\n",
        "        out = F.relu(self.bn2(out))\n",
        "\n",
        "        return out\n",
        "\n",
        "      \n",
        "class ResNetBlock(nn.Module):\n",
        "\n",
        "    def __init__(self, in_planes, planes, stride=1):\n",
        "        super(ResNetBlock, self).__init__()\n",
        "\n",
        "        assert((stride == 1) or (stride == 2 and planes == in_planes * 2))\n",
        "\n",
        "        padding = 0\n",
        "        self.conv1 = nn.Conv2d(in_planes, planes, 3, stride=stride, padding=1, bias=False)\n",
        "        self.bn1 = nn.BatchNorm2d(planes)\n",
        "        self.conv2 = nn.Conv2d(planes, planes, 3, stride=1, padding=1, bias=False)\n",
        "        if stride == 2: \n",
        "            self.shortcut = nn.Conv2d(in_planes, planes, 1, stride=2, padding=0, bias=False)\n",
        "        else: \n",
        "            self.shortcut = nn.Identity()\n",
        "        self.bn2 = nn.BatchNorm2d(planes)\n",
        "\n",
        "    def forward(self, x):\n",
        "        out = F.relu(self.bn1(self.conv1(x)))\n",
        "        out = self.conv2(out) + self.shortcut(x)\n",
        "        out = F.relu(self.bn2(out))\n",
        "\n",
        "        return out\n",
        "\n",
        "class ResNet(nn.Module):\n",
        "    def __init__(self, block, num_blocks, num_classes=10, num_filters=16, input_dim=3):\n",
        "        super(ResNet, self).__init__()\n",
        "        self.in_planes = num_filters\n",
        "\n",
        "        self.conv1 = nn.Conv2d(input_dim, num_filters, kernel_size=3, stride=1, padding=1, bias=False)\n",
        "        self.bn1 = nn.BatchNorm2d(num_filters)\n",
        "\t\t\n",
        "        layers_resnet = []\n",
        "        for i in range(len(num_blocks)) :\n",
        "            if i == 0 :\n",
        "                planes = self.in_planes\n",
        "                print(planes)\n",
        "                layers_resnet.append(self._make_layer(block, planes, num_blocks[i], stride=1))\n",
        "            else :\n",
        "                planes = self.in_planes * 2\n",
        "                print(planes)\n",
        "                layers_resnet.append(self._make_layer(block, planes, num_blocks[i], stride=2))\n",
        "\n",
        "        self.resnet = nn.Sequential(*layers_resnet)\n",
        "\t\t\n",
        "        self.fc = nn.Sequential(\n",
        "            nn.Linear(self.in_planes, num_classes),\n",
        "            nn.LogSoftmax(1)\n",
        "        )\n",
        "\n",
        "    def _make_layer(self, block, planes, num_blocks, stride):\n",
        "        layers = []\n",
        "\n",
        "        if stride != 1 :\n",
        "            downsample = block(self.in_planes, planes, stride)\n",
        "            layers.append(downsample)\n",
        "            num_blocks -= 1\n",
        "            self.in_planes = planes\n",
        "\n",
        "        for i in range(num_blocks) :\n",
        "            layers.append(block(self.in_planes, planes, 1))\n",
        "            self.in_planes = planes\n",
        "        \n",
        "        return nn.Sequential(*layers)\n",
        "\n",
        "    def forward(self, x):\n",
        "        out = F.relu(self.bn1(self.conv1(x)))\n",
        "        out = self.resnet(out)\n",
        "        out = F.adaptive_avg_pool2d(out, (1,1))\n",
        "        out = out.view(out.size(0), -1)\n",
        "        out = self.fc(out)\n",
        "        return out\n",
        "\n",
        "\n",
        "# (1 + 2*(1 + 1) + 2*(1 + 1) + 2*(1 + 1) + 2*(1 + 1)) + 1 = 18\n",
        "def ResNet18(input_dim=3):\n",
        "    return ResNet(ResNetBlock, [6, 6, 6], input_dim=input_dim)\n",
        "\n",
        "def PlainNet18(input_dim=3):\n",
        "    return ResNet(SimpleBlock, [6, 6, 6], input_dim=input_dim)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aWZQOsCfTm9D",
        "colab_type": "text"
      },
      "source": [
        "## On CIFAR-10"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fv0YOntO4EQl",
        "colab_type": "code",
        "outputId": "d5c94f83-7e11-4835-ae0f-ec6880232c56",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "batch_size=64\n",
        "transform = transforms.Compose(\n",
        "    [\n",
        "     transforms.ToTensor(),\n",
        "     transforms.Normalize((0.5,), (0.5,)),\n",
        "])\n",
        "trainset = torchvision.datasets.CIFAR10(root=\"./data\", train=True, download=True, transform=transform)\n",
        "trainloader = torch.utils.data.DataLoader(trainset,batch_size=batch_size,shuffle=False)\n",
        "testset = torchvision.datasets.CIFAR10(root=\"./data\", train=False, download=True, transform=transform)\n",
        "testloader = torch.utils.data.DataLoader(testset,batch_size=10000,shuffle=False)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Files already downloaded and verified\n",
            "Files already downloaded and verified\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qQGOG7dZitlZ",
        "colab_type": "code",
        "outputId": "96292b34-b0ef-4c3e-aa11-75aeb7c24e22",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 151,
          "referenced_widgets": [
            "f07a6b9698d841698c38abcfc4017653",
            "d50821987164451383836b569c077b0c",
            "f4e8e515f4da4911ad72f9578323555f",
            "65ab8ca766f44de9b855f287827dd1db",
            "f7c2a6b6114f415f8e51649b7f7682c0",
            "3433a9abb9fb4527890283c2e6fdac3c",
            "9ada21cf2880481bba50cc424642e5f3",
            "8520902a82dc42adabd81004f1f71df2"
          ]
        }
      },
      "source": [
        "model_resnet18 = ResNet18()\n",
        "model_resnet18.apply(init)\n",
        "\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    model_resnet18.cuda()\n",
        "losses, val_errs = trainer(trainloader, testloader, model_resnet18, use_cuda, epoch=10)\n",
        "print(accuracy(testloader, model_resnet18, use_cuda))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "16\n",
            "32\n",
            "64\n",
            "using cuda\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "f07a6b9698d841698c38abcfc4017653",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=10), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "0.7337\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sCnlwU266xJf",
        "colab_type": "code",
        "outputId": "dc6294df-0c3d-4adf-ff57-df1550e34da1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 151,
          "referenced_widgets": [
            "46d7bbaccad0458ab52d5e44ad6aeac4",
            "3e36bcdc654d4b3db4ce02f57dda70d8",
            "d026ff66c2a2494fa434e9fb2679d191",
            "76b9c5fa459840a78ab2dd059c7612c7",
            "67df71fcc551486b9ef5642102d1b484",
            "a05c20fe5e204169acb751f87ee64acd",
            "a281c4933fcc4bf8a01b3a5baf274c9f",
            "82977c4a15fb4d148537a14d25c626cb"
          ]
        }
      },
      "source": [
        "model_plainnet18 = PlainNet18()\n",
        "model_plainnet18.apply(init)\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    model_plainnet18.cuda()\n",
        "losses_, val_errs_ = trainer(trainloader, testloader, model_plainnet18, use_cuda, epoch=10)\n",
        "print(accuracy(testloader, model_plainnet18, use_cuda))\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "16\n",
            "32\n",
            "64\n",
            "using cuda\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "46d7bbaccad0458ab52d5e44ad6aeac4",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=10), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "0.6632\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CNaWaWJxBXCB",
        "colab_type": "code",
        "outputId": "0dc17460-c6ba-4aa6-d506-515376b3d8a5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        }
      },
      "source": [
        "plt.plot(losses, label='ResNet training loss')\n",
        "plt.plot(val_errs, label='ResNet validation error')\n",
        "plt.plot(losses_, label='Plainnet training loss')\n",
        "plt.plot(val_errs_, label='Plainnet validation error')\n",
        "plt.legend()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fe2a0259cf8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "lPmg588UT7au"
      },
      "source": [
        "## On FashionMNIST"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "outputId": "f100d1cc-d66f-4cec-9042-d71bb0b9feaf",
        "id": "3EpULhPSUApK",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369,
          "referenced_widgets": [
            "8dc6c54e8c5f4ad99f14d05144505a9c",
            "dbd533d84cb24e8994983e118c68167b",
            "5bbecd4ebe214721b47fb9140d9e23b5",
            "93c6f9402bc54c88a68994220bd6043f",
            "1cc41ec45ae64da1b1db581373fd88e1",
            "e915be758aa8423dbe138cf6f781b8e3",
            "27ed6d5eab2148b2b3030299ec99c76c",
            "50202f20508545e79d13fca8d3606001",
            "e433bd147d4c443f998e61fa389551cc",
            "cd901762a5784799b4e55d1d44c871a8",
            "25631a6c8663440782fa6934fcc469a8",
            "d52cb9d568eb4f699bd3914e1f7fcb9b",
            "35c150ba05d74ec39e8e37522062c46c",
            "8e3fe029227d4400b4b08d97ea743a56",
            "7ed5c59c9d7d4911865ae879985f408e",
            "af59ac319e324358a2d956064deeb5c8",
            "b53601df547f458cbb2f85be3a7bf973",
            "788802b31c7c4725885c8bf74af760e6",
            "bdbe8136586d476c895307ef2bdd1229",
            "df908da626ea4fb1a5d27b775b49b850",
            "4ce3ab4979604731bdf1a167d34df045",
            "f5b8780f065747c5930550119832f3d4",
            "7e6e1c9320524f83960cd598cce01e9e",
            "87b4dd731fc745e689fe2317e9e7bc4b",
            "6c658ea12ad1459dbe1692093ae7373f",
            "9e0156f13c314988b65e9209d9dba79a",
            "de2f3b7035744258a9cc11b96bdd6c3c",
            "d494d3201e1e42b7ae4fdb26460396a8",
            "ffda2f4c8da1416f8e3b43d00f0a7906",
            "0e844fca3d594e43804e776d7e08ca5d",
            "8185daed45054680bca8dd9fc9781730",
            "202a3bd512564a1e80095467812c60eb"
          ]
        }
      },
      "source": [
        "batch_size=64\n",
        "transform = transforms.Compose(\n",
        "    [\n",
        "     transforms.ToTensor()\n",
        "])\n",
        "trainset = torchvision.datasets.FashionMNIST(root=\"./data\", train=True, download=True, transform=transform)\n",
        "trainloader = torch.utils.data.DataLoader(trainset,batch_size=batch_size,shuffle=False)\n",
        "testset = torchvision.datasets.FashionMNIST(root=\"./data\", train=False, download=True, transform=transform)\n",
        "testloader = torch.utils.data.DataLoader(testset,batch_size=batch_size,shuffle=False)\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz to ./data/FashionMNIST/raw/train-images-idx3-ubyte.gz\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "8dc6c54e8c5f4ad99f14d05144505a9c",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Extracting ./data/FashionMNIST/raw/train-images-idx3-ubyte.gz to ./data/FashionMNIST/raw\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz to ./data/FashionMNIST/raw/train-labels-idx1-ubyte.gz\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "e433bd147d4c443f998e61fa389551cc",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Extracting ./data/FashionMNIST/raw/train-labels-idx1-ubyte.gz to ./data/FashionMNIST/raw\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz to ./data/FashionMNIST/raw/t10k-images-idx3-ubyte.gz\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "b53601df547f458cbb2f85be3a7bf973",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "Extracting ./data/FashionMNIST/raw/t10k-images-idx3-ubyte.gz to ./data/FashionMNIST/raw\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz to ./data/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "6c658ea12ad1459dbe1692093ae7373f",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Extracting ./data/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz to ./data/FashionMNIST/raw\n",
            "Processing...\n",
            "Done!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "outputId": "0140348d-4a41-4bd7-d617-93aa51e4170f",
        "id": "4k9Cnh1AUTK9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 168,
          "referenced_widgets": [
            "7a38788e9405460dbfb9a81ddfdafdd6",
            "131f96954d23497f9705235384f32714",
            "376dd3f3010b4c24b8b864df14c7fa77",
            "b8af58d01d9748c2bfa2cde80e6748c4",
            "d251bfd320284028afcdf07d63f1f87d",
            "155bdd082cea403f8e04db76c83c0a0c",
            "324d740bd6414f0d84fe1029a9b29240",
            "56d3a8457f4b4b7f8723f2ae58e43847"
          ]
        }
      },
      "source": [
        "model_resnet18 = ResNet18(1)\n",
        "model_resnet18.apply(init)\n",
        "\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    model_resnet18.cuda()\n",
        "losses, val_errs = trainer(trainloader, testloader, model_resnet18, use_cuda, epoch=10)\n",
        "print(accuracy(testloader, model_resnet18, use_cuda))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "16\n",
            "32\n",
            "\n",
            "64\n",
            "using cuda\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "7a38788e9405460dbfb9a81ddfdafdd6",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=10), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "0.9117\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "outputId": "74a2fde1-ee17-435f-98ee-611b63f18084",
        "id": "XnhaSaA3UW2e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 151,
          "referenced_widgets": [
            "1b61b9178bd1405197ab6e852903e97f",
            "7f1840aa27864dea8df473889e4fa437",
            "bdb74454631b454bb33163b34928c80d",
            "29ecf9fd4aa141569e940c23961e46fd",
            "806b3a1ce73147e3a4c47b8e291a2244",
            "6b8d03fb7d43417eb8a9b9e0cd277f53",
            "cf17fcd276124a8e9cb4b919ccd75b82",
            "9f05380c02fa4717817d3953b3f7c435"
          ]
        }
      },
      "source": [
        "model_plainnet18 = PlainNet18(1)\n",
        "model_plainnet18.apply(init)\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    model_plainnet18.cuda()\n",
        "losses_, val_errs_ = trainer(trainloader, testloader, model_plainnet18, use_cuda, epoch=10)\n",
        "print(accuracy(testloader, model_plainnet18, use_cuda))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "16\n",
            "32\n",
            "64\n",
            "using cuda\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "1b61b9178bd1405197ab6e852903e97f",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=10), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "0.9084\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DRA1WR_KBzjW",
        "colab_type": "code",
        "outputId": "2950e47d-2f8e-45c5-9b0a-c5c0323ea1cb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        }
      },
      "source": [
        "plt.plot(losses, label='ResNet training loss')\n",
        "plt.plot(val_errs, label='ResNet validation error')\n",
        "plt.plot(losses_, label='Plainnet training loss')\n",
        "plt.plot(val_errs_, label='Plainnet validation error')\n",
        "plt.legend()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fe2a0347588>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2aJ34FjC6VXp",
        "colab_type": "text"
      },
      "source": [
        "The result is a lot better than a usual plain network. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CV1qfvIuQo8J",
        "colab_type": "text"
      },
      "source": [
        "# 4. InceptionNet"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JpF7rZIpQwvz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "\n",
        "\n",
        "class Inception(nn.Module):\n",
        "    def __init__(self, in_planes, kernel_1_x, kernel_3_in, kernel_3_x, kernel_5_in, kernel_5_x, pool_planes):\n",
        "        super(Inception, self).__init__()\n",
        "        self.conv1 = nn.Sequential(\n",
        "              nn.Conv2d(in_planes, kernel_1_x, 1),\n",
        "              nn.BatchNorm2d(kernel_1_x),\n",
        "              nn.ReLU()\n",
        "        )\n",
        "        self.conv3 = nn.Sequential(\n",
        "            nn.Conv2d(in_planes, kernel_3_in, 1),\n",
        "            nn.BatchNorm2d(kernel_3_in),\n",
        "            nn.ReLU(),\n",
        "            nn.Conv2d(kernel_3_in, kernel_3_x, 3, padding=1),\n",
        "            nn.BatchNorm2d(kernel_3_x),\n",
        "            nn.ReLU()\n",
        "        )\n",
        "        self.conv5 = nn.Sequential(\n",
        "            nn.Conv2d(in_planes, kernel_5_in, 1),\n",
        "            nn.BatchNorm2d(kernel_5_in),\n",
        "            nn.ReLU(),\n",
        "            nn.Conv2d(kernel_5_in, kernel_5_x, 3, padding=1),\n",
        "            nn.BatchNorm2d(kernel_5_x),\n",
        "            nn.ReLU()\n",
        "        )\n",
        "        self.pool = nn.Sequential(\n",
        "            nn.MaxPool2d(3, stride=1, padding=1),\n",
        "            nn.Conv2d(in_planes, pool_planes, 1),\n",
        "            nn.BatchNorm2d(pool_planes),\n",
        "            nn.ReLU()\n",
        "        )\n",
        "\n",
        "    def forward(self, x):\n",
        "        # print(self.conv1(x).shape)\n",
        "        # print(self.conv3(x).shape)\n",
        "        out = self.conv1(x)\n",
        "        out = torch.cat((out, self.conv3(x)), 1)\n",
        "        # print(out.shape)\n",
        "        out = torch.cat((out, self.conv5(x)), 1)\n",
        "        # print(out.shape)\n",
        "        return torch.cat((out, self.pool(x)), 1)\n",
        "\n",
        "\n",
        "class GoogLeNet(nn.Module):\n",
        "    def __init__(self, input_dim=3):\n",
        "        super(GoogLeNet, self).__init__()\n",
        "        self.pre_layers = nn.Sequential(\n",
        "            nn.Conv2d(input_dim, 192, kernel_size=3, padding=1),\n",
        "            nn.BatchNorm2d(192),\n",
        "            nn.ReLU(True),\n",
        "        )\n",
        "        \n",
        "        self.layer1 = Inception(192,  64,  96, 128, 16, 32, 32)\n",
        "        \n",
        "        self.layer2 = Inception(256, 128, 128, 192, 32, 96, 64)\n",
        "        \n",
        "        self.layer3 = Inception(480, 192,  96, 208, 16,  48,  64)\n",
        "        \n",
        "        self.max_pool = nn.MaxPool2d(3, stride=2, padding=1)\n",
        "        \n",
        "        self.avgpool = nn.AdaptiveAvgPool2d((1,1))\n",
        "        self.linear = nn.Linear(512, 10)\n",
        "        \n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.pre_layers(x)\n",
        "        x = self.layer1(x)\n",
        "        x = self.max_pool(x)\n",
        "        x = self.layer2(x)\n",
        "        x = self.max_pool(x)\n",
        "        x = self.layer3(x)\n",
        "        x = self.avgpool(x)\n",
        "        x = x.view(x.size(0), -1)\n",
        "        x = self.linear(x)\n",
        "        return x"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HWKV5Qc2g8Fh",
        "colab_type": "text"
      },
      "source": [
        "## On CIFAR-10"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "outputId": "579704b5-bedc-4959-cb0b-872ea8db354a",
        "id": "MObmVQy2geVK",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "batch_size=64\n",
        "transform = transforms.Compose(\n",
        "    [\n",
        "     transforms.ToTensor(),\n",
        "     transforms.Normalize((0.5,), (0.5,)),\n",
        "])\n",
        "trainset = torchvision.datasets.CIFAR10(root=\"./data\", train=True, download=True, transform=transform)\n",
        "trainloader = torch.utils.data.DataLoader(trainset,batch_size=batch_size,shuffle=False)\n",
        "testset = torchvision.datasets.CIFAR10(root=\"./data\", train=False, download=True, transform=transform)\n",
        "testloader = torch.utils.data.DataLoader(testset,batch_size=1000,shuffle=False)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Files already downloaded and verified\n",
            "Files already downloaded and verified\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gP2lyOZbeVcS",
        "colab_type": "code",
        "outputId": "9ba696b2-89d1-4467-e107-fa39e3675928",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 100,
          "referenced_widgets": [
            "a0a9bca1b29d4ee3bffe47f4fe4dc491",
            "d585bad3453c47048ac625a2079b262b",
            "fdf1f6209ad940168c39f33cb621b4a2",
            "9693f7501b524c9ab1580df0aa7feab6",
            "3e25ec94c7fd4117a732ae99b4adeae7",
            "a47912d8d6914632a747921179a64909",
            "507457deba8c462885fa1871defc50dc",
            "2159a93e0a034b6ab25a3ea3a878bcca"
          ]
        }
      },
      "source": [
        "model_googlenet = GoogLeNet()\n",
        "model_googlenet.apply(init)\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    model_googlenet.cuda()\n",
        "losses, val_errs = trainer(trainloader, testloader, model_googlenet, use_cuda, epoch=10)\n",
        "print(accuracy(testloader, model_googlenet, use_cuda))\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "using cuda\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "a0a9bca1b29d4ee3bffe47f4fe4dc491",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=10), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "0.7337\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SJE-zFjNIcxD",
        "colab_type": "code",
        "outputId": "029b2492-373a-4903-fbdd-2848899d2ec6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        }
      },
      "source": [
        "plt.plot(losses, label='GoogLeNet training loss')\n",
        "plt.plot(val_errs, label='GoogLeNet validation error')\n",
        "plt.legend()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fe2a011d828>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qNwmxACshBek",
        "colab_type": "text"
      },
      "source": [
        "## On FashionMNIST"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "RruBZTI7gmCV",
        "colab": {}
      },
      "source": [
        "batch_size=64\n",
        "transform = transforms.Compose(\n",
        "    [\n",
        "     transforms.ToTensor()\n",
        "])\n",
        "trainset = torchvision.datasets.FashionMNIST(root=\"./data\", train=True, download=True, transform=transform)\n",
        "trainloader = torch.utils.data.DataLoader(trainset,batch_size=batch_size,shuffle=False)\n",
        "testset = torchvision.datasets.FashionMNIST(root=\"./data\", train=False, download=True, transform=transform)\n",
        "testloader = torch.utils.data.DataLoader(testset,batch_size=batch_size,shuffle=False)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "outputId": "6e898068-0a83-49c7-cc18-3e9248d122d4",
        "id": "P7zvPYVlgp5N",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 100,
          "referenced_widgets": [
            "01ba15a12a6d4de08f6c57c23b66de21",
            "e82ec39c040b4543945243b12f0ad2c2",
            "e7be80fc0ca9443c917e0cb6a99c89db",
            "6f78ca37d82947898f8036228de320c9",
            "75bdfa6f78b444f9abc0651dce0edc6b",
            "d076df6dcec64b68947a944d1bb86760",
            "d9a91775898b4a5cb3206c257746e696",
            "5f07b8bc85c5420c831f6d82363644a4"
          ]
        }
      },
      "source": [
        "model_googlenet = GoogLeNet(1)\n",
        "model_googlenet.apply(init)\n",
        "#net = AlexNet(10)\n",
        "use_cuda = True\n",
        "if use_cuda and torch.cuda.is_available():\n",
        "    print(\"using cuda\")\n",
        "    model_googlenet.cuda()\n",
        "losses, val_errs = trainer(trainloader, testloader, model_googlenet, use_cuda, epoch=10)\n",
        "print(accuracy(testloader, model_googlenet, use_cuda))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "using cuda\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "01ba15a12a6d4de08f6c57c23b66de21",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=10), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "0.8963\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eCvvx-PlQggo",
        "colab_type": "code",
        "outputId": "b28f61fc-cae1-4cfc-eb97-84d7c87c6e9a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "plt.plot(losses, label='GoogLeNet training loss')\n",
        "plt.plot(val_errs, label='GoogLeNet validation error')\n",
        "plt.legend()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fe2a01cfa20>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}