Modern_CNN_Architectures.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "5839992e885a4c479e5cc2daacd37ff5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_aebc6d898897497d9a770fc85b28a3f5",
              "IPY_MODEL_4138a2eb2bc948e2ada9128fcb585273",
              "IPY_MODEL_ed3de5b1efa8416dbcc38106054be627"
            ],
            "layout": "IPY_MODEL_52ae0b3b49364d798dacf31275b50ed8"
          }
        },
        "aebc6d898897497d9a770fc85b28a3f5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_585f7420a99b4024a33c5bd1b8c18dc8",
            "placeholder": "​",
            "style": "IPY_MODEL_05f7029da78e4d5593336f8d6a86ef8c",
            "value": "Dl Completed...: 100%"
          }
        },
        "4138a2eb2bc948e2ada9128fcb585273": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_23e9e9437dc040a5be551c8336cd2ca2",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_3b02999701274357be9f621bdd101ebd",
            "value": 1
          }
        },
        "ed3de5b1efa8416dbcc38106054be627": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a4ccab6028e64a928be5c039d008500f",
            "placeholder": "​",
            "style": "IPY_MODEL_837c54300daa441596e87974d800c38f",
            "value": " 1/1 [00:10&lt;00:00, 10.41s/ url]"
          }
        },
        "52ae0b3b49364d798dacf31275b50ed8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "585f7420a99b4024a33c5bd1b8c18dc8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "05f7029da78e4d5593336f8d6a86ef8c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "23e9e9437dc040a5be551c8336cd2ca2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "3b02999701274357be9f621bdd101ebd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "a4ccab6028e64a928be5c039d008500f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "837c54300daa441596e87974d800c38f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "f6f32c41c64b41dc95c6e415e458c116": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_9810ab2b2195476fa90d096cd9c6e5e1",
              "IPY_MODEL_7d3546e2261b4ac2af92ea5234fddc29",
              "IPY_MODEL_cac5448e55b94e5b96585c3509a7a5fb"
            ],
            "layout": "IPY_MODEL_1cd1d0fa70a14aed8d149b4cea53342b"
          }
        },
        "9810ab2b2195476fa90d096cd9c6e5e1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2ac5b24d64c74b2a8a9416bba1fa441a",
            "placeholder": "​",
            "style": "IPY_MODEL_3c8d8f6f247f476fa64ee44c5565b220",
            "value": "Dl Size...: 100%"
          }
        },
        "7d3546e2261b4ac2af92ea5234fddc29": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_861816482bda4cb3a0127058b13d685f",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_56b9dde8640c4e08a96ad53952929527",
            "value": 1
          }
        },
        "cac5448e55b94e5b96585c3509a7a5fb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_630c6f79d7c14ca19dd51bb049f50ab3",
            "placeholder": "​",
            "style": "IPY_MODEL_fb849e5baa9a45f7975779b0cdc0af20",
            "value": " 786/786 [00:10&lt;00:00, 82.53 MiB/s]"
          }
        },
        "1cd1d0fa70a14aed8d149b4cea53342b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2ac5b24d64c74b2a8a9416bba1fa441a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3c8d8f6f247f476fa64ee44c5565b220": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "861816482bda4cb3a0127058b13d685f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "56b9dde8640c4e08a96ad53952929527": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "630c6f79d7c14ca19dd51bb049f50ab3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fb849e5baa9a45f7975779b0cdc0af20": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "f3b83543335d4525918140f9fa670781": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_64ac586a9ff14536a1d39e0eed670842",
              "IPY_MODEL_442f413984e54320a73789a139c46c02",
              "IPY_MODEL_792a4571060848c485ffaba1777c1163"
            ],
            "layout": "IPY_MODEL_18df3dd37bc946d9b5001f23fb6e45b6"
          }
        },
        "64ac586a9ff14536a1d39e0eed670842": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ec4314ef910c473e931ef0bcdd592e9c",
            "placeholder": "​",
            "style": "IPY_MODEL_b318329efb214015a925964af9d570b0",
            "value": "Generating splits...: 100%"
          }
        },
        "442f413984e54320a73789a139c46c02": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_679bd91a9a2b449381e8bddb790afef0",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_f0123f7bba0d4d92a877c2f89255a1da",
            "value": 1
          }
        },
        "792a4571060848c485ffaba1777c1163": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_250d77d4571540288c3638a1dbfceb93",
            "placeholder": "​",
            "style": "IPY_MODEL_f227d27a3ad74949a96be5c0b0a55ad9",
            "value": " 1/1 [00:12&lt;00:00, 12.50s/ splits]"
          }
        },
        "18df3dd37bc946d9b5001f23fb6e45b6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "ec4314ef910c473e931ef0bcdd592e9c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b318329efb214015a925964af9d570b0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "679bd91a9a2b449381e8bddb790afef0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f0123f7bba0d4d92a877c2f89255a1da": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "250d77d4571540288c3638a1dbfceb93": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f227d27a3ad74949a96be5c0b0a55ad9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "07d185b4bfcc44f3ad67ff32f6bfed10": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_ea855e833ee74b5ea76631bd2cd9bfff",
              "IPY_MODEL_d18355e2c05d4785bece2b6c65e0252e",
              "IPY_MODEL_a75931b7752c46bbb4e6f93338effac7"
            ],
            "layout": "IPY_MODEL_65d9209db45d48efaaeba2d2b95d314a"
          }
        },
        "ea855e833ee74b5ea76631bd2cd9bfff": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_768d6df02fb745b6af7e4b1340146f9d",
            "placeholder": "​",
            "style": "IPY_MODEL_6d8da11ee3434f22af8fd205da2313b5",
            "value": "Generating train examples...:  91%"
          }
        },
        "d18355e2c05d4785bece2b6c65e0252e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_359e3cca25794f0da63b0ced5f324267",
            "max": 23262,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_96cbc2951f674b4f9b32a6202a18f24c",
            "value": 23262
          }
        },
        "a75931b7752c46bbb4e6f93338effac7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_70efc159082e433499b6c9cb53425e0c",
            "placeholder": "​",
            "style": "IPY_MODEL_d7de298de6ab41fc8aefeefc4ecc6622",
            "value": " 21224/23262 [00:09&lt;00:00, 2356.37 examples/s]"
          }
        },
        "65d9209db45d48efaaeba2d2b95d314a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "768d6df02fb745b6af7e4b1340146f9d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6d8da11ee3434f22af8fd205da2313b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "359e3cca25794f0da63b0ced5f324267": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "96cbc2951f674b4f9b32a6202a18f24c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "70efc159082e433499b6c9cb53425e0c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d7de298de6ab41fc8aefeefc4ecc6622": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "8f74bdcdc1954040a525e86b1e43fcc7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_9caa32ebf38d4793b6ea61d1beb71d45",
              "IPY_MODEL_2e8449772ec34ab3b761ccba95918c7b",
              "IPY_MODEL_33a952d9d784496a9d42c9ecdc122537"
            ],
            "layout": "IPY_MODEL_93a6d40a5dfe497ba1ac847f02564059"
          }
        },
        "9caa32ebf38d4793b6ea61d1beb71d45": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_618b462829044b3e81c4b9bdfb4b0977",
            "placeholder": "​",
            "style": "IPY_MODEL_381d46508b3a46f8859a21491cc78e58",
            "value": "Shuffling /root/tensorflow_datasets/cats_vs_dogs/4.0.0.incompleteAV7M9C/cats_vs_dogs-train.tfrecord*...: 100%"
          }
        },
        "2e8449772ec34ab3b761ccba95918c7b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_481edd60ca6d41bea5a1c3093e6561fc",
            "max": 23262,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_42016a20de984e78b6a808b212a475d7",
            "value": 23262
          }
        },
        "33a952d9d784496a9d42c9ecdc122537": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_843585c448b640279f1ca954025b5777",
            "placeholder": "​",
            "style": "IPY_MODEL_17efa89b1455463fae3bda2fe2812ef9",
            "value": " 23261/23262 [00:02&lt;00:00, 7849.42 examples/s]"
          }
        },
        "93a6d40a5dfe497ba1ac847f02564059": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "618b462829044b3e81c4b9bdfb4b0977": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "381d46508b3a46f8859a21491cc78e58": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "481edd60ca6d41bea5a1c3093e6561fc": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "42016a20de984e78b6a808b212a475d7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "843585c448b640279f1ca954025b5777": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "17efa89b1455463fae3bda2fe2812ef9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Praveen76/Modern-SOTA-CNN-Architectures/blob/main/Modern_CNN_Architectures.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Learning Objectives:\n",
        "\n",
        "At the end of the experiment, you will be able to:\n",
        "1.   Build important blocks for modern CNNs\n",
        "      - Residual connections\n",
        "      - Batch normalisation\n",
        "      - Depthwise separable convolution\n",
        "2.   Interpret what CNNs learn\n",
        "      - Visualising activations\n",
        "      - Visualising filters\n",
        "      - Visualising heatmaps\n"
      ],
      "metadata": {
        "id": "4bpgBd9QYP0_"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 1. Important building blocks for modern CNNs\n",
        "\n",
        "Here we will study about 3 important building blocks:\n",
        "* Residual connection\n",
        "* Batch normalization\n",
        "* Depthwise separable convolution\n",
        "\n"
      ],
      "metadata": {
        "id": "gqTI0D3bYjsJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Residual connections\n",
        "\n",
        "Why do we need them?\n",
        "\n",
        "* CNNs can become extremely deep.\n",
        "* Prone to the vanishing gradient problem\n",
        "\n",
        "Solution:\n",
        "* allow gradients to flow through another shortcut\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "Q7q-wwT5Y7Oh"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/ResidualConnections.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T044830Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=11ab6a5fd6bfaa8dbebe2c67bdbf3ab8169105675cb93bd0bd85ba55e0a2882c)"
      ],
      "metadata": {
        "id": "Ryb9kj388SSs"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Import libraries"
      ],
      "metadata": {
        "id": "7EBDiTOcWlv0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "from tensorflow import keras\n",
        "from tensorflow.keras import layers\n",
        "from tensorflow.keras.utils import plot_model\n",
        "import tensorflow as tf\n",
        "import tensorflow_datasets as tfds"
      ],
      "metadata": {
        "id": "GRCDcQ_VBeAc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Build model with residual connection**"
      ],
      "metadata": {
        "id": "Jjbyoe08y6YU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "inputs = keras.Input(shape=(32, 32, 3), name=\"input\")\n",
        "x = layers.Conv2D(32, 3, activation=\"relu\", padding=\"same\", name=\"C1\")(inputs)  # Q: No. of filters and kernel size? A: 32, 3x3\n",
        "residual = x\n",
        "x = layers.Conv2D(32, 3, activation=\"relu\", padding=\"same\", name=\"C2a\")(x)\n",
        "x = layers.add([x, residual]   )   # Q: Do x and residual have the same shape? A: Yes\n",
        "x = layers.Conv2D(32, 3)(x)\n",
        "model1 = keras.Model(inputs=inputs, outputs=x)\n",
        "plot_model(model1, show_shapes=True)"
      ],
      "metadata": {
        "id": "nYkBqRTP_TJd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "outputId": "69a01687-45c3-4ad2-afb6-c0fea60f5ac0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model1.summary()"
      ],
      "metadata": {
        "id": "be3LFj5gDg3k",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "aecd67d7-274f-41d9-e6ca-43405aa114c0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                Output Shape                 Param #   Connected to                  \n",
            "==================================================================================================\n",
            " input (InputLayer)          [(None, 32, 32, 3)]          0         []                            \n",
            "                                                                                                  \n",
            " C1 (Conv2D)                 (None, 32, 32, 32)           896       ['input[0][0]']               \n",
            "                                                                                                  \n",
            " C2a (Conv2D)                (None, 32, 32, 32)           9248      ['C1[0][0]']                  \n",
            "                                                                                                  \n",
            " add (Add)                   (None, 32, 32, 32)           0         ['C2a[0][0]',                 \n",
            "                                                                     'C1[0][0]']                  \n",
            "                                                                                                  \n",
            " conv2d (Conv2D)             (None, 30, 30, 32)           9248      ['add[0][0]']                 \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 19392 (75.75 KB)\n",
            "Trainable params: 19392 (75.75 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Residual branch may contain 1 layer to make sure addition is possible, i.e. accomodate sizes."
      ],
      "metadata": {
        "id": "De25vNGd-0Ae"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Build model with residual connection - layer in the residual branch\n",
        "\n",
        "inputs = keras.Input(shape=(32, 32, 3), name=\"input\")\n",
        "x = layers.Conv2D(32, 3, activation=\"relu\", name=\"C1\")(inputs)  # Q: No. of filters and kernel size?\n",
        "residual = x\n",
        "x = layers.Conv2D(64, 3, activation=\"relu\", padding=\"same\", name=\"C2l1\")(x)\n",
        "x = layers.Conv2D(64, 3, activation=\"relu\", padding=\"same\", name=\"C2l2\")(x)\n",
        "\n",
        "residual = layers.Conv2D(64, 1, name=\"C2b\")(residual)           # Q: Why 1? A: To maintain same shapes while addition\n",
        "x = layers.add([x, residual]   )                      # Q: Do x and residual have the same shape?\n",
        "model2 = keras.Model(inputs=inputs, outputs=x)"
      ],
      "metadata": {
        "id": "Rjv_A_9PY6A3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_model(model2, show_shapes=True)"
      ],
      "metadata": {
        "id": "9M3gPZRmksDR",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "outputId": "3057f8fc-f91e-4717-eecd-02c130107edb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model2.summary()"
      ],
      "metadata": {
        "id": "gzTDlBrsDrws",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "cf7fa9e4-136a-4a4e-f83e-2be9fb7e1123"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_1\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                Output Shape                 Param #   Connected to                  \n",
            "==================================================================================================\n",
            " input (InputLayer)          [(None, 32, 32, 3)]          0         []                            \n",
            "                                                                                                  \n",
            " C1 (Conv2D)                 (None, 30, 30, 32)           896       ['input[0][0]']               \n",
            "                                                                                                  \n",
            " C2l1 (Conv2D)               (None, 30, 30, 64)           18496     ['C1[0][0]']                  \n",
            "                                                                                                  \n",
            " C2l2 (Conv2D)               (None, 30, 30, 64)           36928     ['C2l1[0][0]']                \n",
            "                                                                                                  \n",
            " C2b (Conv2D)                (None, 30, 30, 64)           2112      ['C1[0][0]']                  \n",
            "                                                                                                  \n",
            " add_1 (Add)                 (None, 30, 30, 64)           0         ['C2l2[0][0]',                \n",
            "                                                                     'C2b[0][0]']                 \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 58432 (228.25 KB)\n",
            "Trainable params: 58432 (228.25 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Important**: Add layers of **same** shape !"
      ],
      "metadata": {
        "id": "HkL86M3HB7Gs"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Q: Can we add layers of shapes (30,30,64) and (30,30,32)?\n",
        "\n",
        "A: No\n",
        "\n",
        "Q: But what happens if you have a pooling layer in between?\n",
        "\n",
        "A: Spatial Dimension reduction due to stride\n",
        "\n",
        "Solution: Use Strides in the  Conv layer in the skip connection."
      ],
      "metadata": {
        "id": "JyQydIMuB8Vw"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Model with residual connections and a max pool layer in between"
      ],
      "metadata": {
        "id": "UMfQcikJ8tqh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# model with residual connections and a max pool layer in between\n",
        "inputs = keras.Input(shape=(32, 32, 3))\n",
        "x = layers.Conv2D(32, 3, activation=\"relu\")(inputs)\n",
        "residual = x\n",
        "x = layers.Conv2D(64, 3, activation=\"relu\", padding=\"same\")(x)\n",
        "x = layers.MaxPooling2D(2, padding=\"same\")(x)  # Max pooling layer reduces the dimsion. #Q: By how much ? A: half\n",
        "residual = layers.Conv2D(64, 1, strides=2)(residual) #Need a stride=2 to accomodate for the maxpool downsampling in the other branch\n",
        "x = layers.add([x, residual])\n",
        "model3 = keras.Model(inputs=inputs, outputs=x)\n",
        "plot_model(model3, show_shapes=True)"
      ],
      "metadata": {
        "id": "5yG1V6Zrml5L",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "outputId": "1800e0dd-fd46-4262-ef65-40baf70c7eb2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model3.summary()"
      ],
      "metadata": {
        "id": "OV4MhFP1Dy5l",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e646fe3f-fa2c-4f59-83e9-c21c23fb4928"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_2\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                Output Shape                 Param #   Connected to                  \n",
            "==================================================================================================\n",
            " input_1 (InputLayer)        [(None, 32, 32, 3)]          0         []                            \n",
            "                                                                                                  \n",
            " conv2d_1 (Conv2D)           (None, 30, 30, 32)           896       ['input_1[0][0]']             \n",
            "                                                                                                  \n",
            " conv2d_2 (Conv2D)           (None, 30, 30, 64)           18496     ['conv2d_1[0][0]']            \n",
            "                                                                                                  \n",
            " max_pooling2d (MaxPooling2  (None, 15, 15, 64)           0         ['conv2d_2[0][0]']            \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2d_3 (Conv2D)           (None, 15, 15, 64)           2112      ['conv2d_1[0][0]']            \n",
            "                                                                                                  \n",
            " add_2 (Add)                 (None, 15, 15, 64)           0         ['max_pooling2d[0][0]',       \n",
            "                                                                     'conv2d_3[0][0]']            \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 21504 (84.00 KB)\n",
            "Trainable params: 21504 (84.00 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "With residual connections, you can build networks of arbitrary depth, without having to worry about vanishing gradients. We will see an example later.\n",
        "\n",
        "**Intuitions on why residual blocks work:**\n",
        "*   Shorter path for gradients\n",
        "*   Easy to learn the identity matrix\n",
        "*   Ensemble of shallow networks"
      ],
      "metadata": {
        "id": "7rycg6y3ehvl"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Batch Normalization**\n",
        "*   Adaptively normalize data even as the mean and variance change over time during training\n",
        "*   During training, it uses the mean and variance of the current batch of data to normalize samples\n",
        "*    During inference (when a big enough batch of representative data may not be available), it uses an exponential moving average of the batch-wise mean and variance of the data seen during training.\n"
      ],
      "metadata": {
        "id": "v9dpxDKLtbkG"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "<div>\n",
        "<img src=\"https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/BatchNormalization.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T044941Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=56ddeee99ba8e8008be5aa69c2df1d7e82d0e44287fb1452d8c2716396bdda04\" width=\"500\"/>\n",
        "</div>"
      ],
      "metadata": {
        "id": "gTucnUKguvcW"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now let's try to calculate the no. of params introduced because of batch normalization in following example."
      ],
      "metadata": {
        "id": "mbssJsObuxz0"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Model with a batch normalization layer"
      ],
      "metadata": {
        "id": "5TClF7208pG7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Model with a batch normalization layer\n",
        "inputs = keras.Input(shape=(32, 32, 3))\n",
        "# Because the output of the Conv2D layer gets normalized, the layer doesn’t need its own bias vector\n",
        "x = layers.Conv2D(32, 3, activation=\"relu\", use_bias=False)(inputs)\n",
        "x = layers.BatchNormalization()(x)\n",
        "x = layers.Conv2D(64, 3, activation=\"relu\", padding=\"same\")(x)\n",
        "model4 = keras.Model(inputs=inputs, outputs=x)\n",
        "plot_model(model4, show_shapes=True)"
      ],
      "metadata": {
        "id": "h5jAYBOip2so",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 422
        },
        "outputId": "f35314d5-483d-4230-fcd5-c81a5a9ef119"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Q: How many paramerters does the BN layer introduce in the above model?\n",
        "\n",
        "A: 128\n",
        "\n",
        "\n",
        "Q: HOW?\n",
        "\n",
        "A: Batch Normalization layer introduces 4 trainable parameters per channel. Number of channels in the preceeding layers is 32. Hence, the total number of parameters is equal to 32 * 4 = 128."
      ],
      "metadata": {
        "id": "cuFhr6H8xa47"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model4.summary()"
      ],
      "metadata": {
        "id": "MtKeg1J9tVk9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4ed7ed62-6ba4-42ac-c814-2c7b8279576c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_3\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_2 (InputLayer)        [(None, 32, 32, 3)]       0         \n",
            "                                                                 \n",
            " conv2d_4 (Conv2D)           (None, 30, 30, 32)        864       \n",
            "                                                                 \n",
            " batch_normalization (Batch  (None, 30, 30, 32)        128       \n",
            " Normalization)                                                  \n",
            "                                                                 \n",
            " conv2d_5 (Conv2D)           (None, 30, 30, 64)        18496     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 19488 (76.12 KB)\n",
            "Trainable params: 19424 (75.88 KB)\n",
            "Non-trainable params: 64 (256.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Intuitions:**\n",
        "*  Batch Normalization is also a (weak) regularization method.\n",
        "    - increases no. of params\n",
        "    - but also adds noise ~ data augmentation ~ dropout"
      ],
      "metadata": {
        "id": "4YLehpMpxJMN"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Depthwise separable convolutions**\n",
        "\n",
        "* This layer performs a **spatial convolution [Depthwise Conv.]** on each channel of\n",
        "its input, independently, before mixing output channels via a **pointwise convolution**.\n",
        "\n",
        "* **Depthwise separable convolution = Depthwise Conv. +  Pointwise Conv.**\n",
        "\n",
        "* This makes your model smaller and  acts as a strong prior. We impose a strong prior by assuming that spatial patterns and cross-channel patterns can be modeled separately.This is equivalent to separating the learning of spatial features and the learning of channel-wise features.\n",
        "* Depthwise separable convolution relies on the assumption that spatial locations in intermediate activations\n",
        "are highly correlated, but different channels are highly independent. So we never use depthwise separable convolution after the input layer. Because RGB channels are **highly correlated**."
      ],
      "metadata": {
        "id": "4AqwncgPvQju"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "<figure>\n",
        "<center>\n",
        "<img src=\"https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/RGB_channels_separation.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045128Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=8099705ef3284a3c743b137d3b2a6269e85c9b876f9d92c84622b0922b0d965a\" width=\"400\"/><img src=\"https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/DepthWiseConvolution.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045229Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=de1e1f82b7a2215648dba333338b2baa0be6b83fc86685e6f0369e04982be431\" width=\"500\"/> <figcaption>## RGB channels are highly correlated. ##________________## Depthwise separable convolutions ##</figcaption>\n",
        "\n",
        "</center>\n",
        "</figure>\n"
      ],
      "metadata": {
        "id": "ezUdJ7lJ4f-p"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's quickly look at the code first"
      ],
      "metadata": {
        "id": "9__AMGKUS6bb"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Building a model with Separable Conv layer\n",
        "inputs = keras.Input(shape=(32, 32, 3))\n",
        "x = layers.Conv2D(32, 3, activation=\"relu\")(inputs)\n",
        "x = layers.SeparableConv2D(64, 3, activation=\"relu\", padding=\"same\")(x)\n",
        "sep_model = keras.Model(inputs=inputs, outputs=x)\n",
        "plot_model(sep_model, show_shapes=True)\n",
        "sep_model.summary()\n",
        "# Q: Verify the no. of params in the separable_conv2D layer A: (32*(3*3)) + (32*1*1*64) + 64 = 2400\n",
        "# Not in keras 'SeparableConv2D', the bias at  Depthwise Conv. is not considered.\n",
        "# Considering all bias : (32*(3*3) + 32) + (32*1*1*64) + 64 = 2432"
      ],
      "metadata": {
        "id": "vwVyRlQ8uOLb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f6b3e98c-cce4-407c-c3f6-e716a125cedb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_4\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_3 (InputLayer)        [(None, 32, 32, 3)]       0         \n",
            "                                                                 \n",
            " conv2d_6 (Conv2D)           (None, 30, 30, 32)        896       \n",
            "                                                                 \n",
            " separable_conv2d (Separabl  (None, 30, 30, 64)        2400      \n",
            " eConv2D)                                                        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 3296 (12.88 KB)\n",
            "Trainable params: 3296 (12.88 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's compare the above with a model where we replace the SeparableConv2D with a Conv2D layer."
      ],
      "metadata": {
        "id": "XCsZV3LCxYJd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "inputs = keras.Input(shape=(32, 32, 3))\n",
        "x = layers.Conv2D(32, 3, activation=\"relu\")(inputs)\n",
        "x = layers.Conv2D(64, 3, activation=\"relu\", padding=\"same\")(x)\n",
        "model = keras.Model(inputs=inputs, outputs=x)\n",
        "plot_model(model, show_shapes=True)\n",
        "model.summary()\n",
        "# Q: Why does sep_model have much less params? A: Depthwise and pointwise convs are done independently"
      ],
      "metadata": {
        "id": "0bS1qIoywrAB",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8b2142ef-3269-499d-bcc8-680d0f8225e8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_5\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_4 (InputLayer)        [(None, 32, 32, 3)]       0         \n",
            "                                                                 \n",
            " conv2d_7 (Conv2D)           (None, 30, 30, 32)        896       \n",
            "                                                                 \n",
            " conv2d_8 (Conv2D)           (None, 30, 30, 64)        18496     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 19392 (75.75 KB)\n",
            "Trainable params: 19392 (75.75 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **A mini Xception-like model**"
      ],
      "metadata": {
        "id": "rfYMjxYWyOL5"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "We'll build a model like the Xception model, but a smaller version.\n",
        "\n",
        "But first let's see what the actual Xception model looks like."
      ],
      "metadata": {
        "id": "tQkHVp72IRkX"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/Entry_Middle_Exit_Flow.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045310Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=0b4a3e52453bd81dcd7f7e8aa58854113091bde2428090e3840db40a9d67fd0b)"
      ],
      "metadata": {
        "id": "OEYS-LaWIDql"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Q: In middle- flow blocks, what arguments do you give to the SepConv layer ?\n",
        "\n",
        "A: HW question\n"
      ],
      "metadata": {
        "id": "yXmbHNJAWhMd"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's use the cats-vs-dogs data and create datasets."
      ],
      "metadata": {
        "id": "FQYxgWbwIvLN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "# Load the cats_vs_dogs dataset\n",
        "dataset, info = tfds.load('cats_vs_dogs', split='train', with_info=True)\n",
        "\n",
        "# Define the percentage of the dataset to be used\n",
        "percentage_to_use = 0.4  # Adjust as needed\n",
        "\n",
        "# Calculate the size of the subset\n",
        "subset_size = int(percentage_to_use * info.splits['train'].num_examples)\n",
        "\n",
        "# Take a subset of the dataset\n",
        "subset_dataset = dataset.take(subset_size)\n",
        "\n",
        "# Function to filter dataset by class label\n",
        "def filter_by_label(example, label):\n",
        "    return example['label'] == label\n",
        "\n",
        "# Stratified sampling for cats (label 0)\n",
        "cats_subset = subset_dataset.filter(lambda example: filter_by_label(example, label=0))\n",
        "\n",
        "# Stratified sampling for dogs (label 1)\n",
        "dogs_subset = subset_dataset.filter(lambda example: filter_by_label(example, label=1))\n",
        "\n",
        "# Combine the stratified subsets\n",
        "stratified_subset = tf.data.Dataset.concatenate(cats_subset, dogs_subset)\n",
        "\n"
      ],
      "metadata": {
        "id": "TUfhxMyOkrA2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 133,
          "referenced_widgets": [
            "5839992e885a4c479e5cc2daacd37ff5",
            "aebc6d898897497d9a770fc85b28a3f5",
            "4138a2eb2bc948e2ada9128fcb585273",
            "ed3de5b1efa8416dbcc38106054be627",
            "52ae0b3b49364d798dacf31275b50ed8",
            "585f7420a99b4024a33c5bd1b8c18dc8",
            "05f7029da78e4d5593336f8d6a86ef8c",
            "23e9e9437dc040a5be551c8336cd2ca2",
            "3b02999701274357be9f621bdd101ebd",
            "a4ccab6028e64a928be5c039d008500f",
            "837c54300daa441596e87974d800c38f",
            "f6f32c41c64b41dc95c6e415e458c116",
            "9810ab2b2195476fa90d096cd9c6e5e1",
            "7d3546e2261b4ac2af92ea5234fddc29",
            "cac5448e55b94e5b96585c3509a7a5fb",
            "1cd1d0fa70a14aed8d149b4cea53342b",
            "2ac5b24d64c74b2a8a9416bba1fa441a",
            "3c8d8f6f247f476fa64ee44c5565b220",
            "861816482bda4cb3a0127058b13d685f",
            "56b9dde8640c4e08a96ad53952929527",
            "630c6f79d7c14ca19dd51bb049f50ab3",
            "fb849e5baa9a45f7975779b0cdc0af20",
            "f3b83543335d4525918140f9fa670781",
            "64ac586a9ff14536a1d39e0eed670842",
            "442f413984e54320a73789a139c46c02",
            "792a4571060848c485ffaba1777c1163",
            "18df3dd37bc946d9b5001f23fb6e45b6",
            "ec4314ef910c473e931ef0bcdd592e9c",
            "b318329efb214015a925964af9d570b0",
            "679bd91a9a2b449381e8bddb790afef0",
            "f0123f7bba0d4d92a877c2f89255a1da",
            "250d77d4571540288c3638a1dbfceb93",
            "f227d27a3ad74949a96be5c0b0a55ad9",
            "07d185b4bfcc44f3ad67ff32f6bfed10",
            "ea855e833ee74b5ea76631bd2cd9bfff",
            "d18355e2c05d4785bece2b6c65e0252e",
            "a75931b7752c46bbb4e6f93338effac7",
            "65d9209db45d48efaaeba2d2b95d314a",
            "768d6df02fb745b6af7e4b1340146f9d",
            "6d8da11ee3434f22af8fd205da2313b5",
            "359e3cca25794f0da63b0ced5f324267",
            "96cbc2951f674b4f9b32a6202a18f24c",
            "70efc159082e433499b6c9cb53425e0c",
            "d7de298de6ab41fc8aefeefc4ecc6622",
            "8f74bdcdc1954040a525e86b1e43fcc7",
            "9caa32ebf38d4793b6ea61d1beb71d45",
            "2e8449772ec34ab3b761ccba95918c7b",
            "33a952d9d784496a9d42c9ecdc122537",
            "93a6d40a5dfe497ba1ac847f02564059",
            "618b462829044b3e81c4b9bdfb4b0977",
            "381d46508b3a46f8859a21491cc78e58",
            "481edd60ca6d41bea5a1c3093e6561fc",
            "42016a20de984e78b6a808b212a475d7",
            "843585c448b640279f1ca954025b5777",
            "17efa89b1455463fae3bda2fe2812ef9"
          ]
        },
        "outputId": "87bfd7de-a0ed-483d-87d6-bc220f9b3abf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading and preparing dataset 786.68 MiB (download: 786.68 MiB, generated: Unknown size, total: 786.68 MiB) to /root/tensorflow_datasets/cats_vs_dogs/4.0.0...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Dl Completed...: 0 url [00:00, ? url/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "5839992e885a4c479e5cc2daacd37ff5"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Dl Size...: 0 MiB [00:00, ? MiB/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "f6f32c41c64b41dc95c6e415e458c116"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating splits...:   0%|          | 0/1 [00:00<?, ? splits/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "f3b83543335d4525918140f9fa670781"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating train examples...:   0%|          | 0/23262 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "07d185b4bfcc44f3ad67ff32f6bfed10"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:1738 images were corrupted and were skipped\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Shuffling /root/tensorflow_datasets/cats_vs_dogs/4.0.0.incompleteAV7M9C/cats_vs_dogs-train.tfrecord*...:   0%|…"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "8f74bdcdc1954040a525e86b1e43fcc7"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dataset cats_vs_dogs downloaded and prepared to /root/tensorflow_datasets/cats_vs_dogs/4.0.0. Subsequent calls will reuse this data.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Function to resize an image\n",
        "def resize_image(sample):\n",
        "    image = sample['image']\n",
        "    resized_image = tf.image.resize(image, (180, 180))  # Adjust the target size as needed\n",
        "    sample['image'] = resized_image\n",
        "    return sample\n",
        "\n",
        "# Apply image resizing to the entire dataset\n",
        "stratified_subset = stratified_subset.map(resize_image)\n",
        "\n",
        "\n",
        "print(f\"stratified_subset = {stratified_subset}\")"
      ],
      "metadata": {
        "id": "CVOWgUhECt5d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f67bd0fe-71bd-4e0f-ea11-4b6c4218aa06"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "stratified_subset = <_MapDataset element_spec={'image': TensorSpec(shape=(180, 180, 3), dtype=tf.float32, name=None), 'image/filename': TensorSpec(shape=(), dtype=tf.string, name=None), 'label': TensorSpec(shape=(), dtype=tf.int64, name=None)}>\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Define the sizes for training, validation, and testing sets\n",
        "train_size = int(0.8 * subset_size)  # 80% for training\n",
        "validation_size = int(0.1 * subset_size)  # 10% for validation\n",
        "test_size = subset_size - train_size - validation_size  # Remaining for testing\n",
        "\n",
        "batch_size = 32\n",
        "\n",
        "# Create a new dataset with a simpler structure using map for training\n",
        "train_dataset = (\n",
        "    stratified_subset\n",
        "    .take(train_size)\n",
        "    .map(lambda x: (x['image'], x['label']))\n",
        "    .batch(batch_size)\n",
        "    .prefetch(buffer_size=tf.data.AUTOTUNE)\n",
        ")\n",
        "\n",
        "# Create a new dataset with a simpler structure using map for validation\n",
        "validation_dataset = (\n",
        "    stratified_subset\n",
        "    .skip(train_size)  # Skip the training set\n",
        "    .take(validation_size)\n",
        "    .map(lambda x: (x['image'], x['label']))\n",
        "    .batch(batch_size)\n",
        "    .prefetch(buffer_size=tf.data.AUTOTUNE)\n",
        ")\n",
        "\n",
        "# Create a new dataset with a simpler structure using map for testing\n",
        "test_dataset = (\n",
        "    stratified_subset\n",
        "    .skip(train_size + validation_size)  # Skip the training and validation sets\n",
        "    .map(lambda x: (x['image'], x['label']))\n",
        "    .batch(batch_size)\n",
        "    .prefetch(buffer_size=tf.data.AUTOTUNE)\n",
        ")\n",
        "\n",
        "# Access the element_spec attribute to get the shape of a batch for training\n",
        "print(\"train_dataset =\", train_dataset, \"\\n\")\n",
        "\n",
        "# Access the element_spec attribute to get the shape of a batch for validation\n",
        "print(\"validation_dataset =\", validation_dataset, \"\\n\")\n",
        "\n",
        "# Access the element_spec attribute to get the shape of a batch for testing\n",
        "print(\"test_dataset =\", test_dataset, \"\\n\")\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3bfTVL4btj_n",
        "outputId": "06f066c4-40bb-44e4-beda-4975e9da56c0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "train_dataset = <_PrefetchDataset element_spec=(TensorSpec(shape=(None, 180, 180, 3), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.int64, name=None))> \n",
            "\n",
            "validation_dataset = <_PrefetchDataset element_spec=(TensorSpec(shape=(None, 180, 180, 3), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.int64, name=None))> \n",
            "\n",
            "test_dataset = <_PrefetchDataset element_spec=(TensorSpec(shape=(None, 180, 180, 3), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.int64, name=None))> \n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import keras\n",
        "from keras import layers\n",
        "inputs = keras.Input(shape=(180, 180, 3))\n",
        "# x = data_augmentation(inputs)\n",
        "\n",
        "x = layers.Rescaling(1./255)(inputs)\n",
        "x = layers.Conv2D(filters=32, kernel_size=5, use_bias=False)(x) # Q: why not use depth-wise sep conv here? A: RGB channels in input image are highly corelated\n",
        "\n",
        "for size in [32, 64, 128, 256, 512]:                # Repeated block. Very common practice\n",
        "    residual = x\n",
        "\n",
        "    x = layers.BatchNormalization()(x)              # We can also apply BN just before the activation\n",
        "    x = layers.Activation(\"relu\")(x)\n",
        "    x = layers.SeparableConv2D(size, 3, padding=\"same\", use_bias=False)(x)\n",
        "\n",
        "    x = layers.BatchNormalization()(x)\n",
        "    x = layers.Activation(\"relu\")(x)\n",
        "    x = layers.SeparableConv2D(size, 3, padding=\"same\", use_bias=False)(x)\n",
        "\n",
        "    x = layers.MaxPooling2D(3, strides=2, padding=\"same\")(x)\n",
        "\n",
        "    residual = layers.Conv2D(\n",
        "        size, 1, strides=2, padding=\"same\", use_bias=False)(residual)\n",
        "    x = layers.add([x, residual])\n",
        "\n",
        "x = layers.GlobalAveragePooling2D()(x)\n",
        "x = layers.Dropout(0.5)(x)\n",
        "outputs = layers.Dense(1, activation=\"sigmoid\")(x)\n",
        "model = keras.Model(inputs=inputs, outputs=outputs)"
      ],
      "metadata": {
        "id": "UnZGLjDhycgf"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_model(model)"
      ],
      "metadata": {
        "id": "BIvwflhlc9Q-",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "a5f3f936-5a40-4f34-d115-dae308b0aa0d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "xDIn4pcAyjcC",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "38d538df-d6c1-4935-a074-c9cb93c1d311"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "233/233 [==============================] - 54s 207ms/step - loss: 0.1018 - accuracy: 0.9915 - val_loss: 5.8770e-05 - val_accuracy: 1.0000\n"
          ]
        }
      ],
      "source": [
        "model.compile(loss=\"binary_crossentropy\",\n",
        "              optimizer=\"rmsprop\",\n",
        "              metrics=[\"accuracy\"])\n",
        "history = model.fit( train_dataset,\n",
        "                      epochs=1,\n",
        "                      validation_data=validation_dataset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ZIhBNi09EvQ0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 881
        },
        "outputId": "0b3890b1-05c0-4f5e-8654-cda07de36a93"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "data = pd.DataFrame(history.history)\n",
        "plt.plot(range(1,len(data)+1),data['accuracy'],'bo',label=\"Training accuracy\")\n",
        "plt.plot(range(1,len(data)+1),data['val_accuracy'],'b',label=\"Validation accuracy\")\n",
        "plt.legend()\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.show()\n",
        "plt.figure()\n",
        "plt.plot(range(1,len(data)+1),data['loss'],'bo',label=\"Training loss\")\n",
        "plt.plot(range(1,len(data)+1),data['val_loss'],'b',label=\"Validation loss\")\n",
        "plt.legend()\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 2. Interpreting what ConvNets learn\n",
        " * Visualizing intermediate ConvNetsoutputs (intermediate activations)\n",
        " * Visualizing ConvNets filters\n",
        " * Visualizing heatmaps of class activation in an image\n",
        "\n",
        "NOTE:\n",
        "* We will focus mostly on the concepts and key ideas\n",
        "* A lot of the code is pre-processing and post-processing. We will not spend time on these parts.\n",
        "* We will see the part of the code that implements the key ideas.\n",
        "\n",
        "Pro:\n",
        "*   Developing ideas\n",
        "*   Developing thought process"
      ],
      "metadata": {
        "id": "UHFkjDr4eAZV"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Visualizing intermediate activations**\n",
        "\n",
        "The output of a layer is called its 'activation'(It's the output of the activation function).\n",
        "\n",
        "These activations can be visualized by plotting the feature maps.\n",
        "\n",
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/Visualize_Intermediate_Activations.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045418Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=89f7d29949ce3b3939d75f1c456b575618b25b273246853b85d95bc6b0b0aceb)\n",
        "\n",
        "We will plot each feature map independently as a 2D image, since they encode relatively indepent features."
      ],
      "metadata": {
        "id": "TtyRbwd7eKAE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.utils import plot_model\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from tensorflow.keras import layers  # <----- Note this"
      ],
      "metadata": {
        "id": "W6EpNnTPtWcL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Model to be used for visualization :** We are going to use a trained model from the previous assignment -  M2_AST_01_Convolutional _Neural_Networks. In that assignment, we created one model with augmentation and saved it through a callback function with the name \"convnet_from_scratch_with_augmentation_keras\". You can download that model from there and use it by providing the proper path after loading it.\n",
        "\n",
        "For the sake of simplicity, we have already provided that model and has been downloaded along with the dataset."
      ],
      "metadata": {
        "id": "mrmyY_4uF8rE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = keras.models.load_model('/content/convnet_from_scratch_with_augmentation_keras')\n",
        "plot_model(model)"
      ],
      "metadata": {
        "id": "zoNajzD7pntO",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 338
        },
        "outputId": "9a969452-4e33-4fe1-a68a-81437a28d632"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "OSError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mOSError\u001b[0m                                   Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-25-5f73266151ef>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mkeras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodels\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'/content/convnet_from_scratch_with_augmentation_keras'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mplot_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/keras/src/saving/saving_api.py\u001b[0m in \u001b[0;36mload_model\u001b[0;34m(filepath, custom_objects, compile, safe_mode, **kwargs)\u001b[0m\n\u001b[1;32m    260\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    261\u001b[0m     \u001b[0;31m# Legacy case.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 262\u001b[0;31m     return legacy_sm_saving_lib.load_model(\n\u001b[0m\u001b[1;32m    263\u001b[0m         \u001b[0mfilepath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcustom_objects\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcustom_objects\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcompile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcompile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    264\u001b[0m     )\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/keras/src/utils/traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     68\u001b[0m             \u001b[0;31m# To get the full stack trace, call:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     69\u001b[0m             \u001b[0;31m# `tf.debugging.disable_traceback_filtering()`\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 70\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwith_traceback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfiltered_tb\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     71\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     72\u001b[0m             \u001b[0;32mdel\u001b[0m \u001b[0mfiltered_tb\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/keras/src/saving/legacy/save.py\u001b[0m in \u001b[0;36mload_model\u001b[0;34m(filepath, custom_objects, compile, options)\u001b[0m\n\u001b[1;32m    232\u001b[0m                     \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath_str\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    233\u001b[0m                         \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mio\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgfile\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexists\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath_str\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 234\u001b[0;31m                             raise IOError(\n\u001b[0m\u001b[1;32m    235\u001b[0m                                 \u001b[0;34mf\"No file or directory found at {filepath_str}\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    236\u001b[0m                             )\n",
            "\u001b[0;31mOSError\u001b[0m: No file or directory found at /content/convnet_from_scratch_with_augmentation_keras"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_loss, test_acc = model.evaluate(test_dataset) #Q: Why don't we have 2 arguments\n",
        "print(f\"Test accuracy is:{test_acc:.3f}\")"
      ],
      "metadata": {
        "id": "2dky0dGDqoew"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Getting the image &  Preprocessing** that will be passsed inside the model for visualization ."
      ],
      "metadata": {
        "id": "LSL_9iBEGlZA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "img_path = keras.utils.get_file(fname=\"cat.jpg\",origin=\"https://img-datasets.s3.amazonaws.com/cat.jpg\")\n",
        "# Preprocessing a single image\n",
        "def get_img_array(img_path, target_size):\n",
        "    img = keras.utils.load_img(img_path, target_size=target_size)\n",
        "    array = keras.utils.img_to_array(img) # converts image to np array\n",
        "    # Add a dimension to transform the array intoa “batch” of a single sample.\n",
        "    array = np.expand_dims(array, axis=0)     #Its shape is now (1, 180, 180, 3)\n",
        "    return array\n",
        "\n",
        "img_tensor = get_img_array(img_path, target_size=(180, 180)) #resize image\n",
        "\n",
        "plt.axis(\"off\")\n",
        "plt.imshow(img_tensor[0].astype(\"uint8\"))\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "vvkpXx6KhZng"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Focus here\n",
        "# Instantiating a model that returns \"layer activations\"\n",
        "layer_outputs = []\n",
        "layer_names = []\n",
        "for layer in model.layers:                                      #Q: what does model.layers return ? A: list of layers of the model\n",
        "    if isinstance(layer, (layers.Conv2D, layers.MaxPooling2D)): #Q: difference between layer and layers? A: layer is a layer object in the model. layers is the module we have imported\n",
        "        layer_outputs.append(layer.output)\n",
        "        # layer_outputs.append(layers.Dense(1)(layer.output))  # To visualise the output in plot_model()\n",
        "        layer_names.append(layer.name)\n",
        "activation_model = keras.Model(inputs=model.input, outputs=layer_outputs) #Q: what does layer_outputs contain? A: output of conv2d and maxpool layers\n",
        "plot_model(activation_model)\n"
      ],
      "metadata": {
        "id": "A18axaoihdg8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "activation_model.summary()"
      ],
      "metadata": {
        "id": "wtSTumbvHGfM"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Compute layer activations\n",
        "activations = activation_model.predict(img_tensor)\n",
        "print(f\"No. of outputs= {len(activations)}\")\n",
        "\n",
        "first_layer_feature_maps = activations[0]    # Q: why is this the first layer's activation? A: index 0\n",
        "print(f\"first_layer_activation.shape= {first_layer_feature_maps.shape}\")"
      ],
      "metadata": {
        "id": "7z909nZiiTxc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Visualise activation\n",
        "import matplotlib.pyplot as plt\n",
        "plt.matshow(first_layer_feature_maps[0, :, :, 0], cmap=\"viridis\")  # Q: which (1st/2nd/..)feature map are we visualing? A: 1st (0 in the last index)"
      ],
      "metadata": {
        "id": "lML8McG4k0L9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "It seems that the filter has detected ______.\n",
        "\n",
        "Let's look at a feature map after each layer."
      ],
      "metadata": {
        "id": "zqjzaGrX0vZg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "for i in range(9):\n",
        "  plt.matshow(activations[i][0, :, :, 2], cmap=\"viridis\")\n"
      ],
      "metadata": {
        "id": "iHK7sEtx2IUo"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Note the dimensions on the above images. Successive feature maps are actually of smaller dimensions but scaled to be the same size during visualization."
      ],
      "metadata": {
        "id": "wXJOiFWx1Bg0"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now let's visualise all the feature maps of all the layers."
      ],
      "metadata": {
        "id": "5SC0nlq_3Lb0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Post-processing code - only visualizaton\n",
        "# Visualizing every channel in every intermediate activation\n",
        "images_per_row = 16\n",
        "for layer_name, layer_activation in zip(layer_names, activations):\n",
        "    n_features = layer_activation.shape[-1]\n",
        "    size = layer_activation.shape[1]\n",
        "    n_cols = n_features // images_per_row\n",
        "    display_grid = np.zeros(((size + 1) * n_cols - 1,\n",
        "                             images_per_row * (size + 1) - 1))\n",
        "    for col in range(n_cols):\n",
        "        for row in range(images_per_row):\n",
        "            channel_index = col * images_per_row + row\n",
        "            channel_image = layer_activation[0, :, :, channel_index].copy()\n",
        "            if channel_image.sum() != 0:\n",
        "                channel_image -= channel_image.mean()\n",
        "                channel_image /= channel_image.std()\n",
        "                channel_image *= 64\n",
        "                channel_image += 128\n",
        "            channel_image = np.clip(channel_image, 0, 255).astype(\"uint8\")\n",
        "            display_grid[\n",
        "                col * (size + 1): (col + 1) * size + col,\n",
        "                row * (size + 1) : (row + 1) * size + row] = channel_image\n",
        "    scale = 1. / size\n",
        "    plt.figure(figsize=(scale * display_grid.shape[1],\n",
        "                        scale * display_grid.shape[0]))\n",
        "    plt.title(layer_name)\n",
        "    plt.grid(False)\n",
        "    plt.axis(\"off\")\n",
        "    plt.imshow(display_grid, aspect=\"auto\", cmap=\"viridis\")"
      ],
      "metadata": {
        "id": "sov3-WP2k3aT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "* The first layer acts as a collection of various edge detectors.\n",
        "\n",
        "* As you go deeper, the activations become increasingly abstract and less visually interpretable. They begin to encode higher-level concepts such as “cat ear” and “cat eye.”\n",
        "\n",
        "* The sparsity of the activations increases with the depth of the layer: in the first layer, almost all filters are activated by the input image, but in the following layers, more and more filters are blank. This means the pattern encoded by the filter isn’t found in the input image"
      ],
      "metadata": {
        "id": "vDL2xEBypLwG"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Visualising ConvNet filters**\n",
        "*   Pick a filter\n",
        "*   Ask the question: What kind of an input image will excite the filter?\n",
        "*   What should the input image be so that you see a (yellow) feature map?\n",
        "*   In other words, we want to visualize those patterns in the input image that the filter picks up and results in high (yellow) values in the feature map.\n"
      ],
      "metadata": {
        "id": "EVsagdgdquAw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Instantiating the Xception convolutional base\n",
        "\n",
        "model = keras.applications.xception.Xception( weights=\"imagenet\", include_top=False)"
      ],
      "metadata": {
        "id": "VA0-i2t6k69H"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Q: Printing the names of  conv and sepConv layers in Xception\n",
        "\n",
        "for layer in model.layers:\n",
        "    if isinstance(layer, (keras.layers.Conv2D, keras.layers.SeparableConv2D)):\n",
        "        print(layer.name)"
      ],
      "metadata": {
        "id": "1fkAZk_fyIaq"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating a feature extractor model\n",
        "\n",
        "layer_name = \"block3_sepconv1\"\n",
        "layer = model.get_layer(name=layer_name)\n",
        "feature_extractor = keras.Model(inputs=model.input, outputs=layer.output)\n",
        "feature_extractor.summary()"
      ],
      "metadata": {
        "id": "rb_zmZoNyIc-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Q: What is the last layer ?\n",
        "\n",
        "Q: How many filters does block3_sepconv1 have?\n",
        "\n",
        "Q: Why are there so many Nones in the shapes?"
      ],
      "metadata": {
        "id": "q33H-EUi7aeB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Using the feature extractor\n",
        "\n",
        "activation = feature_extractor(keras.applications.xception.preprocess_input(img_tensor))\n",
        "# Image is preprocessed specific to Inception model before passing inside the feature_extractor"
      ],
      "metadata": {
        "id": "lB30XfKvyIff"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Here comes the key idea:\n",
        "*   Define an objective function: mean pixel value of feature map\n",
        "*   Use gradient \"Ascent\" on the \"input image space\" to maximize this objective\n",
        "Here's an analogy: (drawing)"
      ],
      "metadata": {
        "id": "aTnndxqY6sTc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "\n",
        "def compute_loss(image, filter_index):      #Q: How many indices do we have? A: 256\n",
        "    activation = feature_extractor(image)\n",
        "    filter_activation = activation[:, 2:-2, 2:-2, filter_index] # leaving out the boundaries\n",
        "    return tf.reduce_mean(filter_activation)"
      ],
      "metadata": {
        "id": "fLxwawZfyIh0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Loss maximization via stochastic gradient ascent\n",
        "# Revise tutorial 1\n",
        "@tf.function\n",
        "def gradient_ascent_step(image, filter_index, learning_rate):\n",
        "    with tf.GradientTape() as tape:\n",
        "        tape.watch(image)\n",
        "        loss = compute_loss(image, filter_index)\n",
        "    grads = tape.gradient(loss, image)    # Q: Is the gradient a vector or scalar? A: vector\n",
        "    grads = tf.math.l2_normalize(grads)    # Normalize the gradient\n",
        "    image += learning_rate * grads          # Q: What makes this gradient \"ascent\"? A: plus sign\n",
        "    return image\n"
      ],
      "metadata": {
        "id": "-_amLc3KyIj8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Function to generate filter visualizations\n",
        "\n",
        "img_width = 200\n",
        "img_height = 200\n",
        "\n",
        "def generate_filter_pattern(filter_index):\n",
        "    iterations = 30\n",
        "    learning_rate = 10.\n",
        "    image = tf.random.uniform(\n",
        "        minval=0.4,\n",
        "        maxval=0.6,\n",
        "        shape=(1, img_width, img_height, 3))\n",
        "    for i in range(iterations):\n",
        "        image = gradient_ascent_step(image, filter_index, learning_rate)\n",
        "    return image[0].numpy()"
      ],
      "metadata": {
        "id": "nHshMEN4yIml"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Utility function to convert a tensor into a valid image\n",
        "\n",
        "def deprocess_image(image):\n",
        "    image -= image.mean()\n",
        "    image /= image.std()\n",
        "    image *= 64\n",
        "    image += 128\n",
        "    image = np.clip(image, 0, 255).astype(\"uint8\")\n",
        "    image = image[25:-25, 25:-25, :]\n",
        "    return image\n",
        "\n",
        "plt.axis(\"off\")\n",
        "plt.imshow(deprocess_image(generate_filter_pattern(filter_index=2)))"
      ],
      "metadata": {
        "id": "oSnWkDvN0EZ1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Post processing- Just visualization\n",
        "# Generating a grid of all filter response patterns in a layer\n",
        "\n",
        "all_images = []\n",
        "for filter_index in range(64):\n",
        "    print(f\"Processing filter {filter_index}\")\n",
        "    image = deprocess_image(\n",
        "        generate_filter_pattern(filter_index)\n",
        "    )\n",
        "    all_images.append(image)\n",
        "\n",
        "margin = 5\n",
        "n = 8\n",
        "cropped_width = img_width - 25 * 2\n",
        "cropped_height = img_height - 25 * 2\n",
        "width = n * cropped_width + (n - 1) * margin\n",
        "height = n * cropped_height + (n - 1) * margin\n",
        "stitched_filters = np.zeros((width, height, 3))\n",
        "\n",
        "for i in range(n):\n",
        "    for j in range(n):\n",
        "        image = all_images[i * n + j]\n",
        "        stitched_filters[\n",
        "            (cropped_width + margin) * i : (cropped_width + margin) * i + cropped_width,\n",
        "            (cropped_height + margin) * j : (cropped_height + margin) * j\n",
        "            + cropped_height,\n",
        "            :,\n",
        "        ] = image\n",
        "\n",
        "keras.utils.save_img(\n",
        "    f\"filters_for_layer_{layer_name}.png\", stitched_filters)"
      ],
      "metadata": {
        "id": "rASYbJnX0HtR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# plt.figure(figsize=(40,40))\n",
        "# plt.matshow(stitched_filters)\n",
        "\n",
        "for i in [0,8,16,32]:\n",
        "  plt.figure()\n",
        "  plt.imshow((all_images[i]))"
      ],
      "metadata": {
        "id": "mqQbBmvZ0Hwi"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/block2.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045443Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=500abe9012fc3dc49107e9dae6501c7d0179863707372fa36bdc8ccd38cdf0f9)"
      ],
      "metadata": {
        "id": "rn1Ra77e4VIK"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/block4.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045521Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=7aef767ada529726781883741aca52c2ad73323305dcca79a5c2726678a9adf2)"
      ],
      "metadata": {
        "id": "EREwT4VT8vRN"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/block8.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045547Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=d2bbe81cc84d4c4e4c3c464a29bdbdca82ea4b4734fab9ba2703358e15821b17)\n",
        "\n"
      ],
      "metadata": {
        "id": "OQLo9U5-8wBs"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "###  **Visualizing heatmaps of class activation**\n",
        "\n",
        "* Visualise which parts of a given image led a ConvNet to its final classification decision\n",
        "* Such techniques are called **class activation map** (CAM) visualisation\n",
        "* Produce heatmaps of class activation over input images."
      ],
      "metadata": {
        "id": "cA6vN4Yb4ijq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Example:\n",
        "\n",
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/Grad_CAM1.jpeg?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045725Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=27bccd47f379728b643a477a6ee29e1b5c6cc70f9866215516ca77e95611901b)"
      ],
      "metadata": {
        "id": "edfJ7xmG-p-v"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "First let's understand the idea behind CAM:\n",
        "\n",
        "(Remember: The CNN is already trained. Now we are just visualising aspects of the trained CNN)\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/Idea_Behind_CAM.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045829Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=96b10ba1a93c2ccac23ba98daa1dd471a02b6358ab5dd97bf03e10ca5e3b9018)"
      ],
      "metadata": {
        "id": "lBl--ze7_go6"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "Note:\n",
        "*   One set of optimal weights for one class\n",
        "\n"
      ],
      "metadata": {
        "id": "VU9idU_-HBwr"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now let's look at an improved version of CAM:\n",
        "\n",
        "\n",
        "*   Grad-CAM\n",
        "\n"
      ],
      "metadata": {
        "id": "8uRA385UHK82"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/Grad_CAM.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045908Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=85bb3d48e78de63b090f1daa9ccc216b1362d46b3d74e2c777c5f65d573acb0d)"
      ],
      "metadata": {
        "id": "BlKVYyd-__qi"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Key Take-away:\n",
        "\n",
        "\n",
        "*   Need the weights. Q: What are these weights?\n",
        "    - make up a weighted sum of featurmaps to get a heat map\n",
        "*   Can learn them through a new sub-problem- CAM\n",
        "*   Can compute them directly through gradients- grad-CAM\n",
        "\n"
      ],
      "metadata": {
        "id": "6wpNI2J4I9jw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Loading the Xception network with pretrained weights\n",
        "\n",
        "model = keras.applications.xception.Xception(weights=\"imagenet\")\n"
      ],
      "metadata": {
        "id": "HOq9rVmh4rLK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# model.summary()"
      ],
      "metadata": {
        "id": "CinrH57K5tdr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Preprocessing an input image for Xception\n",
        "\n",
        "img_path = keras.utils.get_file(fname=\"cat.jpg\",\n",
        "                                origin=\"https://img-datasets.s3.amazonaws.com/cat.jpg\")\n",
        "\n",
        "def get_img_array(img_path, target_size):\n",
        "    img = keras.utils.load_img(img_path, target_size=target_size)\n",
        "    array = keras.utils.img_to_array(img)\n",
        "    array = np.expand_dims(array, axis=0)\n",
        "    array = keras.applications.xception.preprocess_input(array)\n",
        "    return array\n",
        "\n",
        "img_array = get_img_array(img_path, target_size=(180, 180))"
      ],
      "metadata": {
        "id": "QlFBo2dE4tQh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Predicting the top three labels\n",
        "preds = model.predict(img_array)\n",
        "print(keras.applications.xception.decode_predictions(preds, top=3)[0])"
      ],
      "metadata": {
        "id": "wVUyOcLu4vT8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 373
        },
        "outputId": "62de4b7a-88e9-4b4f-f325-67b1b2716df1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 1s 568ms/step\n"
          ]
        },
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-39-fcb7ad605d42>\u001b[0m in \u001b[0;36m<cell line: 3>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# Predicting the top three labels\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mpreds\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimg_array\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkeras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapplications\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxception\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode_predictions\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpreds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtop\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/keras/src/applications/xception.py\u001b[0m in \u001b[0;36mdecode_predictions\u001b[0;34m(preds, top)\u001b[0m\n\u001b[1;32m    370\u001b[0m \u001b[0;34m@\u001b[0m\u001b[0mkeras_export\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"keras.applications.xception.decode_predictions\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    371\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mdecode_predictions\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpreds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtop\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 372\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mimagenet_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode_predictions\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpreds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtop\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    373\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    374\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/keras/src/applications/imagenet_utils.py\u001b[0m in \u001b[0;36mdecode_predictions\u001b[0;34m(preds, top)\u001b[0m\n\u001b[1;32m    152\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    153\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpreds\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0;36m2\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mpreds\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0;36m1000\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 154\u001b[0;31m         raise ValueError(\n\u001b[0m\u001b[1;32m    155\u001b[0m             \u001b[0;34m\"`decode_predictions` expects \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    156\u001b[0m             \u001b[0;34m\"a batch of predictions \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: `decode_predictions` expects a batch of predictions (i.e. a 2D array of shape (samples, 1000)). Found array with shape: (1, 1)"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# printing out the top label\n",
        "np.argmax(preds[0])\n"
      ],
      "metadata": {
        "id": "oXruuv1D4xLa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 192
        },
        "outputId": "918a78ae-3352-40a1-c8ee-22fcc301a9d4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-29-7d56f2f00eae>\u001b[0m in \u001b[0;36m<cell line: 2>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# printing out the top label\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpreds\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m: name 'preds' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Setting up a model that returns the last convolutional output\n",
        "\n",
        "last_conv_layer_name = \"block14_sepconv2_act\"\n",
        "classifier_layer_names = [ \"avg_pool\",\n",
        "                            \"predictions\",]\n",
        "last_conv_layer = model.get_layer(last_conv_layer_name)\n",
        "last_conv_layer_model = keras.Model(model.inputs, last_conv_layer.output)\n",
        "# plot_model(last_conv_layer_model)"
      ],
      "metadata": {
        "id": "mUwDupeh5UZJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# last_conv_layer_model.summary()"
      ],
      "metadata": {
        "id": "nDRbr73T6FET"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Reapplying the classifier on top of the last convolutional output\n",
        "\n",
        "\n",
        "classifier_input = keras.Input(shape=last_conv_layer.output.shape[1:])\n",
        "x = classifier_input\n",
        "for layer_name in classifier_layer_names:\n",
        "    x = model.get_layer(layer_name)(x)\n",
        "classifier_model = keras.Model(classifier_input, x)\n"
      ],
      "metadata": {
        "id": "VzruX5lB6Wwb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 245
        },
        "outputId": "a8ba62f2-eeed-408f-a08b-817ff6c6a41c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-31-0cee025090cb>\u001b[0m in \u001b[0;36m<cell line: 4>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 4\u001b[0;31m \u001b[0mclassifier_input\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mkeras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mInput\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlast_conv_layer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      5\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mclassifier_input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mlayer_name\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mclassifier_layer_names\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'last_conv_layer' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Retrieving the gradients of the top predicted class\n",
        "\n",
        "import tensorflow as tf\n",
        "\n",
        "with tf.GradientTape() as tape:\n",
        "    last_conv_layer_output = last_conv_layer_model(img_array)\n",
        "    tape.watch(last_conv_layer_output)\n",
        "    preds = classifier_model(last_conv_layer_output)\n",
        "    top_pred_index = tf.argmax(preds[0])\n",
        "    top_class_channel = preds[:, top_pred_index]\n",
        "\n",
        "grads = tape.gradient(top_class_channel, last_conv_layer_output) # this is the grad of top class wrt output feature map"
      ],
      "metadata": {
        "id": "qNOfPpMk9Cwl",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 245
        },
        "outputId": "b29bc7ba-122c-4d80-c897-b33f8a53e558"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-32-f72a288c85c0>\u001b[0m in \u001b[0;36m<cell line: 5>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mGradientTape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mtape\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m     \u001b[0mlast_conv_layer_output\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlast_conv_layer_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimg_array\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m     \u001b[0mtape\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlast_conv_layer_output\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m     \u001b[0mpreds\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mclassifier_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlast_conv_layer_output\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'last_conv_layer_model' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "![picture](https://datascienceimages.s3.eu-north-1.amazonaws.com/Modern_CNN_Architectures/Grad_CAM.png?response-content-disposition=inline&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEHQaCmV1LW5vcnRoLTEiRzBFAiEAmOZQhT%2BYM40tIrgSCC0juQ3jYb2AAM%2F1Fbz6dD8McLICIHg%2Fvrzjk8lvADAhmWpZrYL9TF3PwvkzKpdWA8hZeRQoKu0CCM3%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEQABoMOTc1MDUwMDY4NjU4IgyeGfj2LlL0RNsVYPMqwQJM4d9h4CmJwEyTBC7qZEzc29kl0a%2FrmErIOHAztB%2FrvmsLwfZSviafkN0S0Rgnb9ZUlQLlOLgmJAh%2Boo5wqCAD6Vph5HKDP0UmiithH%2BlXOUIiDE51ykTUEA3w1iSSHMyOs7vDBsLp%2FMFpvpCG3V2XnlQ5Iqgo422ee763aDs%2FzuanCRZpfGDaLTo%2BgTWY0vQntjWJILYrZeOw0tpLpzZArag7KROagc0J8JjBvK0M%2Fz02ApRdrd4sspTDVFraDlGzFJekxjfCDEkf0gSQy7gutoZ2hujhd%2BonPc2Sob988wEY1BUTJWathwBoXphrbMZIS6LhUp1rTnDl8w7hv54XZ4ORo9w4oXVsZdG%2FDbp2dUX6yVHFY3mjMGdU6fEQ1j2AlzySCTm4T%2FzJpXst4CisnbG6opNd6Os2b78VkYfQnKAwtdPmsQY6swKV5d%2BjoEU9poWR0HUudhbWdbIOfv1xKG64htIRW3NpNgi0OonJi5zOPTw5aWoAvi5hbnA7akONMJtwKRTgl9t30qtpFA8DwvmNL%2BUt%2FNsayqUy%2BrKPtD%2FtfVYyu82iXyxz4EyCSXPTgcCREW61a0z%2F8HtRH76JayEUAc7PAypjS861xn7SFWDzldlYWE8H2bDvjOQdf82VBKm67cP5CPWaxKpPkpTJFAR82hOQeFordG1wctY1YO5AOZnqv8Wd%2B7Tz9rQYNYk9xUihZZHgrvYzMuQWKcj%2By2eMBR7uWxP5LR48KBnL6edR4ZCv5H5LYCGjIREzJAImRO9qbDptQLHW%2B943UkCxOSqNdCIQjgRoWe84M93lB6rTQjLowcqWYU9YJDY9V73Jfw8Qz3mx6HxtnhdV&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20240507T045908Z&X-Amz-SignedHeaders=host&X-Amz-Expires=300&X-Amz-Credential=ASIA6GBMDKKZK2QSB44A%2F20240507%2Feu-north-1%2Fs3%2Faws4_request&X-Amz-Signature=85bb3d48e78de63b090f1daa9ccc216b1362d46b3d74e2c777c5f65d573acb0d)"
      ],
      "metadata": {
        "id": "eW-IgUQmP3O9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Gradient pooling and channel-importance weighting\n",
        "\n",
        "# take an avg of the grads - formula for w_k\n",
        "pooled_grads = tf.reduce_mean(grads, axis=(0, 1, 2)).numpy()\n",
        "last_conv_layer_output = last_conv_layer_output.numpy()[0]\n",
        "\n",
        "# multiply weights with feature maps\n",
        "for i in range(pooled_grads.shape[-1]):\n",
        "    last_conv_layer_output[:, :, i] *= pooled_grads[i]\n",
        "\n",
        "# compute weighted sum\n",
        "heatmap = np.mean(last_conv_layer_output, axis=-1)"
      ],
      "metadata": {
        "id": "g7gD_e-T9HDU",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 245
        },
        "outputId": "ac382a73-a383-441f-b750-0764f7f17047"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-33-af1678a25676>\u001b[0m in \u001b[0;36m<cell line: 4>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;31m# take an avg of the grads - formula for w_k\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 4\u001b[0;31m \u001b[0mpooled_grads\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreduce_mean\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgrads\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnumpy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      5\u001b[0m \u001b[0mlast_conv_layer_output\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlast_conv_layer_output\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnumpy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'grads' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Heatmap post-processing\n",
        "\n",
        "heatmap = np.maximum(heatmap, 0)\n",
        "heatmap /= np.max(heatmap)\n",
        "plt.matshow(heatmap)"
      ],
      "metadata": {
        "id": "Ff0Ii0Nq9Mmu",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 245
        },
        "outputId": "9086ae62-f3b9-44ab-d9ac-1b3be2c6b874"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-34-bb4f57d270ca>\u001b[0m in \u001b[0;36m<cell line: 3>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# Heatmap post-processing\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mheatmap\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmaximum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mheatmap\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      4\u001b[0m \u001b[0mheatmap\u001b[0m \u001b[0;34m/=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mheatmap\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmatshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mheatmap\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'heatmap' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Superimposing the heatmap on the original picture\n",
        "\n",
        "import matplotlib.cm as cm\n",
        "\n",
        "img = keras.utils.load_img(img_path)\n",
        "img = keras.utils.img_to_array(img)\n",
        "\n",
        "heatmap = np.uint8(255 * heatmap)\n",
        "\n",
        "jet = cm.get_cmap(\"jet\")\n",
        "jet_colors = jet(np.arange(256))[:, :3]\n",
        "jet_heatmap = jet_colors[heatmap]\n",
        "\n",
        "jet_heatmap = keras.utils.array_to_img(jet_heatmap)\n",
        "jet_heatmap = jet_heatmap.resize((img.shape[1], img.shape[0]))\n",
        "jet_heatmap = keras.utils.img_to_array(jet_heatmap)\n",
        "\n",
        "superimposed_img = jet_heatmap * 0.4 + img\n",
        "superimposed_img = keras.utils.array_to_img(superimposed_img)\n",
        "\n",
        "save_path = \"cat.jpg\"\n",
        "superimposed_img.save(save_path)"
      ],
      "metadata": {
        "id": "keru57IC9PeA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.matshow(superimposed_img)"
      ],
      "metadata": {
        "id": "sUWpJwxl9U28"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's see more gradCAM results\n",
        "\n",
        "Some good readings:\n",
        "* [blog](https://towardsdatascience.com/understand-your-algorithm-with-grad-cam-d3b62fce353#:~:text=Gradient%2Dweighted%20Class%20Activation%20Mapping,regions%20in%20the%20image%20for)"
      ],
      "metadata": {
        "id": "pxUW2yKcCyOF"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Self-Practice Problem:\n",
        "Solve an image classification problem on the cats-vs-dogs dataset by training 'mini-Exception-like' model based on the instructions given below:\n",
        "\n",
        "\n",
        "1.  Set the global random seed to 42.\n",
        "2.  We are using a **cat-vs-dogs** dataset here. You will have to download it using instruction.\n",
        "\n",
        "  Download the data in your notebook\n",
        "\n",
        "\n",
        "3.  Building the Model based on this [model_summary](https://indianinstituteofscience-my.sharepoint.com/:t:/g/personal/rohitc1_iisc_ac_in/EZ36t8eQFu9MrBnPjwueKcABD-2_8AZyDpyJ3vJEqUqlLQ?e=njNCGf) and its corresponding [model_plot](https://indianinstituteofscience-my.sharepoint.com/:i:/g/personal/rohitc1_iisc_ac_in/EU2WCnpqi8BEtfzltqI2vc4B5OFx53lMwn2tv6gqMebTig?e=lSsbz6). Ensure that you follow the trailing instructions:(16)\n",
        "\n",
        "      i).   For the initial layers of model mentioned in this [summary](https://indianinstituteofscience-my.sharepoint.com/:i:/g/personal/ksumanth_iisc_ac_in/EYDi7MfYgkRGvb-NqKKuGiABon9CyOUMEiffHac1sXyEsg?e=gYNdVB), random flip (horizontal), random rotation of 0.1, random zoom of 0.2, rescaling by 1./255, and set kernel_size = 5 and use_bias=False in the convolution layer.\n",
        "\n",
        "      ii).  Define a block of following layers:\n",
        "\n",
        "          *   Batch Normalization layer\n",
        "          *   Activation layer with relu as activation function\n",
        "          *   Depth wise separable layer (kernel size = 3)\n",
        "          *   Batch Normalization layer\n",
        "          *   Activation layer with relu as activation function\n",
        "          *   Depth wise separable layer (kernel size = 5)\n",
        "          *   Batch Normalization layer\n",
        "          *   Activation layer with relu as activation function\n",
        "          *   Depth wise separable layer (kernel size = 7)\n",
        "          *   Maxpool2D layer (poolsize =3, stride=2)\n",
        "          *   Convolution layer\n",
        "          *   'Add layer' due to a residual connection. Infer connection points of the skip connection from the model summary and model plot.\n",
        "\n",
        "          **Infer unspecified arguments from the summary**\n",
        "\n",
        "      iii). The block defined in (ii) repeats 4 times. Note that in each repitition, the number of filters changes. Infer this from the model plot/summary.\n",
        "\n",
        "      iv).  The last two layers are GlobalAveragePooling and Dense layers. The dense layer is the output layer (infer the number of neurons and the activation function).\n",
        "4. Compile model with rmsprop as an optimizer with appropriate loss and metric for this respective problem.\n",
        "5. Fit the model with a batch_size of 32 for 20 epochs. (Don't use EarlyStopping callback). Use the validation dataset from the data you downloaded. We have specified a small no. of epochs because training may take time. Try running colab on GPU by going to Edit > Notebook accelerator > Hardware Accelerator > GPU.\n",
        "6. Return the history as a DataFrame. Show loss and accuracy for training and validation through appropriate plots.\n",
        "7. Evaluate the Model on test dataset from the data you downloaded.\n",
        "\n",
        "**Note**:\n",
        "\n",
        "1. If you are using any parameter values or arguments apart from the ones mentioned or the ones that you must infer, state explicitly where and why you are using them.\n",
        "\n",
        "2. Also verify that the total no. of params of your model are the same as that mentioned in model_summary txt file given to you\n"
      ],
      "metadata": {
        "id": "PVfYIE6WqEcy"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "5ygvBjbiuIMU"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "uQnqMas5uIhZ"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}