feed_forward_nn.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "feed_forward_nn.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "f96cdf6685db4f4aa62534f54b26bca6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_d2fb0b7b8eb940838c4a8e9b3d7b713e",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_b6e7694af3ec44b1b724814cc8fe1efe",
              "IPY_MODEL_3e1e2b13d35744e9a9b698464ec385a0"
            ]
          }
        },
        "d2fb0b7b8eb940838c4a8e9b3d7b713e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "b6e7694af3ec44b1b724814cc8fe1efe": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_9811462d814f48df9183462fd543f85c",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 1000,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1000,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_b5a04e75ce694e0b956f9ab44b71b9f7"
          }
        },
        "3e1e2b13d35744e9a9b698464ec385a0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_217cc3b1433c4662b690261da6ccaa82",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 1000/1000 [00:39&lt;00:00, 25.58epoch/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_2479b84f50ec4c3eb08324839bc4e044"
          }
        },
        "9811462d814f48df9183462fd543f85c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "b5a04e75ce694e0b956f9ab44b71b9f7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "217cc3b1433c4662b690261da6ccaa82": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "2479b84f50ec4c3eb08324839bc4e044": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "dff6115fc6604d3796017e9c0dad637d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_8b2ce44ced6a4621b420404888abd884",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_51d4fae1383e46868bea67b765edebf5",
              "IPY_MODEL_46728af4498b4bbcbbe8e02c44ce24f3"
            ]
          }
        },
        "8b2ce44ced6a4621b420404888abd884": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "51d4fae1383e46868bea67b765edebf5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_e920a23e10604a21a9fc3f02fd727f87",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 2000,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 2000,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_2b15fd3be57149a09ec794063f901a43"
          }
        },
        "46728af4498b4bbcbbe8e02c44ce24f3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_e009543d551346e6ae06d58630173e05",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 2000/2000 [09:32&lt;00:00,  3.49epoch/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_e80f947e2ce944c08cbe3d7a53a4c7a4"
          }
        },
        "e920a23e10604a21a9fc3f02fd727f87": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "2b15fd3be57149a09ec794063f901a43": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "e009543d551346e6ae06d58630173e05": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "e80f947e2ce944c08cbe3d7a53a4c7a4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "0abe98dcc1394c5eb3d19116e9c977e3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_77de76af9bcd426f849212663e169edb",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_7a7b1dadd7b24cc789f6eea7fee25ce9",
              "IPY_MODEL_c333945e501649f6b1ba3b697d629d61"
            ]
          }
        },
        "77de76af9bcd426f849212663e169edb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "7a7b1dadd7b24cc789f6eea7fee25ce9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_7b55ad4afa7a48cf96ede8d79b15e28b",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 1000,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1000,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_8193861f1d85416290bacde8ecd2fd1b"
          }
        },
        "c333945e501649f6b1ba3b697d629d61": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_0434d40777b6463daad8cfdb6e5e7b8e",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 1000/1000 [01:02&lt;00:00, 16.13epoch/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_13abbb44d0fd48a29e390b01441b33a6"
          }
        },
        "7b55ad4afa7a48cf96ede8d79b15e28b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "8193861f1d85416290bacde8ecd2fd1b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "0434d40777b6463daad8cfdb6e5e7b8e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "13abbb44d0fd48a29e390b01441b33a6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "5fe185cb3042459a8bf9a86cfaacb800": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_49966cfbd9a24ae381e448a905125267",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_a8b795defaef4f59a1efa6d1deaca54e",
              "IPY_MODEL_1b690d0a5c2b4cfeae5607ea92ef1af7"
            ]
          }
        },
        "49966cfbd9a24ae381e448a905125267": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "a8b795defaef4f59a1efa6d1deaca54e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_b4cf3ec616e9440a932e8996f945945c",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 2000,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 2000,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_4b296623d0e042e9abc8df02e6bdf1ae"
          }
        },
        "1b690d0a5c2b4cfeae5607ea92ef1af7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_64e8ee1d59a54a029bcddf2ccd790fff",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 2000/2000 [02:24&lt;00:00, 13.82epoch/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_ae2b49fa855145b1af64b6efde4da6df"
          }
        },
        "b4cf3ec616e9440a932e8996f945945c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "4b296623d0e042e9abc8df02e6bdf1ae": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "64e8ee1d59a54a029bcddf2ccd790fff": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "ae2b49fa855145b1af64b6efde4da6df": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6kfHkRrDtbTL",
        "colab_type": "text"
      },
      "source": [
        "# Importing Packages"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "toXDJOG4BYhM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.colors\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score, mean_squared_error, log_loss\n",
        "from tqdm import tqdm_notebook\n",
        "\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "from sklearn.datasets import make_blobs"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Jyr-6D7atVfS",
        "colab_type": "text"
      },
      "source": [
        "# Class for Sigmoid Neuron"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lmi97_mvl5vI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class SigmoidNeuron:\n",
        "    def __init__(self):\n",
        "        self.w = None\n",
        "        self.b = None\n",
        "    \n",
        "    def perceptron(self, x):\n",
        "        return np.dot(x, self.w.T) + self.b\n",
        "\n",
        "    def sigmoid(self, x):\n",
        "        return 1.0/(1.0+np.exp(-x))\n",
        "\n",
        "    def grad_w_mse(self,x,y):\n",
        "        y_pred = self.sigmoid(self.perceptron(x))\n",
        "        return (y_pred-y) * y_pred * (1-y_pred) * x\n",
        "    \n",
        "    def grad_b_mse(self,x,y):\n",
        "        y_pred = self.sigmoid(self.perceptron(x))\n",
        "        return (y_pred-y) * y_pred * (1-y_pred)\n",
        "\n",
        "    def grad_w_ce(self,x,y):\n",
        "        y_pred = self.sigmoid(self.perceptron(x))\n",
        "        if y==0:\n",
        "            return y_pred * x\n",
        "        elif y==1:\n",
        "            return -1 * (1-y_pred) * x\n",
        "        else:\n",
        "            raise ValueError(\"y should be either 0 or 1\")\n",
        "    \n",
        "    def grad_b_ce(self,x,y):\n",
        "        y_pred = self.sigmoid(self.perceptron(x))\n",
        "        if y==0:\n",
        "            return y_pred\n",
        "        elif y==1:\n",
        "            return -1 * (1-y_pred)\n",
        "        else:\n",
        "            raise ValueError(\"y should be either 0 or 1\")\n",
        "\n",
        "    def fit(self, X, Y, epochs=1, learning_rate=1, initialise=True, loss_fn='mse', display_loss=False):\n",
        "\n",
        "        if initialise:\n",
        "            self.w = np.random.randn(1, X.shape[1])\n",
        "            self.b = 0\n",
        "        \n",
        "        if display_loss:\n",
        "            loss = {}\n",
        "\n",
        "        for i in tqdm_notebook(range(epochs), total=epochs, unit='epoch'):\n",
        "            dw = 0\n",
        "            db = 0\n",
        "            for x,y in zip(X,Y):\n",
        "                if loss_fn == \"mse\":\n",
        "                    dw += self.grad_w_mse(x,y)\n",
        "                    db += self.grad_b_mse(x,y)\n",
        "                elif loss_fn == \"ce\":\n",
        "                    dw += self.grad_w_ce(x,y)\n",
        "                    db += self.grad_b_ce(x,y)\n",
        "            \n",
        "            m = X.shape[1]\n",
        "            self.w -= learning_rate*dw/m\n",
        "            self.b -= learning_rate*db/m\n",
        "\n",
        "            if display_loss:\n",
        "                Y_pred = self.sigmoid(self.perceptron(X))\n",
        "                if loss_fn == \"mse\":\n",
        "                    loss[i] = mean_squared_error(Y, Y_pred)\n",
        "                elif loss_fn == \"ce\":\n",
        "                    loss[i] = log_loss(Y, Y_pred)\n",
        "\n",
        "        if display_loss:\n",
        "            plt.plot(np.array(list(loss.values())).astype(float))\n",
        "            plt.xlabel('Epochs')\n",
        "            if loss_fn == \"mse\":\n",
        "                plt.ylabel(\"Mean Squared Error\")\n",
        "            elif loss_fn == \"ce\":\n",
        "                plt.ylabel(\"Log Loss\")\n",
        "            plt.show()\n",
        "\n",
        "    def predict(self, X):\n",
        "        Y_pred = []\n",
        "        for x in X:\n",
        "            y_pred = self.sigmoid(self.perceptron(x))\n",
        "            Y_pred.append(y_pred)\n",
        "        return np.array(Y_pred)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vJivOD1UrRhP",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "my_cmap = matplotlib.colors.LinearSegmentedColormap.from_list(\"\", [\"red\", \"yellow\", \"green\"])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ahvB7SVRtgu7",
        "colab_type": "text"
      },
      "source": [
        "# Generating Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wIdjpcEdrde6",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "outputId": "6b628107-44ea-4509-f239-43b00102ad5d"
      },
      "source": [
        "data, labels = make_blobs(n_samples=1000, centers=4, n_features=2, random_state=0)\n",
        "print(data.shape, labels.shape)"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(1000, 2) (1000,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pi_mVn0ArwdE",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "outputId": "5aae754c-ada2-49cb-91fc-f98d82b75d50"
      },
      "source": [
        "plt.scatter(data[:,0], data[:,1], c=labels, cmap=my_cmap)\n",
        "plt.show()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9GNqWorKsN1z",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "labels_orig = labels\n",
        "labels = np.mod(labels_orig, 2)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MXvnDpIvsi5P",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "outputId": "9f68433b-ff47-4239-c154-67132607ddfa"
      },
      "source": [
        "plt.scatter(data[:,0], data[:,1], c=labels, cmap=my_cmap)\n",
        "plt.show()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EthJq8BksuZB",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "outputId": "bd947691-8723-4128-a33f-e538e0044ac0"
      },
      "source": [
        "X_train, X_val, Y_train, Y_val = train_test_split(data, labels, stratify=labels, random_state=0)\n",
        "print(X_train.shape, X_val.shape)"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(750, 2) (250, 2)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lxgRKlWCtp--",
        "colab_type": "text"
      },
      "source": [
        "# Sigmoid Neuron Classification"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bavxg1jZtFRg",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362,
          "referenced_widgets": [
            "f96cdf6685db4f4aa62534f54b26bca6",
            "d2fb0b7b8eb940838c4a8e9b3d7b713e",
            "b6e7694af3ec44b1b724814cc8fe1efe",
            "3e1e2b13d35744e9a9b698464ec385a0",
            "9811462d814f48df9183462fd543f85c",
            "b5a04e75ce694e0b956f9ab44b71b9f7",
            "217cc3b1433c4662b690261da6ccaa82",
            "2479b84f50ec4c3eb08324839bc4e044"
          ]
        },
        "outputId": "c9ee0a14-b34e-4f6f-e2db-2d54b8a54140"
      },
      "source": [
        "sn = SigmoidNeuron()\n",
        "sn.fit(X_train, Y_train, epochs=1000, learning_rate=0.5, display_loss=True, loss_fn='mse')"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:47: TqdmDeprecationWarning: This function will be removed in tqdm==5.0.0\n",
            "Please use `tqdm.notebook.tqdm` instead of `tqdm.tqdm_notebook`\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "f96cdf6685db4f4aa62534f54b26bca6",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=1000), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0Sa2jlFBx3g1",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "d9dc4bc1-6f9f-4df0-be4c-801cd05d5092"
      },
      "source": [
        "Y_pred_train = sn.predict(X_train)\n",
        "Y_pred_binarised_train = (Y_pred_train >= 0.5).astype(np.int).ravel()\n",
        "Y_pred_val = sn.predict(X_val)\n",
        "Y_pred_binarised_val = (Y_pred_val >= 0.5).astype(np.int).ravel()\n",
        "accuracy_train = accuracy_score(Y_pred_binarised_train, Y_train)\n",
        "accuracy_val = accuracy_score(Y_pred_binarised_val, Y_val)\n",
        "\n",
        "print(\"Accuracy of train data: \", round(accuracy_train,2))\n",
        "print(\"Accuracy of validation data: \", round(accuracy_val,2))"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy of train data:  0.67\n",
            "Accuracy of validation data:  0.66\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cj3b93FU0437",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "dad2e32d-5464-48b4-9474-18b27e36cc9c"
      },
      "source": [
        "plt.scatter(X_train[:,0], X_train[:,1], c=Y_pred_binarised_train, s=15*(np.abs(Y_pred_binarised_train-Y_train)+0.2), cmap=my_cmap)\n",
        "plt.show()"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cJkNnB4x3PKH",
        "colab_type": "text"
      },
      "source": [
        "# Class for Feed Forward Network"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ep8oTZMa26dy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class FFN:\n",
        "    def __init__(self):\n",
        "        self.w1 = np.random.randn()\n",
        "        self.w2 = np.random.randn()\n",
        "        self.w3 = np.random.randn()\n",
        "        self.w4 = np.random.randn()\n",
        "        self.w5 = np.random.randn()\n",
        "        self.w6 = np.random.randn()\n",
        "        self.b1 = 0\n",
        "        self.b2 = 0\n",
        "        self.b3 = 0\n",
        "\n",
        "    def sigmoid(self,x):\n",
        "        return 1.0/(1.0+np.exp(-x))\n",
        "    \n",
        "    def forward_pass(self,x):\n",
        "        self.x1, self.x2 = x\n",
        "        self.a1 = self.w1*self.x1 + self.w2*self.x2 + self.b1\n",
        "        self.h1 = self.sigmoid(self.a1)\n",
        "        self.a2 = self.w3*self.x1 + self.w4*self.x2 + self.b2\n",
        "        self.h2 = self.sigmoid(self.a2)\n",
        "        self.a3 = self.w5*self.h1 + self.w6*self.h2 + self.b3\n",
        "        self.h3 = self.sigmoid(self.a3)\n",
        "        return self.h3\n",
        "    \n",
        "    def grad(self,x,y):\n",
        "        self.forward_pass(x)\n",
        "\n",
        "        self.dw5 = (self.h3-y)*(self.h3)*(1-self.h3)*self.h1\n",
        "        self.dw6 = (self.h3-y)*(self.h3)*(1-self.h3)*self.h2\n",
        "        self.db3 = (self.h3-y)*(self.h3)*(1-self.h3)\n",
        "\n",
        "        self.dw1 = (self.h3-y)*(self.h3)*(1-self.h3)*self.w5*(self.h1)*(1-self.h1)*self.x1\n",
        "        self.dw2 = (self.h3-y)*(self.h3)*(1-self.h3)*self.w5*(self.h1)*(1-self.h1)*self.x2\n",
        "        self.db1 = (self.h3-y)*(self.h3)*(1-self.h3)*self.w5*(self.h1)*(1-self.h1)\n",
        "\n",
        "        self.dw3 = (self.h3-y)*(self.h3)*(1-self.h3)*self.w6*(self.h2)*(1-self.h2)*self.x1\n",
        "        self.dw4 = (self.h3-y)*(self.h3)*(1-self.h3)*self.w6*(self.h2)*(1-self.h2)*self.x2\n",
        "        self.db2 = (self.h3-y)*(self.h3)*(1-self.h3)*self.w6*(self.h2)*(1-self.h2)\n",
        "\n",
        "    def fit(self, X, Y, epochs=1, learning_rate=1, initialise=True, display_loss=False):\n",
        "        if initialise:\n",
        "            self.w1 = np.random.randn()\n",
        "            self.w2 = np.random.randn()\n",
        "            self.w3 = np.random.randn()\n",
        "            self.w4 = np.random.randn()\n",
        "            self.w5 = np.random.randn()\n",
        "            self.w6 = np.random.randn()\n",
        "            self.b1 = 0\n",
        "            self.b2 = 0\n",
        "            self.b3 = 0\n",
        "        \n",
        "        if display_loss:\n",
        "            loss = {}\n",
        "\n",
        "        for i in tqdm_notebook(range(epochs), total=epochs, unit='epoch'):\n",
        "            dw1, dw2, dw3, dw4, dw5, dw6, db1, db2, db3 = [0]*9\n",
        "            for x,y in zip(X,Y):\n",
        "                self.grad(x,y)\n",
        "                dw1 += self.dw1\n",
        "                dw2 += self.dw2\n",
        "                dw3 += self.dw3\n",
        "                dw4 += self.dw4\n",
        "                dw5 += self.dw5\n",
        "                dw6 += self.dw6\n",
        "                db1 += self.db1\n",
        "                db2 += self.db2\n",
        "                db3 += self.db3\n",
        "\n",
        "            m = X.shape[1]\n",
        "            self.w1 -= learning_rate*dw1/m\n",
        "            self.w2 -= learning_rate*dw2/m\n",
        "            self.w3 -= learning_rate*dw3/m\n",
        "            self.w4 -= learning_rate*dw4/m\n",
        "            self.w5 -= learning_rate*dw5/m\n",
        "            self.w6 -= learning_rate*dw6/m\n",
        "            self.b1 -= learning_rate*db1/m\n",
        "            self.b2 -= learning_rate*db2/m\n",
        "            self.b3 -= learning_rate*db3/m\n",
        "\n",
        "            if display_loss:\n",
        "                Y_pred = self.predict(X)\n",
        "                loss[i] = mean_squared_error(Y_pred, Y)\n",
        "        \n",
        "        if display_loss:\n",
        "            plt.plot(np.array(list(loss.values())).astype(np.float))\n",
        "            plt.xlabel('Epochs')\n",
        "            plt.ylabel('Mean Squared Error')\n",
        "            plt.show()\n",
        "    \n",
        "    def predict(self,X):\n",
        "        Y_pred = []\n",
        "        for x in X:\n",
        "            y_pred = self.forward_pass(x)\n",
        "            Y_pred.append(y_pred)\n",
        "        return np.array(Y_pred)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A4eB22CxDB2R",
        "colab_type": "text"
      },
      "source": [
        "# FFN Classification"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6ItCoSps8HNn",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362,
          "referenced_widgets": [
            "dff6115fc6604d3796017e9c0dad637d",
            "8b2ce44ced6a4621b420404888abd884",
            "51d4fae1383e46868bea67b765edebf5",
            "46728af4498b4bbcbbe8e02c44ce24f3",
            "e920a23e10604a21a9fc3f02fd727f87",
            "2b15fd3be57149a09ec794063f901a43",
            "e009543d551346e6ae06d58630173e05",
            "e80f947e2ce944c08cbe3d7a53a4c7a4"
          ]
        },
        "outputId": "354ff65c-87b1-450a-da5c-30bef629eb91"
      },
      "source": [
        "ffn = FFN()\n",
        "ffn.fit(X_train, Y_train, epochs=2000, learning_rate=0.01, display_loss=True)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:56: TqdmDeprecationWarning: This function will be removed in tqdm==5.0.0\n",
            "Please use `tqdm.notebook.tqdm` instead of `tqdm.tqdm_notebook`\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "dff6115fc6604d3796017e9c0dad637d",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=2000), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XKkhcTfSCYx8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "e43e789b-09fb-407b-ceb3-a31912297f59"
      },
      "source": [
        "Y_pred_train = ffn.predict(X_train)\n",
        "Y_pred_binarised_train = (Y_pred_train >= 0.5).astype(np.int).ravel()\n",
        "Y_pred_val = ffn.predict(X_val)\n",
        "Y_pred_binarised_val = (Y_pred_val >= 0.5).astype(np.int).ravel()\n",
        "accuracy_train = accuracy_score(Y_pred_binarised_train, Y_train)\n",
        "accuracy_val = accuracy_score(Y_pred_binarised_val, Y_val)\n",
        "\n",
        "print(\"Accuracy of train data: \", round(accuracy_train,2))\n",
        "print(\"Accuracy of validation data: \", round(accuracy_val,2))"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy of train data:  0.98\n",
            "Accuracy of validation data:  0.94\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lgVDs4XqDL_c",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "c33d79b2-b774-4bc4-dc4f-d46d08504ba7"
      },
      "source": [
        "plt.scatter(X_train[:,0], X_train[:,1], c=Y_pred_binarised_train, s=15*(np.abs(Y_pred_binarised_train-Y_train)+0.2), cmap=my_cmap)\n",
        "plt.show()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MYYe5c_zD2wS",
        "colab_type": "text"
      },
      "source": [
        "# Class for generic FFN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bLnDzCF0DUAZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class gen_FFN:\n",
        "\n",
        "    def __init__(self, n_inputs, hidden_sizes=[2]):\n",
        "        self.nx = n_inputs\n",
        "        self.ny = 1\n",
        "        self.nh = len(hidden_sizes)\n",
        "        self.sizes = [self.nx] + hidden_sizes + [self.ny]\n",
        "    \n",
        "        self.W = {}\n",
        "        self.B = {}\n",
        "        for i in range(self.nh+1):\n",
        "            self.W[i+1] = np.random.randn(self.sizes[i], self.sizes[i+1])\n",
        "            self.B[i+1] = np.zeros((1, self.sizes[i+1]))\n",
        "  \n",
        "    def sigmoid(self, x):\n",
        "        return 1.0/(1.0 + np.exp(-x))\n",
        "  \n",
        "    def forward_pass(self, x):\n",
        "        self.A = {}\n",
        "        self.H = {}\n",
        "        self.H[0] = x.reshape(1, -1)\n",
        "        for i in range(self.nh+1):\n",
        "            self.A[i+1] = np.matmul(self.H[i], self.W[i+1]) + self.B[i+1]\n",
        "            self.H[i+1] = self.sigmoid(self.A[i+1])\n",
        "        return self.H[self.nh+1]\n",
        "    \n",
        "    def grad_sigmoid(self, x):\n",
        "        return x*(1-x) \n",
        "        \n",
        "    def grad(self, x, y):\n",
        "        self.forward_pass(x)\n",
        "        self.dW = {}\n",
        "        self.dB = {}\n",
        "        self.dH = {}\n",
        "        self.dA = {}\n",
        "        L = self.nh + 1\n",
        "        self.dA[L] = (self.H[L] - y)\n",
        "        for k in range(L, 0, -1):\n",
        "            self.dW[k] = np.matmul(self.H[k-1].T, self.dA[k])\n",
        "            self.dB[k] = self.dA[k]\n",
        "            self.dH[k-1] = np.matmul(self.dA[k], self.W[k].T)\n",
        "            self.dA[k-1] = np.multiply(self.dH[k-1], self.grad_sigmoid(self.H[k-1]))\n",
        "        \n",
        "    def fit(self, X, Y, epochs=1, learning_rate=1, initialise=True, display_loss=False):\n",
        "        \n",
        "        # initialise w, b\n",
        "        if initialise:\n",
        "            for i in range(self.nh+1):\n",
        "                self.W[i+1] = np.random.randn(self.sizes[i], self.sizes[i+1])\n",
        "                self.B[i+1] = np.zeros((1, self.sizes[i+1]))\n",
        "        \n",
        "        if display_loss:\n",
        "            loss = {}\n",
        "        \n",
        "        for e in tqdm_notebook(range(epochs), total=epochs, unit=\"epoch\"):\n",
        "            dW = {}\n",
        "            dB = {}\n",
        "            for i in range(self.nh+1):\n",
        "                dW[i+1] = np.zeros((self.sizes[i], self.sizes[i+1]))\n",
        "                dB[i+1] = np.zeros((1, self.sizes[i+1]))\n",
        "            for x, y in zip(X, Y):\n",
        "                self.grad(x, y)\n",
        "                for i in range(self.nh+1):\n",
        "                    dW[i+1] += self.dW[i+1]\n",
        "                    dB[i+1] += self.dB[i+1]\n",
        "                \n",
        "            m = X.shape[1]\n",
        "            for i in range(self.nh+1):\n",
        "                self.W[i+1] -= learning_rate * dW[i+1] / m\n",
        "                self.B[i+1] -= learning_rate * dB[i+1] / m\n",
        "            \n",
        "            if display_loss:\n",
        "                Y_pred = self.predict(X)\n",
        "                loss[e] = mean_squared_error(Y_pred, Y)\n",
        "        \n",
        "        if display_loss:\n",
        "            plt.plot(np.array(list(loss.values())).astype(np.float))\n",
        "            plt.xlabel('Epochs')\n",
        "            plt.ylabel('Mean Squared Error')\n",
        "            plt.show()\n",
        "        \n",
        "    def predict(self, X):\n",
        "        Y_pred = []\n",
        "        for x in X:\n",
        "            y_pred = self.forward_pass(x)\n",
        "            Y_pred.append(y_pred)\n",
        "        return np.array(Y_pred).squeeze()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fpL5qUvSGs8e",
        "colab_type": "text"
      },
      "source": [
        "# Generic FFN Classification"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "djLzpiKgGJ-3",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362,
          "referenced_widgets": [
            "0abe98dcc1394c5eb3d19116e9c977e3",
            "77de76af9bcd426f849212663e169edb",
            "7a7b1dadd7b24cc789f6eea7fee25ce9",
            "c333945e501649f6b1ba3b697d629d61",
            "7b55ad4afa7a48cf96ede8d79b15e28b",
            "8193861f1d85416290bacde8ecd2fd1b",
            "0434d40777b6463daad8cfdb6e5e7b8e",
            "13abbb44d0fd48a29e390b01441b33a6"
          ]
        },
        "outputId": "8d7f1428-5f28-4472-8a98-bb6a11e431cc"
      },
      "source": [
        "gffn = gen_FFN(2, [2,3])\n",
        "gffn.fit(X_train, Y_train, epochs=1000, learning_rate=0.01, display_loss=True)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:55: TqdmDeprecationWarning: This function will be removed in tqdm==5.0.0\n",
            "Please use `tqdm.notebook.tqdm` instead of `tqdm.tqdm_notebook`\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "0abe98dcc1394c5eb3d19116e9c977e3",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=1000), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XsqzJcNEHLni",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "40f6f650-03b2-470a-e5ab-2385247354f2"
      },
      "source": [
        "Y_pred_train = gffn.predict(X_train)\n",
        "Y_pred_binarised_train = (Y_pred_train >= 0.5).astype(np.int).ravel()\n",
        "Y_pred_val = gffn.predict(X_val)\n",
        "Y_pred_binarised_val = (Y_pred_val >= 0.5).astype(np.int).ravel()\n",
        "accuracy_train = accuracy_score(Y_pred_binarised_train, Y_train)\n",
        "accuracy_val = accuracy_score(Y_pred_binarised_val, Y_val)\n",
        "\n",
        "print(\"Accuracy of train data: \", round(accuracy_train,2))\n",
        "print(\"Accuracy of validation data: \", round(accuracy_val,2))"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy of train data:  0.97\n",
            "Accuracy of validation data:  0.93\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PZJXGFzOHln9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "384e69ba-fd2f-413a-ed9d-07a6436b8019"
      },
      "source": [
        "plt.scatter(X_train[:,0], X_train[:,1], c=Y_pred_binarised_train, s=15*(np.abs(Y_pred_binarised_train-Y_train)+0.2), cmap=my_cmap)\n",
        "plt.show()"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mMqHKHJXIIRQ",
        "colab_type": "text"
      },
      "source": [
        "# Class for Multi-Class FFN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "I00oluVaHsUC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class multi_FFN:\n",
        "  \n",
        "    def __init__(self, n_inputs, n_outputs, hidden_sizes=[3]):\n",
        "        self.nx = n_inputs\n",
        "        self.ny = n_outputs\n",
        "        self.nh = len(hidden_sizes)\n",
        "        self.sizes = [self.nx] + hidden_sizes + [self.ny] \n",
        "\n",
        "        self.W = {}\n",
        "        self.B = {}\n",
        "        for i in range(self.nh+1):\n",
        "            self.W[i+1] = np.random.randn(self.sizes[i], self.sizes[i+1])\n",
        "            self.B[i+1] = np.zeros((1, self.sizes[i+1]))\n",
        "      \n",
        "    def sigmoid(self, x):\n",
        "        return 1.0/(1.0 + np.exp(-x))\n",
        "  \n",
        "    def softmax(self, x):\n",
        "        exps = np.exp(x)\n",
        "        return exps / np.sum(exps)\n",
        "\n",
        "    def forward_pass(self, x):\n",
        "        self.A = {}\n",
        "        self.H = {}\n",
        "        self.H[0] = x.reshape(1, -1)\n",
        "        for i in range(self.nh):\n",
        "            self.A[i+1] = np.matmul(self.H[i], self.W[i+1]) + self.B[i+1]\n",
        "            self.H[i+1] = self.sigmoid(self.A[i+1])\n",
        "        self.A[self.nh+1] = np.matmul(self.H[self.nh], self.W[self.nh+1]) + self.B[self.nh+1]\n",
        "        self.H[self.nh+1] = self.softmax(self.A[self.nh+1])\n",
        "        return self.H[self.nh+1]\n",
        "  \n",
        "    def predict(self, X):\n",
        "        Y_pred = []\n",
        "        for x in X:\n",
        "            y_pred = self.forward_pass(x)\n",
        "            Y_pred.append(y_pred)\n",
        "        return np.array(Y_pred).squeeze()\n",
        " \n",
        "    def grad_sigmoid(self, x):\n",
        "        return x*(1-x) \n",
        "  \n",
        "    def cross_entropy(self,label,pred):\n",
        "        yl=np.multiply(pred,label)\n",
        "        yl=yl[yl!=0]\n",
        "        yl=-np.log(yl)\n",
        "        yl=np.mean(yl)\n",
        "        return yl\n",
        " \n",
        "    def grad(self, x, y):\n",
        "        self.forward_pass(x)\n",
        "        self.dW = {}\n",
        "        self.dB = {}\n",
        "        self.dH = {}\n",
        "        self.dA = {}\n",
        "        L = self.nh + 1\n",
        "        self.dA[L] = (self.H[L] - y)\n",
        "        for k in range(L, 0, -1):\n",
        "            self.dW[k] = np.matmul(self.H[k-1].T, self.dA[k])\n",
        "            self.dB[k] = self.dA[k]\n",
        "            self.dH[k-1] = np.matmul(self.dA[k], self.W[k].T)\n",
        "            self.dA[k-1] = np.multiply(self.dH[k-1], self.grad_sigmoid(self.H[k-1])) \n",
        "    \n",
        "    def fit(self, X, Y, epochs=100, initialize='True', learning_rate=0.01, display_loss=False):\n",
        "      \n",
        "        if display_loss:\n",
        "            loss = {}\n",
        "      \n",
        "        if initialize:\n",
        "            for i in range(self.nh+1):\n",
        "                self.W[i+1] = np.random.randn(self.sizes[i], self.sizes[i+1])\n",
        "                self.B[i+1] = np.zeros((1, self.sizes[i+1]))\n",
        "        \n",
        "        for epoch in tqdm_notebook(range(epochs), total=epochs, unit=\"epoch\"):\n",
        "            dW = {}\n",
        "            dB = {}\n",
        "            for i in range(self.nh+1):\n",
        "                dW[i+1] = np.zeros((self.sizes[i], self.sizes[i+1]))\n",
        "                dB[i+1] = np.zeros((1, self.sizes[i+1]))\n",
        "            for x, y in zip(X, Y):\n",
        "                self.grad(x, y)\n",
        "                for i in range(self.nh+1):\n",
        "                    dW[i+1] += self.dW[i+1]\n",
        "                    dB[i+1] += self.dB[i+1]\n",
        "                  \n",
        "            m = X.shape[1]\n",
        "            for i in range(self.nh+1):\n",
        "                self.W[i+1] -= learning_rate * (dW[i+1]/m)\n",
        "                self.B[i+1] -= learning_rate * (dB[i+1]/m)\n",
        "        \n",
        "            if display_loss:\n",
        "                Y_pred = self.predict(X) \n",
        "                loss[epoch] = self.cross_entropy(Y, Y_pred)\n",
        "    \n",
        "        if display_loss:\n",
        "            plt.plot(np.array(list(loss.values())).astype(np.float))\n",
        "            plt.xlabel('Epochs')\n",
        "            plt.ylabel('CE')\n",
        "            plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BHgs0T0kO6_3",
        "colab_type": "text"
      },
      "source": [
        "# Multi-Class FFN Classification"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LKF1daRaJ1z8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "7b37f692-c151-445c-a640-79c794088e0b"
      },
      "source": [
        "X_train, X_val, Y_train, Y_val = train_test_split(data, labels_orig, stratify=labels_orig, random_state=0)\n",
        "print(X_train.shape, X_val.shape, labels_orig.shape)"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(750, 2) (250, 2) (1000,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JVmeHJYtKXKi",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "f1c93378-65f8-4c5a-b747-afa87aa52734"
      },
      "source": [
        "enc = OneHotEncoder()\n",
        "y_OH_train = enc.fit_transform(np.expand_dims(Y_train,1)).toarray()\n",
        "y_OH_val = enc.fit_transform(np.expand_dims(Y_val,1)).toarray()\n",
        "print(y_OH_train.shape, y_OH_val.shape)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(750, 4) (250, 4)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hZSVRMDZONwd",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362,
          "referenced_widgets": [
            "5fe185cb3042459a8bf9a86cfaacb800",
            "49966cfbd9a24ae381e448a905125267",
            "a8b795defaef4f59a1efa6d1deaca54e",
            "1b690d0a5c2b4cfeae5607ea92ef1af7",
            "b4cf3ec616e9440a932e8996f945945c",
            "4b296623d0e042e9abc8df02e6bdf1ae",
            "64e8ee1d59a54a029bcddf2ccd790fff",
            "ae2b49fa855145b1af64b6efde4da6df"
          ]
        },
        "outputId": "d008ff85-ea79-4b09-cef9-cc732d6e08a6"
      },
      "source": [
        "mffn = multi_FFN(2,4,[2,3])\n",
        "mffn.fit(X_train, y_OH_train, epochs=2000, learning_rate=0.005, display_loss=True)"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:74: TqdmDeprecationWarning: This function will be removed in tqdm==5.0.0\n",
            "Please use `tqdm.notebook.tqdm` instead of `tqdm.tqdm_notebook`\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "5fe185cb3042459a8bf9a86cfaacb800",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=2000), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xytXzxmkOzuw",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "99fd140b-1e20-4c5c-8acf-4232f3468d1d"
      },
      "source": [
        "Y_pred_train = mffn.predict(X_train)\n",
        "Y_pred_train = np.argmax(Y_pred_train,1)\n",
        "Y_pred_val = mffn.predict(X_val)\n",
        "Y_pred_val = np.argmax(Y_pred_val,1)\n",
        "accuracy_train = accuracy_score(Y_pred_train, Y_train)\n",
        "accuracy_val = accuracy_score(Y_pred_val, Y_val)\n",
        "\n",
        "print(\"Accuracy of train data: \", round(accuracy_train,2))\n",
        "print(\"Accuracy of validation data: \", round(accuracy_val,2))"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy of train data:  0.92\n",
            "Accuracy of validation data:  0.91\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_sdAKYAjPIIA",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "1f7f73c9-d93e-42f6-b099-f708e80d9ed2"
      },
      "source": [
        "plt.scatter(X_train[:,0], X_train[:,1], c=Y_pred_train, s=15*(np.abs(Y_pred_train-Y_train)+0.2), cmap=my_cmap)\n",
        "plt.show()"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}