dqn_recsys_Getting_Started.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.3"
    },
    "colab": {
      "name": "Copy of 1. Getting Started.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "23e88ceff0394c999aeab0882b8fa980": {
          "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_54146cce4c064c1eb7047433d4bf92df",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_d74bbbe633c746108b624c58a9eefab3",
              "IPY_MODEL_be7c5292092e44ff86eb88caa49f48d4"
            ]
          }
        },
        "54146cce4c064c1eb7047433d4bf92df": {
          "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
          }
        },
        "d74bbbe633c746108b624c58a9eefab3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_6370b8f271364e6fbc9689243202ec54",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 20000263,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 20000263,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_daead702a64f443bbe42a4072f2d08dd"
          }
        },
        "be7c5292092e44ff86eb88caa49f48d4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_c96744df1d2546708b7e19863e4f02a7",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 20000263/20000263 [00:59&lt;00:00, 335938.99it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_ccad919fe71b4a7d87dce176532e7541"
          }
        },
        "6370b8f271364e6fbc9689243202ec54": {
          "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"
          }
        },
        "daead702a64f443bbe42a4072f2d08dd": {
          "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
          }
        },
        "c96744df1d2546708b7e19863e4f02a7": {
          "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"
          }
        },
        "ccad919fe71b4a7d87dce176532e7541": {
          "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
          }
        },
        "4201df14764746c1b8f64180d3fc8941": {
          "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_e233867a32f34d6db756fee75e8f4d5a",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_3d5c66bf06d04236ad659ee17748ae13",
              "IPY_MODEL_cc4dd6428ad044bdbad1876380d3835e"
            ]
          }
        },
        "e233867a32f34d6db756fee75e8f4d5a": {
          "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
          }
        },
        "3d5c66bf06d04236ad659ee17748ae13": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_df618a451a9d44cbb844d7f2eae137ae",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 20000263,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 20000263,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_16904b99baf24b269b54f42762ee9148"
          }
        },
        "cc4dd6428ad044bdbad1876380d3835e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_a84f1816e8464ec5861c5bac40608737",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 20000263/20000263 [00:27&lt;00:00, 722373.56it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_06f0f90620c842ada986300532bce825"
          }
        },
        "df618a451a9d44cbb844d7f2eae137ae": {
          "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"
          }
        },
        "16904b99baf24b269b54f42762ee9148": {
          "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
          }
        },
        "a84f1816e8464ec5861c5bac40608737": {
          "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"
          }
        },
        "06f0f90620c842ada986300532bce825": {
          "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
          }
        },
        "5d672131c55a4b41ad3fbbc182de90c3": {
          "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_792c34944bf143c0aa03207f546edc96",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_fee95341375648108d471b32a253dbaa",
              "IPY_MODEL_1a81bbfb96404fbe8a4b0f90ad848baa"
            ]
          }
        },
        "792c34944bf143c0aa03207f546edc96": {
          "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
          }
        },
        "fee95341375648108d471b32a253dbaa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_897821c51d7848678fa904f29b6f2173",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 138493,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 138493,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_db11ca5faffe482e882d313075ec295b"
          }
        },
        "1a81bbfb96404fbe8a4b0f90ad848baa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_1b058edd60964a9cb4ac5b8095b56ee7",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 138493/138493 [00:07&lt;00:00, 17828.38it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_226709a29a8a41b4b0bf862bc23a8188"
          }
        },
        "897821c51d7848678fa904f29b6f2173": {
          "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"
          }
        },
        "db11ca5faffe482e882d313075ec295b": {
          "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
          }
        },
        "1b058edd60964a9cb4ac5b8095b56ee7": {
          "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"
          }
        },
        "226709a29a8a41b4b0bf862bc23a8188": {
          "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
          }
        },
        "ebc0aedec1234f64b08fec5eebd55640": {
          "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_e439f00548154b238ac017b3fb4383a4",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_c0e4c5235a6a4f0ba9b4aca9505fcebf",
              "IPY_MODEL_3e5d99f3a5c845469281cc12b590f11c"
            ]
          }
        },
        "e439f00548154b238ac017b3fb4383a4": {
          "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
          }
        },
        "c0e4c5235a6a4f0ba9b4aca9505fcebf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_2ed25f0c743c4d62861cb4ccaacb2c8d",
            "_dom_classes": [],
            "description": " 19%",
            "_model_name": "FloatProgressModel",
            "bar_style": "danger",
            "max": 5263,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 999,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_ae85e9d3a65a4bbdb78c24cfec85f986"
          }
        },
        "3e5d99f3a5c845469281cc12b590f11c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_13b6032fab1f4253944c2667d127b996",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 999/5263 [01:40&lt;04:35, 15.49it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_642babd951c94ca39c8d710554b21da6"
          }
        },
        "2ed25f0c743c4d62861cb4ccaacb2c8d": {
          "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"
          }
        },
        "ae85e9d3a65a4bbdb78c24cfec85f986": {
          "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
          }
        },
        "13b6032fab1f4253944c2667d127b996": {
          "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"
          }
        },
        "642babd951c94ca39c8d710554b21da6": {
          "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
          }
        },
        "1e1f5004b9e645b2b4a461e46c57a5b0": {
          "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_f48624751ba744bf81c6c192158e8870",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_9627518a53a5413289733c5490257830",
              "IPY_MODEL_6c7ae3455db8481cbb74bfd11ad5cf91"
            ]
          }
        },
        "f48624751ba744bf81c6c192158e8870": {
          "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
          }
        },
        "9627518a53a5413289733c5490257830": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_0b340d160fce4f90b56c97d69c11f483",
            "_dom_classes": [],
            "description": " 19%",
            "_model_name": "FloatProgressModel",
            "bar_style": "danger",
            "max": 5263,
            "_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_56acbf1a5e364e6c9ee4756d2cf99586"
          }
        },
        "6c7ae3455db8481cbb74bfd11ad5cf91": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_4aabb7b60442413a86f407645c0ef9df",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 1000/5263 [03:20&lt;28:46,  2.47it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_1a4231ab3d764a8c8ac5f1cfb20401a4"
          }
        },
        "0b340d160fce4f90b56c97d69c11f483": {
          "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"
          }
        },
        "56acbf1a5e364e6c9ee4756d2cf99586": {
          "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
          }
        },
        "4aabb7b60442413a86f407645c0ef9df": {
          "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"
          }
        },
        "1a4231ab3d764a8c8ac5f1cfb20401a4": {
          "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
          }
        },
        "378543ea2df74ec8878ad172b226d3c1": {
          "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_347799a89f1040828aedebfd56e36e2c",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_bb1cd48005ab4cf8baa53e24a6f4013c",
              "IPY_MODEL_23fdbda10c05450594e54b8d07354664"
            ]
          }
        },
        "347799a89f1040828aedebfd56e36e2c": {
          "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
          }
        },
        "bb1cd48005ab4cf8baa53e24a6f4013c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_a78332665eb5445ea059de408af0c004",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 38,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 38,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_b02fc1ba86974f72930c4a40e2aa5824"
          }
        },
        "23fdbda10c05450594e54b8d07354664": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_65ad3824a8e44a8ea665ffbfe3b55212",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 38/38 [00:16&lt;00:00,  2.31it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_19d3734d5ca9494eb1ca2dbeedd6d2d0"
          }
        },
        "a78332665eb5445ea059de408af0c004": {
          "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"
          }
        },
        "b02fc1ba86974f72930c4a40e2aa5824": {
          "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
          }
        },
        "65ad3824a8e44a8ea665ffbfe3b55212": {
          "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"
          }
        },
        "19d3734d5ca9494eb1ca2dbeedd6d2d0": {
          "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
          }
        },
        "5cd60ae929a540508a731e0d6c75672f": {
          "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_a4f92f4751e34672a63320dc8672e8c1",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_a29b05b001fa452b9e20f61daf281a02",
              "IPY_MODEL_1c24077eda83415eb4f8bee745dc4cc2"
            ]
          }
        },
        "a4f92f4751e34672a63320dc8672e8c1": {
          "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
          }
        },
        "a29b05b001fa452b9e20f61daf281a02": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_cc78ff25334c4d2191c3b939397c7179",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 38,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 38,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_5c27ff9029e341e987958cf0fa30ca44"
          }
        },
        "1c24077eda83415eb4f8bee745dc4cc2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_55277456381242adb6cc4adfc2da9346",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 38/38 [00:09&lt;00:00,  3.98it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_4454548f89594b5a8f417f2d420af2b1"
          }
        },
        "cc78ff25334c4d2191c3b939397c7179": {
          "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"
          }
        },
        "5c27ff9029e341e987958cf0fa30ca44": {
          "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
          }
        },
        "55277456381242adb6cc4adfc2da9346": {
          "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"
          }
        },
        "4454548f89594b5a8f417f2d420af2b1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jeonniga/-/blob/master/dqn_recsys_Getting_Started.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-fP2MWpvcMAp"
      },
      "source": [
        "# Pre Setup\n",
        "You ideally should have pytorch pre installed with google colab\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JLzu-ZMPXVh7"
      },
      "source": [
        "from IPython.display import clear_output\n",
        "\n",
        "! git clone https://github.com/awarebayes/RecNN\n",
        "! pip install -r ./RecNN/requirements.txt\n",
        "! pip install ./RecNN\n",
        "! pip install gdown\n",
        "\n",
        "clear_output()"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2RxhPDDeXYuV"
      },
      "source": [
        "! wget http://files.grouplens.org/datasets/movielens/ml-20m.zip\n",
        "! gdown https://drive.google.com/uc?id=1EQ_zXBR3DKpmJR3jBgLvt-xoOvArGMsL\n",
        "! unzip ml-20m.zip\n",
        "clear_output()"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FQ5yY1-0kjyY"
      },
      "source": [
        "## Pandas backends\n",
        "\n",
        "You can get better loading speed performance if you use Modin.\n",
        "It does not come with recnn by default so you have to install it. \n",
        "\n",
        "It will not affect the performance in a google colab notebook, but if your pc has plenty of cores, it will."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TLgEitQSkjVn"
      },
      "source": [
        "# !pip install modin[ray]\n",
        "# recnn.pd.set(\"modin\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3ZA9MbXhn2Iu"
      },
      "source": [
        "## Restart runtime\n",
        "Dont panic if it says \"Your session crashed\". Run and jump over to \"Getting started section\""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YSEGL7lhlWFF"
      },
      "source": [
        "import os\n",
        "os.kill(os.getpid(), 9)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_m-oFm2_XTpU"
      },
      "source": [
        "# Getting Started with recnn\n",
        "This is my school project. It focuses on Reinforcement Learning for personalized news recommendation. The main distinction is that it tries to solve online off-policy learning with dynamically generated item embeddings. I want to create a library with SOTA algorithms for reinforcement learning recommendation, providing the level of abstraction you like. In this notebook, you can play around with library's key components.\n",
        "\n",
        "P.S. this tutorial is available offline\n",
        "\n",
        "\n",
        "Also some bugs may occure: tqdm.pandas import error\n",
        "\n",
        "Go to runtime, click restart"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gKf6hfmxXTpW"
      },
      "source": [
        "from IPython.display import clear_output\n",
        "import recnn\n",
        "\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "from tqdm.auto import tqdm\n",
        "import torch_optimizer as optim"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y1lDsRU6XTpa"
      },
      "source": [
        "# Environments\n",
        "\n",
        "Main abstraction of the library for datasets is called environment, similar to how other reinforcement\n",
        "learning libraries name it. This interface is created to provide SARSA like input for your RL Models. When you are working with recommendation env, you have two choices: using static length inputs (say 10 items) or dynamic length time series with sequential encoders (many to one rnn). Static length is provided via FrameEnv, and dynamic length along with sequential state representation encoder is implemented in SeqEnv. Let’s take a look at FrameEnv first:\n",
        "\n",
        "In order to initialize an env, you need to provide embeddings and ratings directories.\n",
        "You can also use caching."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jtpcUIldXTpb",
        "outputId": "9c480ab9-d9ae-4b7d-8764-2ecf94694b1c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 166,
          "referenced_widgets": [
            "23e88ceff0394c999aeab0882b8fa980",
            "54146cce4c064c1eb7047433d4bf92df",
            "d74bbbe633c746108b624c58a9eefab3",
            "be7c5292092e44ff86eb88caa49f48d4",
            "6370b8f271364e6fbc9689243202ec54",
            "daead702a64f443bbe42a4072f2d08dd",
            "c96744df1d2546708b7e19863e4f02a7",
            "ccad919fe71b4a7d87dce176532e7541",
            "4201df14764746c1b8f64180d3fc8941",
            "e233867a32f34d6db756fee75e8f4d5a",
            "3d5c66bf06d04236ad659ee17748ae13",
            "cc4dd6428ad044bdbad1876380d3835e",
            "df618a451a9d44cbb844d7f2eae137ae",
            "16904b99baf24b269b54f42762ee9148",
            "a84f1816e8464ec5861c5bac40608737",
            "06f0f90620c842ada986300532bce825",
            "5d672131c55a4b41ad3fbbc182de90c3",
            "792c34944bf143c0aa03207f546edc96",
            "fee95341375648108d471b32a253dbaa",
            "1a81bbfb96404fbe8a4b0f90ad848baa",
            "897821c51d7848678fa904f29b6f2173",
            "db11ca5faffe482e882d313075ec295b",
            "1b058edd60964a9cb4ac5b8095b56ee7",
            "226709a29a8a41b4b0bf862bc23a8188"
          ]
        }
      },
      "source": [
        "frame_size = 10\n",
        "batch_size = 25\n",
        "# embeddgings: https://drive.google.com/open?id=1EQ_zXBR3DKpmJR3jBgLvt-xoOvArGMsL\n",
        "dirs = recnn.data.env.DataPath(\n",
        "    base=\"\",\n",
        "    embeddings=\"ml20_pca128.pkl\",\n",
        "    ratings=\"ml-20m/ratings.csv\",\n",
        "    cache=\"cached_frame_env.pkl\", # cache will generate after you run\n",
        "    use_cache=True\n",
        ")\n",
        "env = recnn.data.env.FrameEnv(dirs, frame_size, batch_size)"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "23e88ceff0394c999aeab0882b8fa980",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, max=20000263.0), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "4201df14764746c1b8f64180d3fc8941",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, max=20000263.0), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "5d672131c55a4b41ad3fbbc182de90c3",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, max=138493.0), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aafirYfDXTpg"
      },
      "source": [
        "You can get train and test batches directly from the environmnent"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JMBU6ix8XTph"
      },
      "source": [
        "train = env.train_batch()\n",
        "test = env.train_batch()\n",
        "state, action, reward, next_state, done = recnn.data.get_base_batch(train, device=torch.device('cpu'))"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1eQlppreXTpj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5f911d93-937f-478e-c708-48bbdf29da95"
      },
      "source": [
        "print(state)"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tensor([[  6.0919,  -1.2394,   1.1477,  ...,   5.0000,   5.0000,   5.0000],\n",
            "        [  5.8175,  -1.2590,  -9.0728,  ...,   5.0000,   5.0000,   5.0000],\n",
            "        [  3.1487,   3.3241, -12.5570,  ...,   5.0000,   5.0000,   5.0000],\n",
            "        ...,\n",
            "        [  5.0938,  -1.5238,  -5.7540,  ...,  -2.0000,   4.0000,   1.0000],\n",
            "        [  6.3019,  -6.2144,  -1.8957,  ...,   4.0000,   1.0000,   2.0000],\n",
            "        [  3.6832,  -1.0374,  -7.8407,  ...,   1.0000,   2.0000,   2.0000]])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0UvlWSc1XTpn"
      },
      "source": [
        "You can also look at the metadata:\n",
        "    user ids and their positions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VjgWGWVtXTpn",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "422c57ff-ae10-4edd-857f-9bcdee79e2f1"
      },
      "source": [
        "train['meta']"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'sizes': tensor([ 236,  199,  214,  532,   70, 1474,   69,   37,   39,   30,   80,   70,\n",
              "           29,   44,   21,   74,   22,  130,   23,   29,   20,  203,  255,  237,\n",
              "          282]),\n",
              " 'users': tensor([ 45451,  33570,  80914,  76741,  53800,  49229, 125875,  99666,  27891,\n",
              "          21545,  77765,  91899,   7776, 111255,  41133,  39113,    197, 133708,\n",
              "          26048, 118842, 103729, 120089,  36593,  67495, 106622])}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IqRb0aeeXTpq"
      },
      "source": [
        "## Working with metadata\n",
        "you can display user_id: their item/rating positions in the batch"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MAy7DAc2XTpq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "db4c33f1-6ba2-4f7e-998a-c109440fd7cf"
      },
      "source": [
        "import numpy as np\n",
        "user_ends = np.cumsum(train['meta']['sizes']).numpy().tolist() + [0]\n",
        "user_starts = [0] + user_ends\n",
        "starts_ends = list(zip(user_starts, user_ends))[:-1]\n",
        "user_range = dict(zip(train['meta']['users'].numpy(), starts_ends))\n",
        "user_range"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{197: (3218, 3240),\n",
              " 7776: (3050, 3079),\n",
              " 21545: (2870, 2900),\n",
              " 26048: (3370, 3393),\n",
              " 27891: (2831, 2870),\n",
              " 33570: (236, 435),\n",
              " 36593: (3645, 3900),\n",
              " 39113: (3144, 3218),\n",
              " 41133: (3123, 3144),\n",
              " 45451: (0, 236),\n",
              " 49229: (1251, 2725),\n",
              " 53800: (1181, 1251),\n",
              " 67495: (3900, 4137),\n",
              " 76741: (649, 1181),\n",
              " 77765: (2900, 2980),\n",
              " 80914: (435, 649),\n",
              " 91899: (2980, 3050),\n",
              " 99666: (2794, 2831),\n",
              " 103729: (3422, 3442),\n",
              " 106622: (4137, 4419),\n",
              " 111255: (3079, 3123),\n",
              " 118842: (3393, 3422),\n",
              " 120089: (3442, 3645),\n",
              " 125875: (2725, 2794),\n",
              " 133708: (3240, 3370)}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "45ctbrwoXTpt"
      },
      "source": [
        "# Recommending"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zIBQiqKTXTpv"
      },
      "source": [
        "Let's initialize main networks, and recommend something!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ym0yOEbBXTpw"
      },
      "source": [
        "value_net  = recnn.nn.Critic(1290, 128, 256, 54e-2)\n",
        "policy_net = recnn.nn.Actor(1290, 128, 256, 6e-1)"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0yXfHvvxXTpz",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1b807965-4d19-4b6f-e99f-a8825b0629bb"
      },
      "source": [
        "recommendation = policy_net(state)\n",
        "value = value_net(state, recommendation)\n",
        "print(recommendation)\n",
        "print(value)"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tensor([[ -0.0429,   4.0228,   4.6988,  ...,   4.4322,  -0.6010,  -5.3859],\n",
            "        [ -2.9168,   3.1831,  -9.1667,  ...,   3.1363,   6.8549,  -7.3664],\n",
            "        [ -2.9340,  -4.4668,  -1.2368,  ...,   0.2389,  -9.8812,  -6.1002],\n",
            "        ...,\n",
            "        [ -1.6067,  -1.5906,  -3.9627,  ...,  -0.1376,  -3.7005, -10.7235],\n",
            "        [ -5.1449,   1.4699,   0.9481,  ...,  -0.5809,  -2.2753,  -0.8915],\n",
            "        [ -1.5316,   2.4869,  -0.8978,  ...,   0.5183,   6.2981,  -3.2225]],\n",
            "       grad_fn=<AddmmBackward>)\n",
            "tensor([[-1.4894],\n",
            "        [-2.7197],\n",
            "        [-4.3218],\n",
            "        ...,\n",
            "        [ 2.6723],\n",
            "        [ 0.2003],\n",
            "        [ 0.7195]], grad_fn=<AddmmBackward>)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ei9hlgwVXTp3"
      },
      "source": [
        "# Algo classes\n",
        "\n",
        "Algo is a high level abstraction for an RL algorithm. You need two networks (policy and value) in order to initialize it. Later on you can tweak parameters and stuff in the algo itself\n",
        ".\n",
        "#### Important: you can set writer to torch.SummaryWriter and get the debug output\n",
        "\n",
        "\n",
        "## Tweak how you want\n",
        "```\n",
        "ddpg = recnn.nn.DDPG(policy_net, value_net)\n",
        "print(ddpg.params)\n",
        "ddpg.params['gamma'] = 0.9\n",
        "ddpg.params['policy_step'] = 3\n",
        "ddpg.optimizers['policy_optimizer'] = torch.optim.Adam(ddpg.nets['policy_net'], your_lr)\n",
        "ddpg.writer = torch.utils.tensorboard.SummaryWriter('./runs')\n",
        "ddpg = ddpg.to(torch.device('cuda'))\n",
        "```\n",
        "\n",
        "ddpg.loss_layout is also handy, it allows you to see how the loss should look like"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6zD_EYCWXTp4"
      },
      "source": [
        "# test function\n",
        "def run_tests():\n",
        "    batch = next(iter(env.test_dataloader))\n",
        "    loss = ddpg.update(batch, learn=False)\n",
        "    return loss"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Uwh2z5zSXTp7"
      },
      "source": [
        "value_net  = recnn.nn.Critic(1290, 128, 256, 54e-2)\n",
        "policy_net = recnn.nn.Actor(1290, 128, 256, 6e-1)"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Of6jbPfWXTp-"
      },
      "source": [
        "cuda = torch.device('cuda')\n",
        "ddpg = recnn.nn.DDPG(policy_net, value_net)\n",
        "ddpg = ddpg.to(cuda)\n",
        "plotter = recnn.utils.Plotter(ddpg.loss_layout, [['value', 'policy']],)"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aW0TTlwrXTqB",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 306,
          "referenced_widgets": [
            "ebc0aedec1234f64b08fec5eebd55640",
            "e439f00548154b238ac017b3fb4383a4",
            "c0e4c5235a6a4f0ba9b4aca9505fcebf",
            "3e5d99f3a5c845469281cc12b590f11c",
            "2ed25f0c743c4d62861cb4ccaacb2c8d",
            "ae85e9d3a65a4bbdb78c24cfec85f986",
            "13b6032fab1f4253944c2667d127b996",
            "642babd951c94ca39c8d710554b21da6"
          ]
        },
        "outputId": "e6d75933-00b5-4e4f-f4ea-0e10b4ccdd75"
      },
      "source": [
        "from IPython.display import clear_output\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "plot_every = 50\n",
        "n_epochs = 2\n",
        "\n",
        "def learn():\n",
        "    for epoch in range(n_epochs):\n",
        "        for batch in tqdm(env.train_dataloader):\n",
        "            loss = ddpg.update(batch, learn=True)\n",
        "            plotter.log_losses(loss)\n",
        "            ddpg.step()\n",
        "            if ddpg._step % plot_every == 0:\n",
        "                clear_output(True)\n",
        "                print('step', ddpg._step)\n",
        "                test_loss = run_tests()\n",
        "                plotter.log_losses(test_loss, test=True)\n",
        "                plotter.plot_loss()\n",
        "            if ddpg._step > 1000:\n",
        "                return\n",
        "            \n",
        "learn()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "step 1000\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YlEWXTnsdDsI"
      },
      "source": [
        "if ImportError: cannot import name '_png':\n",
        "\n",
        "Headbar -> Runtime -> Restart Runtime\n",
        "\n",
        "Or: Contol+M\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J0E2gbTjXTqF"
      },
      "source": [
        "# Update Functions\n",
        "## Here is what's going on inside algo class"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6O9hkgUaXTqG"
      },
      "source": [
        "value_net  = recnn.nn.Critic(1290, 128, 256, 54e-2)\n",
        "policy_net = recnn.nn.Actor(1290, 128, 256, 6e-1)\n",
        "# these are target networks that we need for ddpg algorigm to work\n",
        "target_value_net = recnn.nn.Critic(1290, 128, 256)\n",
        "target_policy_net = recnn.nn.Actor(1290, 128, 256)\n",
        "\n",
        "target_policy_net.eval()\n",
        "target_value_net.eval()\n",
        "\n",
        "\n",
        "# soft update\n",
        "recnn.utils.soft_update(value_net, target_value_net, soft_tau=1.0)\n",
        "recnn.utils.soft_update(policy_net, target_policy_net, soft_tau=1.0)\n",
        "\n",
        "# define optimizers\n",
        "value_optimizer = optim.RAdam(value_net.parameters(),\n",
        "                              lr=1e-5, weight_decay=1e-2)\n",
        "policy_optimizer = optim.RAdam(policy_net.parameters(), lr=1e-5 , weight_decay=1e-2)\n",
        "\n",
        "nets = {\n",
        "    'value_net': value_net.to(cuda),\n",
        "    'target_value_net': target_value_net.to(cuda),\n",
        "    'policy_net': policy_net.to(cuda),\n",
        "    'target_policy_net': target_policy_net.to(cuda),\n",
        "}\n",
        "\n",
        "optimizer = {\n",
        "    'policy_optimizer': policy_optimizer,\n",
        "    'value_optimizer':  value_optimizer\n",
        "}\n",
        "\n",
        "debug = {}\n",
        "writer = recnn.utils.misc.DummyWriter()\n"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "C_uJw8cQXTqI",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7e908476-8202-4dc7-9fdd-f781939d5082"
      },
      "source": [
        "step = 0\n",
        "params = {\n",
        "    'gamma'      : 0.99,\n",
        "    'min_value'  : -10,\n",
        "    'max_value'  : 10,\n",
        "    'policy_step': 10,\n",
        "    'soft_tau'   : 0.001,\n",
        "}\n",
        "batch = {'state': state, 'action': action, 'reward': reward, 'next_state': next_state, 'done':done}\n",
        "loss = recnn.nn.update.ddpg_update(batch, params, nets, optimizer, cuda, debug, writer, step=step)\n",
        "print(loss)"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'value': 13.708120346069336, 'policy': -1.1476987600326538, 'step': 0}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6WL3CmNdXTqL"
      },
      "source": [
        "## Training Loop\n",
        "Let's wrap this thing in the training loop. I also will use Colab's gpu for training.\n",
        "\n",
        "In an error pops up, go to runtime -> change runtime type -> select GPU. And also restart the session."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JS2MWwCwXTqM"
      },
      "source": [
        "cuda = torch.device('cuda')\n",
        "loss = {\n",
        "    'test': {'value': [], 'policy': [], 'step': []},\n",
        "    'train': {'value': [], 'policy': [], 'step': []}\n",
        "    }\n",
        "\n",
        "plotter = recnn.utils.Plotter(loss, [['value', 'policy']],)"
      ],
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U1qNjAfXXTqQ"
      },
      "source": [
        "# test function\n",
        "def run_tests():\n",
        "    batch = next(iter(env.test_dataloader))\n",
        "    loss = recnn.nn.ddpg_update(batch, params, nets, optimizer,\n",
        "                       cuda, debug, writer, step=step, learn=False)\n",
        "    return loss"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "N4F_vYzgXTqT",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303,
          "referenced_widgets": [
            "1e1f5004b9e645b2b4a461e46c57a5b0",
            "f48624751ba744bf81c6c192158e8870",
            "9627518a53a5413289733c5490257830",
            "6c7ae3455db8481cbb74bfd11ad5cf91",
            "0b340d160fce4f90b56c97d69c11f483",
            "56acbf1a5e364e6c9ee4756d2cf99586",
            "4aabb7b60442413a86f407645c0ef9df",
            "1a4231ab3d764a8c8ac5f1cfb20401a4"
          ]
        },
        "outputId": "723d3a9a-7712-4f8b-ca40-e7f50890afe8"
      },
      "source": [
        "from IPython.display import clear_output\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "plot_every = 50\n",
        "n_epochs = 2\n",
        "\n",
        "def learn():\n",
        "    step = 0\n",
        "    for epoch in range(n_epochs):\n",
        "        for batch in tqdm(env.train_dataloader):\n",
        "            loss = recnn.nn.ddpg_update(batch, params,\n",
        "                                               nets, optimizer, cuda, debug,\n",
        "                                               writer, step=step)\n",
        "            plotter.log_losses(loss)\n",
        "            step += 1\n",
        "            if step % plot_every == 0:\n",
        "                clear_output(True)\n",
        "                print('step', step)\n",
        "                test_loss = run_tests()\n",
        "                plotter.log_losses(test_loss, test=True)\n",
        "                plotter.plot_loss()\n",
        "            if step > 1000:\n",
        "                return\n",
        "            \n",
        "        \n",
        "learn()"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "step 1000\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k_ebfkkLXTqa"
      },
      "source": [
        "def plot_grad_flow(named_parameters):\n",
        "    ave_grads = []\n",
        "    layers = []\n",
        "    for n, p in named_parameters:\n",
        "        if(p.requires_grad) and (\"bias\" not in n):\n",
        "            print('plotting', n)\n",
        "            layers.append(n)\n",
        "            ave_grads.append(p.grad.abs().mean())\n",
        "    plt.plot(ave_grads, alpha=0.3, color=\"b\")\n",
        "    plt.hlines(0, 0, len(ave_grads)+1, linewidth=1, color=\"k\" )\n",
        "    plt.xticks(range(0,len(ave_grads), 1), layers, rotation=\"vertical\")\n",
        "    plt.xlim(xmin=0, xmax=len(ave_grads))\n",
        "    plt.xlabel(\"Layers\")\n",
        "    plt.ylabel(\"average gradient\")\n",
        "    plt.title(\"Gradient flow\")\n",
        "    plt.grid(True)"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YTHj8oN2XTqh",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 410
        },
        "outputId": "e26d3c08-4329-401f-f0b5-7f83596a80ec"
      },
      "source": [
        "plot_grad_flow(ddpg.nets['policy_net'].named_parameters())"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "plotting linear1.weight\n",
            "plotting linear2.weight\n",
            "plotting linear3.weight\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAFUCAYAAAAu8yfrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deZzVd33v8debYQt7CANhTQgkEwgQshBJ4kJy9bZVY3ur3tZWe22tqbW1tmptr62tWtt71V57tWo1atVbt7q0VdNotepk34AQSCCTBZIA2YCwDYQZls/94/sbMyHDzAHO7/zO73fez8fjPGbO/oHvbz7nd77L56uIwMzMqmdY0QGYmVk+nODNzCrKCd7MrKKc4M3MKsoJ3sysopzgzcwqygneWoqkhyW9NPv9PZI+16D3laQvSNop6Q5JKyRtacR7W+tygremIelXJd0uaZ+kp7Lf3ypJebxfRPxNRPz2yb6OpDMlhaThgzzshcDLgFkRccnJvqdZLZzgrSlIeifwMeAjwOnANOAtwOXAyGM8p61hAZ68M4CHI2Jf0YFY63CCt8JJmgh8AHhrRHwrIvZGcldE/HpE9GSP+6Kkf5B0naR9wBWSXiHpLkl7JG2W9L6jXvsNkh6RtEPSnx113/skfbnf9eWSbpG0S9Ldklb0u69T0l9JulnSXkk/lDQlu/uG7OcuSd2SLj3qfd4EfA64NLv//QP8HyzI3mOXpHslvSq7fW5227Ds+mclPdXvef8k6Q+P6z/cWoYTvDWDS4FRwHdqeOyvAX8NjAduAvYBvwFMAl4B/K6kXwKQtBD4B+ANwAzgNGDWQC8qaSbw78AHgcnAu4BvS2o/6r1/E5hK+lbxruz2F2c/J0XEuIi4tf9rR8TnSd9Gbs3u/8uj3nsE8D3gh9lrvw34iqSOiNgE7AEu6Pde3ZIWZNdfAlw/2H+Yta6mS/CS/jHrf72nTq83Jzvb2iBpvaQz6/G6VldTgO0Rcajvhn5n0s9IenG/x34nIm6OiCMRcSAiOiNiXXZ9LfA1UtIDeA1wbUTckH0LeC9w5BgxvB64LiKuy17rR8BK4OX9HvOFiLg/Ip4BvgEsrcu/HpYD44D/HRG9EfET4Frgddn91wMvkXR6dv1b2fW5wATg7jrFYRXTdAke+CLw83V8vf8HfCQiFgCXAE8N8XhrvB3AlP6DlBFxWURMyu7rf5xu7v9ESS+Q9FNJ2yTtJp0p93WdzOj/+Kz/e8cxYjgDeG32obJL0i7SwOj0fo95ot/v+0lJuR5mAJsjov+HzyPAzOz364EVpLP3G4BO0ofYS4Abj3qe2c80XYKPiBuAp/vfJmmepB9IWiXpRknn1vJa2Vf04dnZGBHRHRH76x+1naRbgR7gF2t47NHlT78KfBeYHRETgU8DfbNuHgdm9z1Q0hhSN81ANgP/FBGT+l3GRsT/PoGYjtdjwOy+fvbMHGBr9vv1wItISf56UtfU5bh7xobQdAn+GK4B3hYRF5H6PT9V4/POIQ18/Us2EPeRks28aAkRsQt4P/ApSa+RNF7SMElLgbFDPH088HREHJB0CamfvM+3gFdKeqGkkaSB3GMd818GrpL0c5LaJI3O5qoP2Gd/lG2krp+zanjsQG4nfSN4t6QR2eDuVcDXASLiAeAZUjfS9RGxB3gSeDVO8DaIpk/wksYBlwHflLQG+AzZ12ZJvyzpngEu/5E9fTjpzOddwDLSH+AbG/6PsCFFxIeBdwDvJiWvJ0lt/SfALYM89a3AByTtBf6C1Dfe95r3Ar9HOst/HNgJDLi4KCI2k75BvIeUsDcDf0wNfyPZt8K/Bm7OuneWD/Wco57fS0rovwBsJ53A/EZE3NfvYdcDO7I4+64LWH0872WtRc244Uc2EHptRCySNAHoiojpgz9rwNdZDnwoIl6SXX8DsDwifq+e8ZqZNaOmP4PPvo5ukvRa+NmS7/NrfPqdwKR+U92uBNbnEKaZWdNpugQv6WukQbcOSVuyRSK/DrxJ0t3AvdQ2GEdEHCZ1z/xY0jrSV9rP5hO5mVlzacouGjMzO3lNdwZvZmb1MVj1u4abNGlSzJ8/v+gw7ATs27ePsWOHmtFozcrtV16rVq3aHhHtA93XVAl+2rRprFy5sugw7AR0dnayYsWKosOwE+T2Ky9JjxzrPnfRmJlVlBO8mVlF5dpFI+lhYC9wGDgUERfn+X5mZvasRvTBXxER2xvwPmZm1o+7aMzMKirvM/gAfigpgM9ExDVHP0DS1cDVAO3t7XR2duYckuWhu7vbbVdibr9qynUlq6SZEbFV0lTgR6SSvzcc6/EdHR3R1dWVWzyWH0+zKze3X3lJWnWs8c1cu2giYmv28yngX0k7KlnF9PRAd3dTLakwM3JM8JLGShrf9zvwX4G67LNqzePAAbjpJli3biI7jrUZnpkVIs8z+GnATVkFyDuAf4+IH+T4ftZgBw/CbbdBby+MHHmYVatSwjez5pDb9+qI2AjUWrfdSubwYbj9dti3D17wAojYy6FDsGoVXHopDPP8LLPC+c/QjtuRI7ByJezaBRdeCFOmwJgxhzn/fHj6adiwoegIzQyc4O04RcCaNfDUU7BkCUzvt5HizJkwdy5s3AiPP15cjGaWOMHbcbnnHti6FRYsgDlznn//woVw6qnpQ6C7u/HxmdmznOCtZl1d8PDDMG8eHKts/7BhcNFF6efKlXDoUENDNLN+nOCtJps2wf33p7P2hQsHf+wpp6Qk390Na9c2Jj4zez4neBvS1q2pa+b001O/ey2mTIGOjvTcTZvyjc/MBuYEb4N68km4666UsC+6CKTanzt/PkybBuvXw86d+cVoZgNzgrdjevrp1I8+YQIsW3b8c9sluOACGD06vU5PTz5xmtnAnOBtQHv2wB13pP705cth+AkuiRsxIn049PbC6tVpmqWZNYYTvD3Pvn2pBEFbW1qVOnLkyb3ehAmp73779jQTx8wawwnenuPAgZTcI1JyP+WU+rzu7NlpBs4DD8ATT9TnNc1scE7w9jMHD6b6Mj09qb7MuHH1ff3Fi2HixLQIat+++r62mT2fE7wBqXjYHXekuevLlsGkSfV/j2HD4OJsW4KVK9N7mll+nODtZ8XDdu5MxcPa2/N7rzFj0nvs2QPr1uX3PmbmBN/yBiselpepU+Gcc2DzZnj00fzfz6xVOcG3uHvvHbx4WF7OOSd9U1i3LpUdNrP6c4JvYfffn8oIDFY8LC9S6qoZNSp1Dx082Nj3N2sFTvAt6uGH05z02bOHLh6Wl5Ej06BrT48XQZnlwQm+BW3dmrpGTj8dzi94U8VJk2DRojQG8MADxcZiVjVO8C3mqadS8bDTTktdJMdTPCwvZ5wBs2albxTbthUdjVl1OMG3kKOLh7W1FR3Rs5YsSXGtWgXPPFN0NGbV4ATfIvqKh40enVapjhhRdETP1daW+uMj0ofQkSNFR2RWfk7wLWD//lSCoK0tVYYcNaroiAY2dmwqL7xrV9pgxMxOjhN8xfX0pOJhhw+n5D5mTNERDe7009OUzUcegS1bio7GrNyc4Cvs4MGU3A8cSN0y48cXHVFtzj03DQLffXfqWjKzE+MEX1FHFw879dSiI6qdlLYHHDnSi6DMToYTfAUdOZJmozz9dOrTzrN4WF5GjUpJfv/+VCvHi6DMjp8TfMVEpK6NJ59MUw9nzCg6ohM3eXJaZfvEE/DQQ0VHY1Y+TvAVs359Gpw899y0gKjszjorfUjdd1/a8s/MaucEXyEPPAAbN6akePbZRUdTP+efn6ZQrl6dBozNrDZO8BXxyCPpLHfWrOKKh+Vl+PA0UHzoUBpb8CIos9rknuAltUm6S9K1eb9Xq3rsMVi7FqZNS2e7zVBfpt7GjYOlS9PA8YYNRUdjVg6NOIN/O+A/yZxs25aKh02enGadDKvwd7IZM1L308aN6UPNzAaXazqQNAt4BfC5PN+nVe3cCXfemc5uL7mkuYqH5WXBgvRhtmYN7N1bdDRmzW14zq//f4F3A8dcQynpauBqgPb2djo7O3MOqRr272/jnnsmMnz4ERYt2s3NNxc7Uby7u7thbdfbO4x7751IV1ewZMlu2to8Sf5kNbL9rHEUOa0gkfRK4OUR8VZJK4B3RcQrB3tOR0dHdHV15RJPlezfDzffnH6//PLmqC/T2dnJihUrGvZ+27enMgzTp6euKTs5jW4/qx9JqyLi4oHuy7OL5nLgVZIeBr4OXCnpyzm+X0soW/GwvEyZkub6P/ZY6pM3s+fLLcFHxP+MiFkRcSbwq8BPIuL1eb1fKzh4MJX9LVvxsLzMn5+qT65fn2bXmNlzVXjORbUcPpwGVPfsSRtjlKl4WJ6WLk3fYlatSt9uzOxZDUnwEdE5VP+7HVtEWsW5Y0cqHjZ1atERNY8RI9IH3sGDKcm7KJnZs3wG3+T6ioc98QQsXgwzZxYdUfOZMCEVVtuxI63mNbPECb7JrV8PmzdDRweceWbR0TSvWbNScbUHH0wfhmbmBN/U+oqHzZ0L55xTdDTNb9EimDQprezdt6/oaMyK5wTfpPqKh82cCeedV3Q05TBsWOqPHzYs7QR1+HDREZkVywm+CT3++LPFw5YurWbxsLyccgpceGGabbR2bdHRmBXLCb7JbNuWZsy0QvGwvLS3pzGLLVvSNyGzVuX00UR27Wq94mF5OfvsNJ30nnvS/6tZK3KCbxLd3akEwahRaZXqiBFFR1RuUuqqGTUq9cf39hYdkVnjOcE3gWeegVtvTd0xl14Ko0cXHVE1jBiRdoLq6UndXl4EZa3GCb5gvb0pubd68bC8TJyYFoht2wb33190NGaN5QRfoEOHUrfMgQOpz33ChKIjqqY5c2D27JTgn3yy6GjMGscJviBHjsAddzxbPGzy5KIjqrbFi9MH6F13pXr6Zq3ACb4AEakwlouHNU5bW/oghTToeuRIsfGYNYITfAH6ioctWuTiYY00dmz6QN29G9atKzoas/wNmeAljarlNqtNX/Gwc85JNWassaZNS3PkH300XcyqrJYz+FtrvM2G8OCD8NBDqSpkR0fR0bSujo605d+6dels3qyqjpngJZ0u6SLgFEkXSLowu6wAPJnvOD36KGzYkLpkFi0qOprWJqUyECNHpv74gweLjsgsH8MHue/ngDcCs4CP9rt9L/CeHGOqnL7iYVOnunhYsxg5Mg263nJLmlmzbJnbxarnmAk+Ir4EfEnSqyPi2w2MqVK2b0+rKCdNeraUrTWHU0+FhQtTvZoHH0x982ZVMtgZfJ9rJf0acGb/x0fEB/IKqir6ioeNHZvqy7h4WPOZOxd27oSurvQh3N5edERm9VPL+eR3gF8EDgH7+l1sEN3dcPvtqStg+XIXD2tm55+fKniuXp3qAplVRS1n8LMi4udzj6RC+oqHSSm5u3hYc+tbBHXjjWkB2mWXuSvNqqGWw/gWSYtzj6QientTfZlDh1JyHzu26IisFuPGpQHwnTvh3nuLjsasPmo5g38h8EZJm4AeQEBExJJcIyuhQ4dSt8z+/Sm5u3hYuUyfDvPmpbUKkyd7lbGVXy0J/hdyj6ICjhxJA6q7d6cpd6edVnREdiLOPTedxd99d/qAHj++6IjMTtyQXTQR8QgwG7gy+31/Lc9rJRFpgG779vQ1f9q0oiOyEzVsWFoENXx4+sA+dKjoiMxOXC21aP4S+BPgf2Y3jQC+nGdQZbN2bVrMdN55MGtW0dHYyRo9OiX5/fthzZqiozE7cbWcif834FVkUyMj4jHAX1wzGzakMgRnnw1nnVV0NFYvp50GCxakD+6HHio6GrMTU0uC742IAAJAkueFZB56KK2APOOM1Hdr1TJvXhp43bAh1e43K5taEvw3JH0GmCTpzcB/Ap/NN6zmt3lzKv07Y0baLciqaenStE/uqlVpa0WzMqllkPVvgW8B3wY6gL+IiL/PO7Bm9sQTaZZFe3vaQMJFqqpr+PA0K+rQoZTkvROUlUkt0ySJiB8BPzqeF5Y0GrgBGJW9z7ci4i+PO8Ims317+kN38bDWMX58KmewejXcd18qUGZWBsdM8JJuiogXStpL1v/edxdpodNQy3h6SFMruyWNAG6S9P2IuO3kwy7G7t3PFg+75JJ0dmetYeZMePrpNO5y6qmpb96s2Q1WLviF2c8TmjGTDcx2Z1dHZJc49jOaW3d3KkEwYkRapTpyZNERWaOdd176kF+zJp3VjxtXdERmgxvsDH7yYE+MiKeHenFJbcAqYD7wyYi4fYDHXA1cDdDe3k5nZ+dQL9twvb3DWLduIocPw+LFu7ntNnfEHq27u7sp267eenrSsdDVFSxevJu2ttKeszxHq7Rfq1E60R7gjlR7JkhdMnOAndnvk4BHI6LmLaMlTQL+FXhbRNxzrMd1dHREV1dX7dE3QG8v3HxzmkFx2WUwcWLRETWnzs5OVqxYUXQYDbFtW6o5NGMGXHhh0dHURyu1X9VIWhURFw903zGHCCNibkScRZoWeVVETImI04BXAj88ngAiYhfwU6BUZYf7Fw+75BInd0va29PG3Vu3wqZNRUdjdmy1zAFZHhHX9V2JiO8Dlw31JEnt2Zk7kk4BXgbcd6KBNtqRI2lD5t2707J1Fw+z/ubPTzWH1q9PxcnMmlEtCf4xSX8u6czs8mfAYzU8bzrwU0lrgTuBH0XEtScTbKNEpI2Yt21L0+NOP73oiKzZSGkNxOjR6USgt7foiMyer5YE/zqgndSH/q/A1Oy2QUXE2oi4ICKWRMSiMu3hum4dPPZYmjUxe3bR0VizGjEiLYLq7U1rI44xnGVWmCFncmezZd7egFiawn33wSOPuHiY1WbCBFiyJE2d7OpyTSJrLkMmeEntwLuB84Cf7S4aEVfmGFchNm6EBx6AOXP8h2q1mz07LYJ64IG0CMr7AVizqKWL5iukwdG5wPuBh0l96pWyeXPai3P69HRGZnY8Fi9Os6zuugv27Ss6GrOklgR/WkR8HjgYEddHxG8BlTp77yseNmVKmtfs4mF2vIYNS7WJIA26Hj5cbDxmUFuCP5j9fFzSKyRdAAy6yrVMduxIA2QTJ6YBMxcPsxM1Zkw6QdizJw3UmxWtlnJZH5Q0EXgn8PfABOCPco2qQXbvhjvuSH+YL3iBi4fZyZs6Fc45B+6/HyZPTuM5ZkUZNKVltWTOzuav7wauaEhUDbBvX1qlOmIEXHqpi4dZ/ZxzTlr8tG5dmmUzaVLREVmrGrRDIiIOU8Oc97I5cABuvTXNW16+PC1WMasXKXXVjBqV+uMPHhz6OWZ5qKXH+WZJn5D0IkkX9l1yjywnBw+msr8HD6bk7pKvloeRI9Oga09P2ijEi6CsCLX0Oi/NfvZfiRqUcCZNX/GwfftSn7uLh1meJk2CRYtg7do0R/6cc4qOyFpNLStZK9Hv3lc8bNeudGY1ZUrREVkrOOOMtAiqqystgmpvLzoiayW1rGR9xwA37wZWRcSa+odUfy4eZkVasiRNnVy1Cl7yEjjllKIjslZRSx/8xcBbgJnZ5XdIdd0/K+ndOcZWN33FwxYu9LQ1a7y2tvStMSJ9izziDcGsQWpJ8LOACyPinRHxTuAiUkXJFwNvzDG2uujqSsXD5s+HefOKjsZa1dixqbzwrl1wzzH3NDOrr1oS/FSgp9/1g8C0iHjmqNubzsaNacHJnDmwYEHR0VirO/30dKLxyCOwZUvR0VgrqGUWzVeA2yV9J7t+FfBVSWOB9blFdpK2bHHxMGs+556bFkGtXZsWQU2YUHREVmVDnsFHxF8BVwO7sstbIuIDEbEvIn497wBPxJNPpvrcLh5mzUZKW0COGOFFUJa/mqqvRMRKYGXOsdTF00+nPxwXD7NmNWpUSvK33JJORJYtKzoiq6pKpb89e9JCJhcPs2Y3eXKa1fXEE/Dgg0VHY1VVmQS/b18qQTB8eCpB4OJh1uzOOgtmzEjbRG7fXnQ0VkU1JXhJZ0h6afb7KZLG5xvW8TlwICX3iFQZ0gtJrCzOPz9NoVy9Oh3HZvU0ZIKX9GbgW8BnsptmAf+WZ1DHo694WE9P6pZx8TArk+HDUx/8oUNppasXQVk91XIG/3vA5cAegIh4gDQ3vnCHD6cNO/btS38krrttZTRuHCxdmiYIbNhQdDRWJbUk+J6I6O27Imk4qZpkofqKh+3cmaZCuoiTldmMGalPfuPGVFbDrB5qSfDXS3oPcIqklwHfBL6Xb1iDi0jTy556Ki1imj69yGjM6mPBgjS7Zs0a6O4uOhqrgloS/J8C24B1pEJj1wF/nmdQQ7n3Xti6Nf1BuHiYVcWwYWl+/PDhcOedqV/e7GTUspL1SER8NiJeGxGvyX4vrIumqws2bUqFw+bPLyoKs3yMHp26HPftg7vvLjoaK7ta6sGv4/l97rtJK1s/GBE78ghsIJs2peJhs2enRSJmVTRlSqpZs2FD2iTkrLOKjsjKqpa1nt8HDgNfza7/KjAGeAL4Iqn4WO62bk1lVk8/Pc0dNquy+fPTBIL169PssMmTi47IyqiWBP/SiOi/yfY6Sasj4kJJr88rsP6eeirtyHTaaamP0sXDrBUsXQo33pjmx7/4xamGjdnxqGWQtU3SJX1XJC0D2rKruQ8D9RUPmzABLrnExcOsdYwYkXaCOngwJfniRr6srGpJl78NfF7SJkkPA58H3pzVg/9fx3qSpNmSfippvaR7Jb39eIPbsyctZBo92sXDrDVNmJCmAu/YkWrWmB2PIVNmRNwJLJY0Mbu+u9/d3xjkqYeAd0bE6qx2zSpJP4qImjYJ2b8/lSBoa0v1Zfz11FrVrFnpm+yDD6ZBV28ab7Wq6ZxY0iuA84DRyjrAI+IDgz0nIh4HHs9+3ytpA2nT7iETfE8P3HprWq16+eUuHma2aBHs3p3Gol784lSgzGwotRQb+zTwK8DbAAGvBc44njeRdCZwAXD7UI89unjY+KaqW2lWjGHDUn+8lMakDh8uOiIrAw21ZknS2ohY0u/nOOD7EfGimt4gPf564K8j4l8GuP9q0paAtLdPvei97/0+3d3DWbBgL5MmeT+zsuju7macS3nmbteuEaxfP4H29h7OPrt+9QzcfuV1xRVXrIqIiwe6r5Yumr4q1fslzQB2ADVVf5E0Avg28JWBkjtARFwDXANwxhkL48wzL+Sii1LxJSuPzs5OVqxYUXQYLWHhwrSie+5cOOO4vksfm9uvmmqZRfM9SZOAjwCrgYd5dtHTMSl11n8e2BARH60lmEOHxJIlTu5mgzn7bJg6NS3827Wr6GismQ2a4CUNA34cEbsi4tukvvdzI+Ivanjty4E3AFdKWpNdXj7YE0aNOlK3MxKzqpLgggvSzLKVK6G3d+jnWGsaNMFHxBHgk/2u9xw1TXKw594UEYqIJRGxNLtcN9hzRozwdjZmtRg5Mm1y09OTtvvzIigbSC1dND+W9GrJBQLMmsnEibB4MWzblorwmR2tlgT/O6RNPnol7ZG0V9KenOMysxrMmZOqq95/f6rZZNZfLfXgx0fEsIgYERETsusTGhGcmQ1t8eJU0mD16rQC3KxPLQudJOn1kt6bXZ/dv/iYmRWrrS0tgoI06HrEQ1mWqaWL5lPApcCvZde76TfwambFGzs2zazZvRvWrSs6GmsWtST4F0TE75EteIqIncDIXKMys+M2bVqaI//oo+liVkuCPyipjWzbPkntgL8EmjWhjo605d+6dels3lpbLQn+48C/AlMl/TVwE/A3uUZlZidESpt2jxyZ+uMPupxTS6tlFs1XgHeTNvd4HPiliPhm3oGZ2YkZNSoNuh44kMoLexFU66plFs3HgckR8cmI+EREbGhAXGZ2Ek49NRUle/LJtFGItaZaumhWAX8u6SFJfytpwLKUZtZc5s6FmTNT5clt24qOxopQSxfNlyLi5cAyoAv4kKQHco/MzE7akiUwblxaBPXMM0VHY41Wyxl8n/nAuaSKkt7+16wEhg9P/fFHjsCqVV4E1Wpq6YP/cHbG/gHgHuDiiLgq98jMrC7GjYOlS2HnTlhf05b3VhW17Oj0EHBpRGzPOxgzy8f06XDWWbBxYxqAnTmz6IisEYZM8BHxGUmnZvVnRve7/YZcIzOzulqwIO0AdffdqTiZN7Svvlq6aH4buAH4D+D92c/35RuWmdXbsGFw0UWpX/7OO+HQoaIjsrzVMsj6dtIMmkci4grgAsA7QZqV0OjRKcnv3w9r1hQdjeWtlgR/ICIOAEgaFRH3AR35hmVmeTnttNRd8/jj8NBDRUdjeaplkHWLpEnAvwE/krQTeCTfsMwsT/PmpVk1GzbApElFR2N5qWWQ9b9lv75P0k+BicAPco3KzHK3dCnccEOaH398S2KsLI6rVSPi+oj4bkT05hWQmTXG8OGwbFkabN2y5ZSiw7Ec+GPbrIWNHw+XXgpnnrmv6FAsB07wZi3u1FPTFEqrHjermVlFOcGbmVWUE7yZWUU5wZuZVZQTvJlZRTnBm5lVlBO8mVlFOcGbmVVUbgle0j9KekrSPXm9h5mZHVueZ/BfBH4+x9c3M7NB5Jbgsy39ns7r9c3MbHC11IPPlaSrgasB2tvb6ezsLDYgOyHd3d1uuxJz+1WTIiK/F5fOBK6NiEW1PL6joyO6urpyi8fy09nZyYoVK4oOw06Q26+8JK2KiIsHus+zaMzMKsoJ3sysovKcJvk14FagQ9IWSW/K673MzOz5chtkjYjX5fXaZmY2NHfRmJlVlBO8mVlFOcGbmVWUE7yZWUU5wZuZVZQTvJlZRTnBm5lVlBO8mVlFOcGbmVWUE7yZWUU5wZuZVZQTvJlZRTnBm5lVlBO8mVlFOcGbmVWUE7yZWUU5wZuZVZQTvJlZRTnBm5lVlBO8mVlFOcGbmVWUE7yZWUU5wZuZVZQTvJlZRTnBm5lVlBO8mVlFOcGbmVWUE7yZWUU5wZuZVZQTvJlZRTnBm5lVVK4JXtLPS+qS9KCkP83zvczM7LlyS/CS2oBPAr8ALAReJ2lhXu9nZmbPlecZ/CXAgxGxMSJ6ga8Dv5jj+5mZWT/Dc3ztmcDmfte3AC84+kGSrgauBhgzZgyScgzJzKx15JngaxIR1wDXAHR0dERXV1fBEdmJ6LGx4B0AAAuxSURBVOzsZMWKFUWHYSfI7Vdeg50U59lFsxWY3e/6rOw2MzNrgDwT/J3A2ZLmShoJ/Crw3Rzfz8zM+smtiyYiDkn6feA/gDbgHyPi3rzez8zMnivXPviIuA64Ls/3MDOzgXklq5lZRTnBm5lVlBO8mVlFOcGbmVWUIqLoGH5G0l7AK53KaQqwvegg7IS5/crrjIhoH+iOwleyHqUrIi4uOgg7fpJWuu3Ky+1XTe6iMTOrKCd4M7OKarYEf03RAdgJc9uVm9uvgppqkNXMzOqn2c7gzcysTpzgzcwqygnezKyinODNWpSkt9dym5VX4Qle0odquc2aj9uu9P7HALe9sdFBWH4Kn0UjaXVEXHjUbWsjYklRMVlt3HblJOl1wK8BLwRu7HfXeOBIRPyXQgKzuiusVIGk3wXeCpwlaW2/u8YDNxcTldXCbVd6twCPk+rP/J9+t+8F1g74DCulws7gJU0ETgX+F/Cn/e7aGxFPFxKU1cRtZ1YOhXfRAEhqA6bR7xtFRDxaXERWK7ddeUn6ZeBDwFRA2SUiYkKhgVndFJ7gs4253wc8CRzJbg734zY/t125SXoQuCoiNhQdi+WjGRL8g8ALImJHoYHYcXPblZukmyPi8qLjsPw0Qz34zcDuooOwE+K2K6GsawZgpaR/Bv4N6Om7PyL+pZDArO6KHGR9R/breUAH8O889yD7aBFx2dDcduUm6QuD3B0R8VsNC8ZyVeQZ/Pjs56PZZWR2sebntiuxiPjNomOwxii8D97MiiHp4wPcvBtYGRHfaXQ8Vn+F98FL+h5w9KfMbmAl8JmIOND4qKwWbrvSGw2cC3wzu/5qYBNwvqQrIuIPC4vM6qLwM3hJHwPaga9lN/0KsIeUOCZExBuKis0G57YrN0m3AZdHxOHs+nBS6YIXAusiYmGR8dnJK/wMHrgsIpb1u/49SXdGxDJJ9xYWldXCbVdupwLjeHYm1FhgckQcltRz7KdZWTRDgh8naU7f6kdJc0gHHUBvcWFZDdx25fZhYI2kTtIq1hcDfyNpLPCfRQZm9dEMCf6dwE2SHiIdZHOBt2YH2ZcKjcyG4rYrsYj4vKTrgEuym94TEY9lv/9xQWFZHRXeBw8gaRRpsAegy4Nz5eG2Kx9J50bEfZIuHOj+iFjd6JgsH0UudLoyIn7Sb1Xdc3g1XfNy25WbpGsi4mpJPx3g7oiIKxselOWiyC6alwA/Aa4a4L4AnCSal9uuxCLi6uznFUXHYvlqii4aM2s8SWOAdwBzsjP6s4GOiLi24NCsTpphT9Zpkj4v6fvZ9YWS3lR0XDY0t13pfYE02+my7PpW4IPFhWP1VniCB74I/AcwI7t+P+AVdOXwRdx2ZTYvIj4MHASIiP2k2VBWEc2Q4KdExDfINoyIiEPA4WJDshq57cqtV9IpZOUmJM2jX1VQK79mmAe/T9JpPHuQLcc1xsvCbVdu7wN+AMyW9BXgcuCNRQZk9VX4IKuki4CPA4uAe0i1TV4TEd7dvcm57cov+4BeTuqauS0ithccktVR4QkeflbkqIN0kHVFxMGCQ7Iaue3KS9KXgeuBGyPivqLjsforPMFLuonsIANujoi9hQZkNXPblZukK4AXZZd5wF3ADRHxsUIDs7pphgQ/l2cPsuWkQZ4bI+KPCg3MhuS2Kz9JbcAy4ArgLcAzEXHu4M+ysih8kDUiNkk6QJqP20s60BYUG5XVwm1XbpJ+TCoRfCvpW9iyiHiq2KisnprhDP4hYDvwVdJBtiYijhQalNXEbVdukv4OuIj0zetm4Abg1oh4ptDArG6aIcG/nbSDzGzgPlKf7g0R8VChgdmQ3HbVIGk8aXrku4DTI2JUsRFZvRSe4PtIGgf8JukgmxURbQWHZDVy25WTpN8njZ9cBDxM+hZ2Y0T8pMi4rH4K74OX9H9IZ4HjgFuAvyAdaNbk3HalNxr4KLAqW4VsFVP4Gbyk15DOGp4sNBA7bm47s+ZWeII3s+Yh6dqIeGXRcVh9NEOxseeR5C3DSsptV3pvLjoAqx+fwZuZVVRTnsGbWX4knSvp+5L+XdI8SV+UtEvSHZK8UK1CmjLBS1pXdAx2bJJmS/q6pBslvUfSiH73/VuRsVlNrgE+BXyZtLfuD4BTgb8CPlFgXFZnhXXRSPrlY90FfDoi2hsZj9VO0o+AbwO3AW8izaO+KiJ2SLorIi4oNEAbVP82kvRgRMzvd9/qiLiwuOisnoqcB//PwFfINos4yugGx2LHpz0iPp39/jZJrwdukPQqBm5Pay79F6J99Kj7RjYyEMtXkQl+LfC3EXHP0XdIemkB8VjtRkgaHREHACLiy5KeIO3POrbY0KwGn5Q0LiK6I+JTfTdKmg/8Z4FxWZ0V2UXzIuCRiHh0gPsujoiVBYRlNZD0R8DqiLj+qNsvAD4cES8rJjKrVVYm+A8i4u+KjsXyU+g0SR9k5eW2Kz9Jd0TEJUXHYfkpfB68D7LyctuVW1YueARpPGxf3+0R4cVqFdEMCd4HWUm57cpN0k8HuDki4sqGB2O5aIYE74OspNx2Zs2t8ARvZsWR9ArgPPpNTY6IDxQXkdVT4fXgwQdZmbntykvSp4ExpL10Pwe8Brij0KCsrgovVZAdZL8CvI20ivW1wBmFBmU1cduV3mUR8RvAzoh4P3ApcE7BMVkdFZ7g8UFWZm67cuvbXHu/pBnAQWB6gfFYnTVDgvdBVl5uu3K7VtIk4CPAatK+rF8rNCKrq2bogz/6IAtSf6A1P7ddiUXEX2W/flvStcDoiNhdZExWX001i0bSKHyQlZLbrnwkjQHeCcyJiDdLOhvoiIhrCw7N6qTwLhpJYyS9V9JnI6IHmCrJe0KWgNuu9L4A9JDGTgC2Ah8sLhyrt8ITPD7IysxtV27zIuLDpLETImI/aTaUVUQzJHgfZOXltiu3XkmnkNXwlzSP9IFtFdEMg6w+yMrLbVduf0narm+2pK8AlwNvLDQiq6vCB1klvQz4c2Ah8EOygywiOouMy4bmtis/SacBy0nfvG6LiO0Fh2R1VHiCBx9kZea2KzdJM0mrj3/2bT4ibiguIqunZuiigVTHZCcpnoWSfJCVh9uupCR9iFRq4l7gSHZzAG6/iig8wfsgKy+3Xen9Emneu8dNKqrwBI8PsjJz25XbRtKGLW6/imqGBO+DrLzcduW2H1gj6cf0a8OI+IPiQrJ6aoYE74OsvNx25fbd7GIVVfgsGkn/Y6DbI+JLjY7Fjo/bzqy5FZ7gzayxJH0jIv67pHVki9T6i4glBYRlOSgswfsgKy+3XblJmh4Rj0sacPetiHik0TFZPopM8D7ISsptZ1YO7qIxazGS9jLANy/SauSIiAkNDslyUuQZvA+yknLbmZWDz+DNzCqqGerBm5lZDpzgzcwqygneWoKk7qJjMGs0J3izOpLUDOU/zAAneGthkq6SdLukuyT9p6RpkoZJekBSe/aYYZIelNSeXb4t6c7scnn2mPdJ+idJNwP/JOk8SXdIWiNpraSzC/2HWstygrdWdhOwPCIuAL4OvDsijgBfBn49e8xLgbsjYhvwMeDvImIZ8Grgc/1eayHw0oh4HfAW4GMRsRS4GNjSkH+N2VH8ddJa2SzgnyVNB0YCm7Lb/xH4DvB/gd8CvpDd/lKyXasyEySNy37/bkQ8k/1+K/BnkmYB/xIRD+T7zzAbmM/grZX9PfCJiFgM/A5p+0EiYjPwpKQrgUuA72ePH0Y641+aXWZGRN/g7b6+F42IrwKvAp4Brstex6zhnOCtlU0Etma/H136+HOkrppvRsTh7LYfAm/re4CkpQO9qKSzgI0R8XHSNwEXX7NCOMFbqxgjaUu/yzuA9wHflLQK2H7U478LjOPZ7hmAPwAuzgZO15P62gfy34F7JK0BFgH/r57/ELNauVSB2QAkXUwaUH1R0bGYnSgPspodRdKfAr/LszNpzErJZ/BmZhXlPngzs4pygjczqygneDOzinKCNzOrKCd4M7OK+v88TRJA2yn0DAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4H_0JBCbeStU"
      },
      "source": [
        "# Bonus: working with your own data\n",
        "RecNN's main goal is to provide a wrapper for Markov Descision Process.\n",
        "\n",
        "It literally does one thing: iterates over table with columns [item, user, rating, timestamp]\n",
        "\n",
        "Anyways, here is how my preparedataset function looks like:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v6cMKTZZXTq1"
      },
      "source": [
        "def prepare_dataset(args_mut, kwargs):\n",
        "\n",
        "    \"\"\"\n",
        "        Basic prepare dataset function. Automatically makes index linear, in ml20 movie indices look like:\n",
        "        [1, 34, 123, 2000], recnn makes it look like [0,1,2,3] for you.\n",
        "    \"\"\"\n",
        "\n",
        "    # get args\n",
        "    frame_size = kwargs.get('frame_size')\n",
        "    key_to_id = args_mut.base.key_to_id\n",
        "    df = args_mut.df\n",
        "\n",
        "    # rating range mapped from [0, 5] to [-5, 5]\n",
        "    df['rating'] = try_progress_apply(df['rating'], lambda i: 2 * (i - 2.5))\n",
        "    # id's tend to be inconsistent and sparse so they are remapped here\n",
        "    df['movieId'] = try_progress_apply(df['movieId'], lambda i: key_to_id.get(i))\n",
        "    users = df[['userId', 'movieId']].groupby(['userId']).size()\n",
        "    users = users[users > frame_size].sort_values(ascending=False).index\n",
        "\n",
        "    if pd.get_type() == \"modin\":\n",
        "        df = df._to_pandas() # pandas groupby is sync and doesnt affect performance\n",
        "    ratings = df.sort_values(by='timestamp').set_index('userId').drop('timestamp', axis=1).groupby('userId')\n",
        "\n",
        "    # groupby user\n",
        "    user_dict = {}\n",
        "\n",
        "    def app(x):\n",
        "        userid = (x.index[0])\n",
        "        user_dict[userid] = {}\n",
        "        user_dict[userid]['items'] = x['movieId'].values\n",
        "        user_dict[userid]['ratings'] = x['rating'].values\n",
        "\n",
        "    # collect\n",
        "    try_progress_apply(ratings, app)\n",
        "\n",
        "    # set args\n",
        "    args_mut.user_dict = user_dict\n",
        "    args_mut.users = users\n",
        "\n",
        "    return args_mut, kwargs"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "D49wWJMw0ocf"
      },
      "source": [
        " Look up the docs: reference/data/dataset_functions for further details."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sQO3CAmEXTq3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 203
        },
        "outputId": "f27753fe-90a3-4cc1-b015-0f0251135d81"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import datetime\n",
        "import random\n",
        "import time\n",
        "\n",
        "def random_string_date():\n",
        "  return datetime.datetime.strptime('{} {} {} {}'.format(random.randint(1, 366),\n",
        "                                                         random.randint(0, 23),\n",
        "                                                         random.randint(1, 59),\n",
        "                                                          2019), '%j %H %M %Y').strftime(\"%m/%d/%Y, %H:%M:%S\")\n",
        "\n",
        "def string_time_to_unix(s):\n",
        "  return int(time.mktime(datetime.datetime.strptime(s, \"%m/%d/%Y, %H:%M:%S\").timetuple()))\n",
        "\n",
        "size = 100000\n",
        "n_emb = 1000\n",
        "n_usr = 1000\n",
        "mydf = pd.DataFrame({'book_id': np.random.randint(0, n_emb, size=size),\n",
        "                     'reader_id': np.random.randint(1, n_usr, size=size),\n",
        "                     'liked': np.random.randint(0, 2, size=size),\n",
        "                     'when': [random_string_date() for i in range(size)]})\n",
        "my_embeddings = dict([(i, torch.tensor(np.random.randn(128)).float()) for i in range(n_emb)])\n",
        "mydf.head()"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>book_id</th>\n",
              "      <th>reader_id</th>\n",
              "      <th>liked</th>\n",
              "      <th>when</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>820</td>\n",
              "      <td>512</td>\n",
              "      <td>1</td>\n",
              "      <td>03/02/2019, 21:02:00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>62</td>\n",
              "      <td>835</td>\n",
              "      <td>0</td>\n",
              "      <td>11/24/2019, 13:48:00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>485</td>\n",
              "      <td>653</td>\n",
              "      <td>1</td>\n",
              "      <td>01/01/2020, 04:40:00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>566</td>\n",
              "      <td>938</td>\n",
              "      <td>1</td>\n",
              "      <td>07/20/2019, 05:39:00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>51</td>\n",
              "      <td>743</td>\n",
              "      <td>0</td>\n",
              "      <td>12/25/2019, 20:43:00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   book_id  reader_id  liked                  when\n",
              "0      820        512      1  03/02/2019, 21:02:00\n",
              "1       62        835      0  11/24/2019, 13:48:00\n",
              "2      485        653      1  01/01/2020, 04:40:00\n",
              "3      566        938      1  07/20/2019, 05:39:00\n",
              "4       51        743      0  12/25/2019, 20:43:00"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NdOFZBGok2q_"
      },
      "source": [
        "! mkdir mydataset\n",
        "import pickle\n",
        "\n",
        "mydf.to_csv('mydataset/mydf.csv', index=False)\n",
        "with open('mydataset/myembeddings.pickle', 'wb') as handle:\n",
        "    pickle.dump(my_embeddings, handle)"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z6j-4zKBf7bq"
      },
      "source": [
        "   def prepare_my_dataset(args_mut, kwargs):\n",
        "\n",
        "        # get args\n",
        "        frame_size = kwargs.get('frame_size')\n",
        "        key_to_id = args_mut.base.key_to_id\n",
        "        df = args_mut.df\n",
        "\n",
        "        df['liked'] = df['liked'].apply(lambda a: (a - 1) * (1 - a) + a)\n",
        "        df['when'] = df['when'].apply(string_time_to_unix)\n",
        "        df['book_id'] = df['book_id'].apply(key_to_id.get)\n",
        "\n",
        "        users = df[['reader_id', 'book_id']].groupby(['reader_id']).size()\n",
        "        users = users[users > frame_size].sort_values(ascending=False).index\n",
        "\n",
        "        # If using modin: pandas groupby is sync and doesnt affect performance\n",
        "        # if pd.get_type() == \"modin\": df = df._to_pandas()  \n",
        "        ratings = df.sort_values(by='when').set_index('reader_id').drop('when', axis=1).groupby('reader_id')\n",
        "\n",
        "        # Groupby user\n",
        "        user_dict = {}\n",
        "\n",
        "        def app(x):\n",
        "            userid = x.index[0]\n",
        "            user_dict[int(userid)] = {}\n",
        "            user_dict[int(userid)]['items'] = x['book_id'].values\n",
        "            user_dict[int(userid)]['ratings'] = x['liked'].values\n",
        "\n",
        "        ratings.apply(app)\n",
        "\n",
        "        args_mut.user_dict = user_dict\n",
        "        args_mut.users = users\n",
        "\n",
        "        return args_mut, kwargs"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Nzr5vRFUf_EV"
      },
      "source": [
        "frame_size = 10\n",
        "batch_size = 25\n",
        "\n",
        "\n",
        "dirs = recnn.data.env.DataPath(\n",
        "    base=\"mydataset/\",\n",
        "    embeddings=\"myembeddings.pickle\",\n",
        "    ratings=\"mydf.csv\",\n",
        "    cache=\"frame_env.pkl\", # cache will generate after you run\n",
        "    use_cache=True # generally you want to save env after it runs\n",
        ")\n",
        "\n",
        "env = recnn.data.env.FrameEnv(\n",
        "    dirs,\n",
        "    frame_size,\n",
        "    batch_size, \n",
        "    prepare_dataset=prepare_my_dataset # <- ! pass YOUR function here\n",
        ") "
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0Vqkc3m1qJwj"
      },
      "source": [
        "# That's it!\n",
        "\n",
        "Now let's use the library"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "82wWJ-2sjd5j",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 321,
          "referenced_widgets": [
            "378543ea2df74ec8878ad172b226d3c1",
            "347799a89f1040828aedebfd56e36e2c",
            "bb1cd48005ab4cf8baa53e24a6f4013c",
            "23fdbda10c05450594e54b8d07354664",
            "a78332665eb5445ea059de408af0c004",
            "b02fc1ba86974f72930c4a40e2aa5824",
            "65ad3824a8e44a8ea665ffbfe3b55212",
            "19d3734d5ca9494eb1ca2dbeedd6d2d0",
            "5cd60ae929a540508a731e0d6c75672f",
            "a4f92f4751e34672a63320dc8672e8c1",
            "a29b05b001fa452b9e20f61daf281a02",
            "1c24077eda83415eb4f8bee745dc4cc2",
            "cc78ff25334c4d2191c3b939397c7179",
            "5c27ff9029e341e987958cf0fa30ca44",
            "55277456381242adb6cc4adfc2da9346",
            "4454548f89594b5a8f417f2d420af2b1"
          ]
        },
        "outputId": "325426f5-3209-4667-93ba-c26438b49f9c"
      },
      "source": [
        "# test function\n",
        "def run_tests():\n",
        "    batch = next(iter(env.test_dataloader))\n",
        "    loss = ddpg.update(batch, learn=False)\n",
        "    return loss\n",
        "  \n",
        "value_net  = recnn.nn.Critic(1290, 128, 256, 54e-2)\n",
        "policy_net = recnn.nn.Actor(1290, 128, 256, 6e-1)\n",
        "\n",
        "cuda = torch.device('cuda')\n",
        "ddpg = recnn.nn.DDPG(policy_net, value_net)\n",
        "ddpg = ddpg.to(cuda)\n",
        "plotter = recnn.utils.Plotter(ddpg.loss_layout, [['value', 'policy']],)\n",
        "\n",
        "from IPython.display import clear_output\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "plot_every = 3\n",
        "n_epochs = 2\n",
        "\n",
        "def learn():\n",
        "    for epoch in range(n_epochs):\n",
        "        for batch in tqdm(env.train_dataloader):\n",
        "            loss = ddpg.update(batch, learn=True)\n",
        "            plotter.log_losses(loss)\n",
        "            ddpg.step()\n",
        "            if ddpg._step % plot_every == 0:\n",
        "                clear_output(True)\n",
        "                print('step', ddpg._step)\n",
        "                test_loss = run_tests()\n",
        "                plotter.log_losses(test_loss, test=True)\n",
        "                plotter.plot_loss()\n",
        "            if ddpg._step > 100:\n",
        "                return\n",
        "            \n",
        "learn()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "step 75\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "exrxI8d4rAYA"
      },
      "source": [
        "# Congratulations you've made it\n",
        "Now you know how to use the library with your own data and is ready to apply it in your project. \n",
        "\n",
        "**Make sure to give it a star**"
      ]
    }
  ]
}