Meet-5/SudokuSolver_P2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "SudokuSolver-Part2-NN.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "b29801c65d39453fa46b904cc8e85a3a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_ba113567a90a4e5a84014b2446b9fb57",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_a4ee399db8314db7ade7287ce0c3708e",
              "IPY_MODEL_78449a8433fe4519aee487478c41e665",
              "IPY_MODEL_40f7db34c30e4829bc67e2f4e9024c66"
            ]
          }
        },
        "ba113567a90a4e5a84014b2446b9fb57": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "a4ee399db8314db7ade7287ce0c3708e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_1475d2a6c4db4deb8688bb01fbd2a556",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "100%",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_bb7a09d10ec94efc9ad35f8eec20b0a8"
          }
        },
        "78449a8433fe4519aee487478c41e665": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_44aa98ba45b24acb8068f0f4685c7689",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 160,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 160,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_c321281b83564b7892941bd0227d2872"
          }
        },
        "40f7db34c30e4829bc67e2f4e9024c66": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_68bdc62c3b634a79bef134c62fda5b47",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 160/160 [00:05&lt;00:00, 30.34it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_48378e4e04c64119ad37d2f96081f385"
          }
        },
        "1475d2a6c4db4deb8688bb01fbd2a556": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "bb7a09d10ec94efc9ad35f8eec20b0a8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "44aa98ba45b24acb8068f0f4685c7689": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "c321281b83564b7892941bd0227d2872": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "68bdc62c3b634a79bef134c62fda5b47": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "48378e4e04c64119ad37d2f96081f385": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "5acafaa90c1a41eb9bbea80ff1c3b520": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_0959b940596d4226a31178d479f853db",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_f4aa0932e6f047baa28e06c815b9e955",
              "IPY_MODEL_05502ecbf3af4cc2b3b4a143e8dd4824",
              "IPY_MODEL_eb7b756b16e44caca91b7307f0d5f28a"
            ]
          }
        },
        "0959b940596d4226a31178d479f853db": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "f4aa0932e6f047baa28e06c815b9e955": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_2e7d6400242e4f168a72c473361b0f35",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "100%",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_89c5be7684e7476fb13300916e01e177"
          }
        },
        "05502ecbf3af4cc2b3b4a143e8dd4824": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_23b263c1e873454f985b97f60965d600",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 40,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 40,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_f939673711f7417d81383d6bc42038e9"
          }
        },
        "eb7b756b16e44caca91b7307f0d5f28a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_42c59a885b5e47b0a598b6e8a4321259",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 40/40 [00:01&lt;00:00, 30.70it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_4845d23f75c9472f91cd839ec09ab47a"
          }
        },
        "2e7d6400242e4f168a72c473361b0f35": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "89c5be7684e7476fb13300916e01e177": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "23b263c1e873454f985b97f60965d600": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "f939673711f7417d81383d6bc42038e9": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "42c59a885b5e47b0a598b6e8a4321259": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "4845d23f75c9472f91cd839ec09ab47a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "cb37b67d81b440ec8c8cc6ddbb5a035a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_fc1c9dbc4c424d9f9ffd3920b33ab170",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_1e21bae50896430d87ad16ce95c5743e",
              "IPY_MODEL_836aa5b422dc4ba38b18669285c87d6c",
              "IPY_MODEL_c61e6523919c4628b7c22a529413ed13"
            ]
          }
        },
        "fc1c9dbc4c424d9f9ffd3920b33ab170": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "1e21bae50896430d87ad16ce95c5743e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_50235fb5a5164f31855b106e2c4b0cb6",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_ca99346708a0417f8d9b92026623561d"
          }
        },
        "836aa5b422dc4ba38b18669285c87d6c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_53d11754b49943bf9e3c293d6dfaa472",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_1494a011a268490aa2060b7ddbe0d4d2"
          }
        },
        "c61e6523919c4628b7c22a529413ed13": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_b5b49d4629674f5a85edd8d40c212f40",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 160/? [00:01&lt;00:00, 112.77it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_7de16fa1ae6e4d198c9d11c6e49e2f04"
          }
        },
        "50235fb5a5164f31855b106e2c4b0cb6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "ca99346708a0417f8d9b92026623561d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "53d11754b49943bf9e3c293d6dfaa472": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "1494a011a268490aa2060b7ddbe0d4d2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": "20px",
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "b5b49d4629674f5a85edd8d40c212f40": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "7de16fa1ae6e4d198c9d11c6e49e2f04": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "c52e3b1cf6e849fdad38f55fa694aae1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_30a19021bc9f42419edfc9123e38ef80",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_d61f7289990342b5ae4bedc789ab1fec",
              "IPY_MODEL_9831bdab9fad446cbb9998a0f657f66e",
              "IPY_MODEL_8c1e8e20472f4962acdab625de5950db"
            ]
          }
        },
        "30a19021bc9f42419edfc9123e38ef80": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "d61f7289990342b5ae4bedc789ab1fec": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_d7275b988f64489aa280b8b4cf604ce3",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": "",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_e356a4b7dd7240cca16af526d61bb2c6"
          }
        },
        "9831bdab9fad446cbb9998a0f657f66e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_5c6a62a4d60644a789c7a31c76762757",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_fc15b298d3334d8c9ef191a3e4ba4d80"
          }
        },
        "8c1e8e20472f4962acdab625de5950db": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_1c2af08b7fc644dda67194fa194d1004",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 40/? [00:00&lt;00:00, 109.78it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_c12f350f02684f24bf438f3d2c98fdf4"
          }
        },
        "d7275b988f64489aa280b8b4cf604ce3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "e356a4b7dd7240cca16af526d61bb2c6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "5c6a62a4d60644a789c7a31c76762757": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "fc15b298d3334d8c9ef191a3e4ba4d80": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": "20px",
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "1c2af08b7fc644dda67194fa194d1004": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "c12f350f02684f24bf438f3d2c98fdf4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Building a Sudoku Solver Part 2: Deep Neural Networks\n",
        "---\n",
        "A quick recap of the steps needed to do solve sudoku from images: \n",
        "\n",
        "1. Finding the largest square in the image\n",
        "2. Unskewing the image\n",
        "3. Extracting the cells\n",
        "4. Deciding if each cell contains a digit or not\n",
        "5. Applying an ML model like CNN or a simple neural network to find the digits\n",
        "6. Solving the sudoku\n",
        "\n",
        "Steps 1-3 use out-of-the-box well-defined computer vision methods. <br />\n",
        "We used logistic regression for step 4 [here](https://github.com/ssncodingclub/Machine-Learning/blob/master/Meet-4/SudokuSolver_Part1.ipynb) and it worked great! <br />But for step 5, it only produced a 57% accuracy, meaning around half the time the digits were misclassified.\n",
        "\n",
        "❓ How do we make a model that is more accurate in classifying the digits present in the sudoku cells that we extract?\n",
        "\n",
        "> We can try using neural networks, a class of models so powerful that they can allow us to skip step 4, directly classyfing the digit present (1-9) and if there is no digit at all. "
      ],
      "metadata": {
        "id": "sAEe-AUlxcNH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Setup code:\n",
        "\n",
        "The code below basically takes the sudoku images, applies step 1-3 and gives us a dataset consisting of 32x32 pixel images of the sudoku cells, having either the digit 1-9 or no digit at all (let's consider this label as 0)"
      ],
      "metadata": {
        "id": "GLBVLi7RvMQ7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Installing required packages"
      ],
      "metadata": {
        "id": "RGp9y3e4WTD3"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sNnxDXXHUlbE",
        "outputId": "0ef97500-f262-45c2-e055-f0b4c14d1c07"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting wget\n",
            "  Downloading wget-3.2.zip (10 kB)\n",
            "Building wheels for collected packages: wget\n",
            "  Building wheel for wget (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for wget: filename=wget-3.2-py3-none-any.whl size=9675 sha256=2adb491ffe3e5fd8ad782b62df7050c74fbbfc45740fd89060992b2fdf09b5a3\n",
            "  Stored in directory: /root/.cache/pip/wheels/a1/b6/7c/0e63e34eb06634181c63adacca38b79ff8f35c37e3c13e3c02\n",
            "Successfully built wget\n",
            "Installing collected packages: wget\n",
            "Successfully installed wget-3.2\n"
          ]
        }
      ],
      "source": [
        "!pip install wget"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Importing required packages"
      ],
      "metadata": {
        "id": "NwUlM_ipWWqO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import cv2\n",
        "import numpy as np\n",
        "import wget\n",
        "import bz2\n",
        "import random\n",
        "\n",
        "from glob import glob\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from tqdm.auto import tqdm"
      ],
      "metadata": {
        "id": "uQV3O_EAU9EQ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Download the zip files"
      ],
      "metadata": {
        "id": "RhTBPcPUYoRV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_zip = wget.download('https://github.com/jeffreywolberg/sudoku_dataset/blob/master/datasets/v2_train.tar.bz2?raw=true')\n",
        "test_zip = wget.download('https://github.com/jeffreywolberg/sudoku_dataset/blob/master/datasets/v2_test.tar.bz2?raw=true')\n",
        "csv_data = wget.download('https://raw.githubusercontent.com/jeffreywolberg/sudoku_dataset/master/outlines_sorted.csv')"
      ],
      "metadata": {
        "id": "AYLwACcFWzC4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Extract and move the zip files to Data folder"
      ],
      "metadata": {
        "id": "ZRhVBLYeYrqc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!if [ -d '/content/Data/' ]; then rm -rf /content/Data else ; fi\n",
        "!mkdir Data \n",
        "!tar -xf /content/v2_train.tar.bz2 -C /content/Data/ && rm -rf /content/v2_train.tar.bz2\n",
        "!tar -xf /content/v2_test.tar.bz2 -C /content/Data/  && rm -rf /content/v2_test.tar.bz2\n",
        "!mv outlines_sorted.csv /content/Data"
      ],
      "metadata": {
        "id": "kOI6OSHCXV-d"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Helper functions"
      ],
      "metadata": {
        "id": "hdvkPmNqc8Bo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def largest4SideContour(image):\n",
        "    contours, h = cv2.findContours(\n",
        "        image, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)\n",
        "\n",
        "    # Sorting contours in descending order based on the area\n",
        "    contours = sorted(contours, key=cv2.contourArea, reverse=True)\n",
        "\n",
        "    # Looking at top 5 contours and checking if any of them are of size 4\n",
        "    for cnt in contours[:min(5, len(contours))]:\n",
        "        if len(approx(cnt)) == 4:\n",
        "            return cnt\n",
        "    return None\n",
        "\n",
        "\n",
        "def approx(cnt):\n",
        "    try:\n",
        "        peri = cv2.arcLength(cnt, True)\n",
        "        app = cv2.approxPolyDP(cnt, 0.01 * peri, True)\n",
        "        return app\n",
        "    except:\n",
        "        return None\n",
        "\n",
        "\n",
        "def get_rectangle_corners(cnt):\n",
        "    pts = cnt.reshape(4, 2)\n",
        "    rect = np.zeros((4, 2), dtype=\"float32\")\n",
        "\n",
        "    # the top-left point has the smallest sum whereas the\n",
        "    # bottom-right has the largest sum\n",
        "    s = pts.sum(axis=1)\n",
        "    rect[0] = pts[np.argmin(s)]\n",
        "    rect[2] = pts[np.argmax(s)]\n",
        "\n",
        "    # compute the difference between the points -- the top-right\n",
        "    # will have the minumum difference and the bottom-left will\n",
        "    # have the maximum difference\n",
        "    diff = np.diff(pts, axis=1)\n",
        "    rect[1] = pts[np.argmin(diff)]\n",
        "    rect[3] = pts[np.argmax(diff)]\n",
        "    return rect\n",
        "\n",
        "\n",
        "def warp_perspective(rect, grid):\n",
        "    (tl, tr, br, bl) = rect\n",
        "    widthA = np.sqrt(((br[0] - bl[0]) ** 2) + ((br[1] - bl[1]) ** 2))\n",
        "    widthB = np.sqrt(((tr[0] - tl[0]) ** 2) + ((tr[1] - tl[1]) ** 2))\n",
        "\n",
        "    # ...and now for the height of our new image\n",
        "    heightA = np.sqrt(((tr[0] - br[0]) ** 2) + ((tr[1] - br[1]) ** 2))\n",
        "    heightB = np.sqrt(((tl[0] - bl[0]) ** 2) + ((tl[1] - bl[1]) ** 2))\n",
        "\n",
        "    # take the maximum of the width and height values to reach\n",
        "    # our final dimensions\n",
        "    maxWidth = max(int(widthA), int(widthB))\n",
        "    maxHeight = max(int(heightA), int(heightB))\n",
        "\n",
        "    # construct our destination points which will be used to\n",
        "    # map the screen to a top-down, \"birds eye\" view\n",
        "    dst = np.array([\n",
        "        [0, 0],\n",
        "        [maxWidth - 1, 0],\n",
        "        [maxWidth - 1, maxHeight - 1],\n",
        "        [0, maxHeight - 1]], dtype=\"float32\")\n",
        "\n",
        "    # calculate the perspective transform matrix and warp\n",
        "    # the perspective to grab the screen\n",
        "    M = cv2.getPerspectiveTransform(rect, dst)\n",
        "    warp = cv2.warpPerspective(grid, M, (maxWidth, maxHeight))\n",
        "    return cv2.resize(warp, (288, 288))\n",
        "\n",
        "\n",
        "def getTopLine(image):\n",
        "    for i, row in enumerate(image):\n",
        "        if np.any(row):\n",
        "            return i\n",
        "    return None\n",
        "\n",
        "\n",
        "def getBottomLine(image):\n",
        "    for i in range(image.shape[0] - 1, -1, -1):\n",
        "        if np.any(image[i]):\n",
        "            return i\n",
        "    return None\n",
        "\n",
        "\n",
        "def getLeftLine(image):\n",
        "    for i in range(image.shape[1]):\n",
        "        if np.any(image[:, i]):\n",
        "            return i\n",
        "    return None\n",
        "\n",
        "\n",
        "def getRightLine(image):\n",
        "    for i in range(image.shape[1] - 1, -1, -1):\n",
        "        if np.any(image[:, i]):\n",
        "            return i\n",
        "    return None\n",
        "\n",
        "\n",
        "def rowShift(image, start, end, length):\n",
        "    shifted = np.zeros(image.shape)\n",
        "    if start + length < 0:\n",
        "        length = -start\n",
        "    elif end + length >= image.shape[0]:\n",
        "        length = image.shape[0] - 1 - end\n",
        "\n",
        "    for row in range(start, end + 1):\n",
        "        shifted[row + length] = image[row]\n",
        "    return shifted\n",
        "\n",
        "\n",
        "def colShift(image, start, end, length):\n",
        "    shifted = np.zeros(image.shape)\n",
        "    if start + length < 0:\n",
        "        length = -start\n",
        "    elif end + length >= image.shape[1]:\n",
        "        length = image.shape[1] - 1 - end\n",
        "\n",
        "    for col in range(start, end + 1):\n",
        "        shifted[:, col + length] = image[:, col]\n",
        "    return shifted\n",
        "\n",
        "\n",
        "def centerX(digit):\n",
        "    topLine = getTopLine(digit)\n",
        "    bottomLine = getBottomLine(digit)\n",
        "    if topLine is None or bottomLine is None:\n",
        "        return digit\n",
        "    centerLine = (topLine + bottomLine) >> 1\n",
        "    imageCenter = digit.shape[0] >> 1\n",
        "    digit = rowShift(\n",
        "        digit, start=topLine, end=bottomLine, length=imageCenter - centerLine)\n",
        "    return digit\n",
        "\n",
        "\n",
        "def centerY(digit):\n",
        "    leftLine = getLeftLine(digit)\n",
        "    rightLine = getRightLine(digit)\n",
        "    if leftLine is None or rightLine is None:\n",
        "        return digit\n",
        "    centerLine = (leftLine + rightLine) >> 1\n",
        "    imageCenter = digit.shape[1] >> 1\n",
        "    digit = colShift(\n",
        "        digit, start=leftLine, end=rightLine, length=imageCenter - centerLine)\n",
        "    return digit"
      ],
      "metadata": {
        "id": "Dog7DWcQdBfk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## More helper functions.."
      ],
      "metadata": {
        "id": "LRscIR-uciUQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def process_image(path: str):\n",
        "    # Load image\n",
        "    try:\n",
        "        img = cv2.imread(path)\n",
        "\n",
        "        img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\n",
        "        img = cv2.adaptiveThreshold(img.astype(np.uint8), 255, cv2.ADAPTIVE_THRESH_MEAN_C,\n",
        "                                    cv2.THRESH_BINARY, 11, 3)\n",
        "        img = 255 - img\n",
        "        kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (2, 2))\n",
        "        img = cv2.morphologyEx(img, cv2.MORPH_CLOSE, kernel)\n",
        "\n",
        "        contours, h = cv2.findContours(img, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)\n",
        "        max_contour = max(contours, key=cv2.contourArea)\n",
        "        x, y, w, h = cv2.boundingRect(max_contour)\n",
        "        img = img[y:y + h, x:x + w]\n",
        "        img = cv2.resize(img, (288, 288), interpolation=cv2.INTER_AREA)\n",
        "        img2 = img.copy()\n",
        "        largest = largest4SideContour(img2)\n",
        "        app = approx(largest)\n",
        "        if app is None:\n",
        "            return img\n",
        "        corners = get_rectangle_corners(app)\n",
        "        img = warp_perspective(corners, img)\n",
        "        return img\n",
        "    except Exception as e:\n",
        "        print(str(e))\n",
        "        return None\n",
        "\n",
        "def ground_truth(path: str):\n",
        "    with(open(path, 'r')) as fin:\n",
        "        lines = fin.readlines()\n",
        "        truth = []\n",
        "\n",
        "        # First line contains the details of phone\n",
        "        # Second line contains size of source image\n",
        "        # Skipping both these lines\n",
        "        for line in lines[2:]:\n",
        "            for value in line.split():\n",
        "                truth.append(int(value))\n",
        "\n",
        "        return np.array(truth) # > 0\n",
        "\n",
        "\n",
        "def clean(cell):\n",
        "    kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (2, 2))\n",
        "    cell = cv2.morphologyEx(cell, cv2.MORPH_CLOSE, kernel)\n",
        "    return cell\n",
        "\n",
        "\n",
        "def centerDigit(digit):\n",
        "    digit = centerX(digit)\n",
        "    digit = centerY(digit)\n",
        "    return digit\n",
        "\n",
        "\n",
        "def get_cells(sudoku, size=288):\n",
        "    sudoku = sudoku / 255.\n",
        "    cells = []\n",
        "    cell_shape = size // 9\n",
        "    for i in range(9):\n",
        "        i = cell_shape * i\n",
        "        for j in range(9):\n",
        "            j = j * cell_shape\n",
        "            cell = sudoku[i: i+cell_shape, j: j+cell_shape]\n",
        "            cell = clean(cell)\n",
        "            cell = centerDigit(cell)\n",
        "            cells.append(cell)\n",
        "\n",
        "    return np.array(cells)\n",
        "\n",
        "\n",
        "def compare_outputs(pred, truth):\n",
        "    errors = 0\n",
        "    for i in range(len(pred)):\n",
        "        if pred[i] != truth[i]:\n",
        "            errors += 1\n",
        "            break\n",
        "\n",
        "    if errors:\n",
        "        return 0\n",
        "    return 1\n",
        "\n",
        "\n",
        "def get_sudoku(path: str, model):\n",
        "    sudoku = process_image(path)\n",
        "\n",
        "    if sudoku is None:\n",
        "        raise Exception\n",
        "\n",
        "    cells = get_cells(sudoku, 288)\n",
        "    cells = np.reshape(cells, (cells.shape[0], 32, 32, 1))\n",
        "\n",
        "    # for cell in cells:\n",
        "    #   cell = cell.reshape((32, 32))\n",
        "    #   plt.imshow(cell)\n",
        "    #   plt.show()\n",
        "    #   v_pred = model.predict(cell.reshape((1, 32, 32, 1)))\n",
        "    #   print('Predicted: ',np.argmax(v_pred, axis=1))\n",
        "    #   input(\"Continue\")\n",
        "\n",
        "    v_pred = model.predict(cells)\n",
        "    v_pred = np.argmax(v_pred, axis=1)\n",
        "    sudoku_ext = np.reshape(v_pred, (9, 9))\n",
        "    return sudoku_ext"
      ],
      "metadata": {
        "id": "b8_SxzS9bKLh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Image before pre-processing"
      ],
      "metadata": {
        "id": "0tl0aDtKmLTw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt "
      ],
      "metadata": {
        "id": "U0x-otCNmP1k"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "img = plt.imread('/content/Data/v2_train/image10.jpg')\n",
        "plt.imshow(img)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "id": "zbP_zUDgmRDW",
        "outputId": "be864ae3-af1e-44e5-f3d8-60734ab68fed"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Image after pre-processing"
      ],
      "metadata": {
        "id": "YJ41oGgJmkZ-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "img = process_image('/content/Data/v2_train/image10.jpg')\n",
        "plt.imshow(img)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "id": "KlmjQxN6luUS",
        "outputId": "2a24e05f-8284-47b9-a502-3ab1efd0bdc0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "img.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OBwgxd-joDjQ",
        "outputId": "9abb2eae-26b3-46d7-f1c5-05b6830b4323"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(288, 288)"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cells_img = get_cells(img, size=288)"
      ],
      "metadata": {
        "id": "FnetgJo2neQg"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "cells_img.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oDYoL4CunisM",
        "outputId": "67f25266-4cc4-4c76-a580-654b2802f935"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(81, 32, 32)"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(cells_img[1])\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "id": "zKHRlotynnkE",
        "outputId": "206c280d-4ee9-4d2a-e012-9735a87be2ec"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Preparing the training dataset"
      ],
      "metadata": {
        "id": "jHvV7rQHeWLq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_train = []\n",
        "y_train = []\n",
        "y_train_all = []\n",
        "\n",
        "for path in tqdm(glob('/content/Data/v2_train/*.jpg')):\n",
        "  path = path[:-4]\n",
        "  processed_img = process_image(path + '.jpg')\n",
        "  if processed_img is None:\n",
        "    continue\n",
        "  X_train.append(processed_img)\n",
        "  y_train.append(ground_truth(path + '.dat') > 0)\n",
        "  y_train_all.append(ground_truth(path + '.dat'))\n",
        "\n",
        "X_train = np.array(X_train)\n",
        "y_train = np.array(y_train)\n",
        "y_train_all = np.array(y_train_all)\n",
        "\n",
        "X_test = []\n",
        "y_test = []\n",
        "y_test_all = []\n",
        "\n",
        "for path in tqdm(glob('/content/Data/v2_test/*.jpg')):\n",
        "  path = path[:-4]\n",
        "  processed_img = process_image(path + '.jpg')\n",
        "  if processed_img is None:\n",
        "    continue\n",
        "  X_test.append(processed_img)\n",
        "  y_test.append(ground_truth(path + '.dat') > 0)\n",
        "  y_test_all.append(ground_truth(path + '.dat'))\n",
        "\n",
        "X_test = np.array(X_test)\n",
        "y_test = np.array(y_test)\n",
        "y_test_all = np.array(y_test_all)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 81,
          "referenced_widgets": [
            "b29801c65d39453fa46b904cc8e85a3a",
            "ba113567a90a4e5a84014b2446b9fb57",
            "a4ee399db8314db7ade7287ce0c3708e",
            "78449a8433fe4519aee487478c41e665",
            "40f7db34c30e4829bc67e2f4e9024c66",
            "1475d2a6c4db4deb8688bb01fbd2a556",
            "bb7a09d10ec94efc9ad35f8eec20b0a8",
            "44aa98ba45b24acb8068f0f4685c7689",
            "c321281b83564b7892941bd0227d2872",
            "68bdc62c3b634a79bef134c62fda5b47",
            "48378e4e04c64119ad37d2f96081f385",
            "5acafaa90c1a41eb9bbea80ff1c3b520",
            "0959b940596d4226a31178d479f853db",
            "f4aa0932e6f047baa28e06c815b9e955",
            "05502ecbf3af4cc2b3b4a143e8dd4824",
            "eb7b756b16e44caca91b7307f0d5f28a",
            "2e7d6400242e4f168a72c473361b0f35",
            "89c5be7684e7476fb13300916e01e177",
            "23b263c1e873454f985b97f60965d600",
            "f939673711f7417d81383d6bc42038e9",
            "42c59a885b5e47b0a598b6e8a4321259",
            "4845d23f75c9472f91cd839ec09ab47a"
          ]
        },
        "id": "pg3LRbzMeZul",
        "outputId": "246253b4-361b-4621-a942-d4f18bba5103"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "b29801c65d39453fa46b904cc8e85a3a",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "  0%|          | 0/160 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "5acafaa90c1a41eb9bbea80ff1c3b520",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "  0%|          | 0/40 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* The 288 x 288 image here is the entire 9x9 sudoku grid, giving us 81 different 32x32 cells either containing a digit from 1-9 or no digit at all.\n",
        "\n",
        "* There are 160 such images for training and 40 for testing. Leading to 12960 (160 * 81) cells to train our model to detect what the cell has and 3240 (40 * 81) cells to validate our model on."
      ],
      "metadata": {
        "id": "lX0TeRSzzhNm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_train.shape, X_test.shape"
      ],
      "metadata": {
        "id": "X83SLFbkfDY4",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8cde76ea-0d9c-4c69-b7e0-fe4a359c2fb2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((160, 288, 288), (40, 288, 288))"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The y here is the digit contained (or 0 for no digit) in each of the 81 cells for each sudoku image. "
      ],
      "metadata": {
        "id": "uZPTc5II0SLu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "y_train.shape, y_test.shape, y_train_all.shape, y_test_all.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dtof0oTjk7Gk",
        "outputId": "0571f4d0-0a12-45f7-c1d4-2e45a72161be"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((160, 81), (40, 81), (160, 81), (40, 81))"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "digit_train_img = []\n",
        "digit_train_truth = []\n",
        "digit_train_truth_all = []\n",
        "\n",
        "for i, sudoku in tqdm(enumerate(X_train)):\n",
        "\t\tdigit_train_img.extend(get_cells(sudoku, size=288))\n",
        "\t\tdigit_train_truth.extend(y_train[i])\n",
        "\t\tdigit_train_truth_all.extend(y_train_all[i])\n",
        "\n",
        "digit_train_truth = np.array(digit_train_truth)\n",
        "digit_train_img = np.array(digit_train_img)\n",
        "digit_train_truth_all = np.array(digit_train_truth_all)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "cb37b67d81b440ec8c8cc6ddbb5a035a",
            "fc1c9dbc4c424d9f9ffd3920b33ab170",
            "1e21bae50896430d87ad16ce95c5743e",
            "836aa5b422dc4ba38b18669285c87d6c",
            "c61e6523919c4628b7c22a529413ed13",
            "50235fb5a5164f31855b106e2c4b0cb6",
            "ca99346708a0417f8d9b92026623561d",
            "53d11754b49943bf9e3c293d6dfaa472",
            "1494a011a268490aa2060b7ddbe0d4d2",
            "b5b49d4629674f5a85edd8d40c212f40",
            "7de16fa1ae6e4d198c9d11c6e49e2f04"
          ]
        },
        "id": "SMHvrDQTlEIM",
        "outputId": "094314a1-b8d4-477a-b9c2-2aed2d593505"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "cb37b67d81b440ec8c8cc6ddbb5a035a",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "0it [00:00, ?it/s]"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "digit_train_img.shape, digit_train_truth.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QNkFi2IeoyrE",
        "outputId": "9eaef6ab-a3d1-4e9f-84c3-b6a77f94a829"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((12960, 32, 32), (12960,))"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "digit_test_img = []\n",
        "digit_test_truth = []\n",
        "digit_test_truth_all = []\n",
        "\n",
        "for i, sudoku in tqdm(enumerate(X_test)):\n",
        "\t\tdigit_test_img.extend(get_cells(sudoku, size=288))\n",
        "\t\tdigit_test_truth.extend(y_test[i])\n",
        "\t\tdigit_test_truth_all.extend(y_test_all[i])\n",
        "\n",
        "digit_test_truth = np.array(digit_test_truth)\n",
        "digit_test_img = np.array(digit_test_img)\n",
        "digit_test_truth_all = np.array(digit_test_truth_all)"
      ],
      "metadata": {
        "id": "jvjkJOVyo7D3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "c52e3b1cf6e849fdad38f55fa694aae1",
            "30a19021bc9f42419edfc9123e38ef80",
            "d61f7289990342b5ae4bedc789ab1fec",
            "9831bdab9fad446cbb9998a0f657f66e",
            "8c1e8e20472f4962acdab625de5950db",
            "d7275b988f64489aa280b8b4cf604ce3",
            "e356a4b7dd7240cca16af526d61bb2c6",
            "5c6a62a4d60644a789c7a31c76762757",
            "fc15b298d3334d8c9ef191a3e4ba4d80",
            "1c2af08b7fc644dda67194fa194d1004",
            "c12f350f02684f24bf438f3d2c98fdf4"
          ]
        },
        "outputId": "9de7ad37-7dd6-46ab-d5a9-c61460d059f4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "c52e3b1cf6e849fdad38f55fa694aae1",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "0it [00:00, ?it/s]"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "digit_test_img.shape, digit_test_truth.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3JFdGdgIwRsK",
        "outputId": "31f1959d-c14e-4df2-90ca-0257b834a9d6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((3240, 32, 32), (3240,))"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "digit_train_img = digit_train_img.reshape(-1, 1024)\n",
        "digit_test_img = digit_test_img.reshape(-1, 1024)"
      ],
      "metadata": {
        "id": "VTyMb2pkw4XK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "digit_train_img.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cB6EUiO_xE0o",
        "outputId": "b2176e30-f11c-478f-ed48-cac4582b5515"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(12960, 1024)"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Finally, we obtain the dataset consisting of 12960 different 32x32 training images and the 12960 labels corresponding to them. Since our models take in a fixed-length vector of number, we flatten the 32x32 matrix into a 1024-element-long vector."
      ],
      "metadata": {
        "id": "-duFvvcc0cBM"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Using neural networks for this task\n",
        "\n"
      ],
      "metadata": {
        "id": "r4PxNs-NXQ0V"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# just to do one-hot encoding\n",
        "from tensorflow.keras.utils import to_categorical"
      ],
      "metadata": {
        "id": "25a18YVRZJK8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "training_data = list(zip(digit_train_img, to_categorical(digit_train_truth_all)))\n",
        "test_data = list(zip(digit_test_img, digit_test_truth_all))"
      ],
      "metadata": {
        "id": "MXKo7syUXd_V"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 💻 Coding a neural network from scratch:\n",
        "\n",
        "This is a simple implementation from [neuralnetworksanddeeplearning.com](http://neuralnetworksanddeeplearning.com/chap1.html) (a great resource to learn NNs deeply) adapted to Python 3. Added code to track training accuracy as well. "
      ],
      "metadata": {
        "id": "lpS645g0f_lg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class Network(object):\n",
        "    def __init__(self, sizes):\n",
        "        self.num_layers = len(sizes)\n",
        "        self.sizes = sizes\n",
        "        self.biases = [np.random.randn(y) for y in sizes[1:]]\n",
        "        self.weights = [np.random.randn(y, x)\n",
        "                        for x, y in zip(sizes[:-1], sizes[1:])]\n",
        "\n",
        "    def feedforward(self, a):\n",
        "        for b, w in zip(self.biases, self.weights):\n",
        "            a = sigmoid(np.dot(w, a)+b)\n",
        "        return a\n",
        "\n",
        "    def SGD(self, training_data, epochs, mini_batch_size, eta,\n",
        "            test_data=None):\n",
        "        if test_data: n_test = len(test_data)\n",
        "        n = len(training_data)\n",
        "        for j in range(epochs):\n",
        "            random.shuffle(training_data)\n",
        "            mini_batches = [\n",
        "                training_data[k:k+mini_batch_size]\n",
        "                for k in range(0, n, mini_batch_size)]\n",
        "            for mini_batch in mini_batches:\n",
        "                self.update_mini_batch(mini_batch, eta)\n",
        "            if test_data:\n",
        "                correct_test = self.evaluate(test_data)\n",
        "                correct_train = self.evaluate_train(training_data)\n",
        "                print(\"Epoch {0}: Val {1}/{2} - Accuracy: {3}, Train {4}/{5} - Accuracy: {6}\".format(\n",
        "                    j, correct_test, n_test, round(correct_test / n_test, 3), \n",
        "                    correct_train, len(training_data), round(correct_train / len(training_data), 3)\n",
        "                ))\n",
        "            else:\n",
        "                print(\"Epoch {0} complete\".format(j))\n",
        "\n",
        "    def update_mini_batch(self, mini_batch, eta):\n",
        "        nabla_b = [np.zeros(b.shape) for b in self.biases]\n",
        "        nabla_w = [np.zeros(w.shape) for w in self.weights]\n",
        "        for x, y in mini_batch:\n",
        "            delta_nabla_b, delta_nabla_w = self.backprop(x, y)\n",
        "            nabla_b = [nb+dnb for nb, dnb in zip(nabla_b, delta_nabla_b)]\n",
        "            nabla_w = [nw+dnw for nw, dnw in zip(nabla_w, delta_nabla_w)]\n",
        "        self.weights = [w-(eta/len(mini_batch))*nw\n",
        "                        for w, nw in zip(self.weights, nabla_w)]\n",
        "        self.biases = [b-(eta/len(mini_batch))*nb\n",
        "                       for b, nb in zip(self.biases, nabla_b)]\n",
        "\n",
        "    def backprop(self, x, y):\n",
        "        nabla_b = [np.zeros(b.shape) for b in self.biases]\n",
        "        nabla_w = [np.zeros(w.shape) for w in self.weights]\n",
        "\n",
        "        activation = x\n",
        "        activations = [x]\n",
        "        zs = [] \n",
        "        for b, w in zip(self.biases, self.weights):\n",
        "            z = np.dot(w, activation)+b\n",
        "            zs.append(z)\n",
        "            activation = sigmoid(z)\n",
        "            activations.append(activation)\n",
        "        delta = self.cost_derivative(activations[-1], y) * \\\n",
        "            sigmoid_prime(zs[-1])\n",
        "        nabla_b[-1] = delta\n",
        "        nabla_w[-1] = np.dot(delta.reshape(-1, 1), activations[-2].reshape(-1, 1).transpose())\n",
        "        for l in range(2, self.num_layers):\n",
        "            z = zs[-l]\n",
        "            sp = sigmoid_prime(z)\n",
        "            delta = np.dot(self.weights[-l+1].transpose(), delta) * sp\n",
        "            nabla_b[-l] = delta\n",
        "            nabla_w[-l] = np.dot(delta.reshape(-1, 1), activations[-l-1].reshape(-1, 1).transpose())\n",
        "        return (nabla_b, nabla_w)\n",
        "\n",
        "    def evaluate(self, test_data):\n",
        "        test_results = [(np.argmax(self.feedforward(x)), y)\n",
        "                        for (x, y) in test_data]\n",
        "        return sum(int(x == y) for (x, y) in test_results)\n",
        "\n",
        "    def evaluate_train(self, training_data):\n",
        "        train_results = [(np.argmax(self.feedforward(x)), np.argmax(y))\n",
        "                        for (x, y) in training_data]\n",
        "        return sum(int(x == y) for (x, y) in train_results)\n",
        "\n",
        "    def cost_derivative(self, output_activations, y):\n",
        "        return (output_activations-y)\n",
        "\n",
        "def sigmoid(z):\n",
        "    return 1/(1 + np.exp(-z))\n",
        "\n",
        "def sigmoid_prime(z):\n",
        "    return sigmoid(z) * (1 - sigmoid(z))"
      ],
      "metadata": {
        "id": "nKySI_OcXTBn"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Training a neural network with one hidden layer:"
      ],
      "metadata": {
        "id": "RfDdKJ0fojM3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "net = Network([1024, 16, 10])\n",
        "net.SGD(training_data, 10, 64, 2, test_data=test_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZI282z8fXVp6",
        "outputId": "2e1ddcf8-b3c6-4eb0-efc1-268d85a55ef2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 0: Val 2087/3240 - Accuracy: 0.644, Train 8287/12960 - Accuracy: 0.639\n",
            "Epoch 1: Val 2115/3240 - Accuracy: 0.653, Train 8390/12960 - Accuracy: 0.647\n",
            "Epoch 2: Val 2136/3240 - Accuracy: 0.659, Train 8489/12960 - Accuracy: 0.655\n",
            "Epoch 3: Val 2141/3240 - Accuracy: 0.661, Train 8524/12960 - Accuracy: 0.658\n",
            "Epoch 4: Val 2147/3240 - Accuracy: 0.663, Train 8559/12960 - Accuracy: 0.66\n",
            "Epoch 5: Val 2157/3240 - Accuracy: 0.666, Train 8598/12960 - Accuracy: 0.663\n",
            "Epoch 6: Val 2170/3240 - Accuracy: 0.67, Train 8619/12960 - Accuracy: 0.665\n",
            "Epoch 7: Val 2169/3240 - Accuracy: 0.669, Train 8636/12960 - Accuracy: 0.666\n",
            "Epoch 8: Val 2192/3240 - Accuracy: 0.677, Train 8675/12960 - Accuracy: 0.669\n",
            "Epoch 9: Val 2182/3240 - Accuracy: 0.673, Train 8683/12960 - Accuracy: 0.67\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Using TensorFlow and Keras\n",
        "<center>\n",
        "<img width=\"600\" src=\"https://miro.medium.com/max/1000/1*LkKz4wtZNBo5i-Vc8DWhTA.png\" alt=\"kerasimage\"/>\n",
        "</center>\n",
        "\n",
        "While for a simple example like above, coding from scratch would work, but as models become more complex, a library like [TensorFlow](www.tensorflow.org) might be more suitable. [Keras](keras.io) is a framework on top of TensorFlow that makes designing and testing neural networks even more simpler!\n"
      ],
      "metadata": {
        "id": "_68KPlGphYnn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.layers import Dense, ReLU\n",
        "from keras.models import Sequential\n",
        "from tensorflow.keras.optimizers import Adam"
      ],
      "metadata": {
        "id": "bIYzVnK6fLdj"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### ✏️ Design a simple feed-forward neural network\n",
        "\n",
        "Our model is a simple feed-forward neural network, where each layer takes input from the previous one and feeds the output to the next, so we use the Sequential model."
      ],
      "metadata": {
        "id": "7wjxsEEapQTi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = Sequential()"
      ],
      "metadata": {
        "id": "MIu60eTlpVn9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "1. To add a fully-connected hidden layer to our model, we can use the **Dense** layer.\n",
        "\n",
        "2. Since this is the first (hidden) layer of our model, we also mention the input shape (```input_shape```) - the dimensions of our input layer - in our case this is 1024 (the 32 x 32 image we have).\n",
        "\n",
        "3. Let's use a hidden layer of size 16, and then add an activation function. In Keras, Activation functions also work like Layers. Here we use the ReLU activation, sigmoid can be used all.\n",
        "\n",
        "4. Finally, our output layer has 10 neurons as we have 10 classes (1-9 being digits, 0 being no digit). We apply an activation function through the activation parameter rather than a separate layer. Both these ways of applying activations functions are acceptable and equivalent."
      ],
      "metadata": {
        "id": "OHKjE2BQpZMf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.add(Dense(16, input_shape=(digit_train_img.shape[1], )))\n",
        "model.add(ReLU()) # Can be sigmoid too\n",
        "model.add(Dense(10, activation='sigmoid'))"
      ],
      "metadata": {
        "id": "z1NdhPVPbacf"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Then we compile our model, mentioning what loss and optimizer we want to use along with any metrics we want to track.\n",
        "\n",
        "Since this is a classification problem, accuracy can be one metric to track."
      ],
      "metadata": {
        "id": "jS3SbsJptFu1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(loss='sparse_categorical_crossentropy', \n",
        "              optimizer=Adam(learning_rate=0.01), \n",
        "              metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "I4Nfol6KpXA7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.fit(digit_train_img, \n",
        "          digit_train_truth_all, \n",
        "          validation_data=(digit_test_img, digit_test_truth_all), \n",
        "          batch_size=128, \n",
        "          epochs=20)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lEDqkgr2b6bd",
        "outputId": "69c7508f-317d-43f3-8c1e-6e25eb8b2626"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "102/102 [==============================] - 1s 5ms/step - loss: 0.9946 - accuracy: 0.6879 - val_loss: 0.7289 - val_accuracy: 0.7651\n",
            "Epoch 2/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.7422 - accuracy: 0.7770 - val_loss: 0.6835 - val_accuracy: 0.7873\n",
            "Epoch 3/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.6723 - accuracy: 0.7969 - val_loss: 0.6332 - val_accuracy: 0.8090\n",
            "Epoch 4/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.6123 - accuracy: 0.8186 - val_loss: 0.5770 - val_accuracy: 0.8324\n",
            "Epoch 5/20\n",
            "102/102 [==============================] - 0s 4ms/step - loss: 0.5778 - accuracy: 0.8279 - val_loss: 0.5866 - val_accuracy: 0.8244\n",
            "Epoch 6/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.5455 - accuracy: 0.8384 - val_loss: 0.5426 - val_accuracy: 0.8336\n",
            "Epoch 7/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.5434 - accuracy: 0.8361 - val_loss: 0.5503 - val_accuracy: 0.8389\n",
            "Epoch 8/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.5086 - accuracy: 0.8539 - val_loss: 0.5535 - val_accuracy: 0.8364\n",
            "Epoch 9/20\n",
            "102/102 [==============================] - 0s 4ms/step - loss: 0.4861 - accuracy: 0.8599 - val_loss: 0.5140 - val_accuracy: 0.8556\n",
            "Epoch 10/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4876 - accuracy: 0.8593 - val_loss: 0.5327 - val_accuracy: 0.8475\n",
            "Epoch 11/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4661 - accuracy: 0.8694 - val_loss: 0.5248 - val_accuracy: 0.8528\n",
            "Epoch 12/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4662 - accuracy: 0.8645 - val_loss: 0.5113 - val_accuracy: 0.8543\n",
            "Epoch 13/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4442 - accuracy: 0.8728 - val_loss: 0.5146 - val_accuracy: 0.8593\n",
            "Epoch 14/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4511 - accuracy: 0.8688 - val_loss: 0.5063 - val_accuracy: 0.8645\n",
            "Epoch 15/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4351 - accuracy: 0.8750 - val_loss: 0.4923 - val_accuracy: 0.8679\n",
            "Epoch 16/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4246 - accuracy: 0.8798 - val_loss: 0.5019 - val_accuracy: 0.8657\n",
            "Epoch 17/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4226 - accuracy: 0.8791 - val_loss: 0.5137 - val_accuracy: 0.8577\n",
            "Epoch 18/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4277 - accuracy: 0.8776 - val_loss: 0.5141 - val_accuracy: 0.8623\n",
            "Epoch 19/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4156 - accuracy: 0.8817 - val_loss: 0.5230 - val_accuracy: 0.8593\n",
            "Epoch 20/20\n",
            "102/102 [==============================] - 0s 3ms/step - loss: 0.4064 - accuracy: 0.8819 - val_loss: 0.4899 - val_accuracy: 0.8710\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fdd7e6c3d90>"
            ]
          },
          "metadata": {},
          "execution_count": 129
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def get_weights(model, i):\n",
        "    ''' Helper function to retrieve a hidden layer nodes weight mapping to the input layer '''\n",
        "    return model.layers[0].get_weights()[0][:, i].reshape(32, 32)"
      ],
      "metadata": {
        "id": "7irkZiaAmlCy"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 📈 Visualizing the trained weights"
      ],
      "metadata": {
        "id": "AI94I9p-lsdZ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "After training our network, taking the first layer's weights (a 1024 * 16) matrix, we can find out what kind of patterns activate/trigger each of the 16 hidden layer nodes.\n",
        "\n",
        "This can be done by reshaping the 1024 x 1 weight vector for each hidden layer node into a 32 x 32 matrix and plotting it."
      ],
      "metadata": {
        "id": "lXDh7F8YmXTv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "ax = plt.figure(figsize=(10, 10))\n",
        "for i in range(16):\n",
        "    plt.subplot(4, 4, i+1)\n",
        "    plt.axis('off')\n",
        "    plt.imshow(get_weights(model, i), cmap='winter')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 575
        },
        "id": "NGMrgO_sjEOx",
        "outputId": "edbbb407-73f4-422d-db7c-f2886d73f9da"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 16 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* It's pretty clear that the model learns different lines and curves involved in classifying the digits (as well as the lack of any digit!)\n",
        "\n",
        "* What makes this interesting is the fact that we **never explicitly mentioned** or conveyed the neural network to learn these features, it learned it by itself. All we ever did was to program a method to reduce the number of mistakes the model makes, this drives/forces the network to learn these patterns in order to achieve that.\n",
        "\n",
        "* By restricting the number of hidden layer nodes to 16, we might not achieve the best performance but it gives us insight into what **patterns** a neural network learns to recognise when it's constrained."
      ],
      "metadata": {
        "id": "0g4Es_MMm6b_"
      }
    }
  ]
}