diff --git a/01_Regression.ipynb b/01_Regression.ipynb index c96fabd..a163fcb 100644 --- a/01_Regression.ipynb +++ b/01_Regression.ipynb @@ -1,4972 +1,151 @@ { - "nbformat": 4, - "nbformat_minor": 0, - "metadata": { - "colab": { - "provenance": [], - "include_colab_link": true + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "view-in-github", + "colab_type": "text" + }, + "source": [ + "\"Open" + ] }, - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" + { + "cell_type": "markdown", + "metadata": { + "id": "7nQSRI4Dim1Y" + }, + "source": [ + "---\n", + "# **PyCaret for Regression**\n", + "---\n", + "- It is a bundle of many Machine Learning algorithms.\n", + "- Only three lines of code is required to compare 20 ML models.\n", + "- Pycaret is available for:\n", + " - Classification\n", + " - Regression\n", + " - Clustering\n", + "---\n", + "\n", + "### **Self learning resource**\n", + "1. Tutorial on Pycaret ** Click Here**\n", + "\n", + "2. Documentation on Pycaret-Regression: ** Click Here **\n", + "\n", + "---\n", + "\n", + "### **In this tutorial we will learn:**\n", + "\n", + "- Getting Data\n", + "- Setting up Environment\n", + "- Create Model\n", + "- Tune Model\n", + "- Plot Model\n", + "- Finalize Model\n", + "- Predict Model\n", + "- Save / Load Model\n", + "---\n", + "\n" + ] }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 + { + "cell_type": "markdown", + "metadata": { + "id": "A30y-VtNim1h" }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.8.5" + "source": [ + "### **(a) Install Pycaret**" + ] }, - "widgets": { - "application/vnd.jupyter.widget-state+json": { - "f887b0172fff4f1a86a96e8254a00bdf": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_dc0576828c0a46aca763bbe6bb9da8ed", - "IPY_MODEL_0288ed14b63c402db714c8b1692c0b63", - "IPY_MODEL_6a60004a82fe4abbaee62d412c61dd71" - ], - "layout": "IPY_MODEL_5d56734e94004fc9b076231a64aac73b" - } - }, - "dc0576828c0a46aca763bbe6bb9da8ed": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_53bab062ba374e0fb8cd8a0b037e2399", - "placeholder": "​", - "style": "IPY_MODEL_30a7f33261234f5a953718bb243adbc8", - "value": "Processing: 100%" - } + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "0288ed14b63c402db714c8b1692c0b63": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_a813fc57be61446faee2a510a4f553d2", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_26040ce648604f1192044c7119b67180", - "value": 81 - } + "id": "cF_mSA9Xim1j", + "outputId": "c99aa740-9f2e-49ce-c4a5-21d557b6623a" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Pycaret installed sucessfully!!\n" + ] + } + ], + "source": [ + "!pip install pycaret &> /dev/null\n", + "print (\"Pycaret installed sucessfully!!\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UjvBYy0Lim1m" + }, + "source": [ + "### **(b) Get the version of the pycaret**" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 35 }, - "6a60004a82fe4abbaee62d412c61dd71": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_fdbad02f20ed45ec9262cfa07d135866", - "placeholder": "​", - "style": "IPY_MODEL_5b98d73256314de08ed3b2a31aeb813e", - "value": " 81/81 [00:30<00:00, 2.14it/s]" - } - }, - "5d56734e94004fc9b076231a64aac73b": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "53bab062ba374e0fb8cd8a0b037e2399": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "30a7f33261234f5a953718bb243adbc8": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "a813fc57be61446faee2a510a4f553d2": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "26040ce648604f1192044c7119b67180": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "fdbad02f20ed45ec9262cfa07d135866": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "5b98d73256314de08ed3b2a31aeb813e": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "a625601ae65c45beab624a2792f17c5e": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_c03e2dc928b04c94b7da9136861b8b37", - "IPY_MODEL_e325e4d9001049bfb453861e084af26a", - "IPY_MODEL_539c563ff63a4ed49fdd37210f7b288b" - ], - "layout": "IPY_MODEL_f2acbf7ee7084343bc26083d628bb114" - } - }, - "c03e2dc928b04c94b7da9136861b8b37": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_551adee0c3d74bab875895a5abbcf63f", - "placeholder": "​", - "style": "IPY_MODEL_09c74c3a72b449288192c35e41f3928f", - "value": "Processing: 100%" - } - }, - "e325e4d9001049bfb453861e084af26a": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_8e0a2ebf90b0472996efbb3f62d1eef8", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_f303621e07064e5d9859413d61fcfbcd", - "value": 81 - } - }, - "539c563ff63a4ed49fdd37210f7b288b": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_ff7b5502ca6c4aceb4af044bd0ea9527", - "placeholder": "​", - "style": "IPY_MODEL_e48469c3a59149f698b79a692c747822", - "value": " 81/81 [00:24<00:00, 2.29it/s]" - } - }, - "f2acbf7ee7084343bc26083d628bb114": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "551adee0c3d74bab875895a5abbcf63f": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "09c74c3a72b449288192c35e41f3928f": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "8e0a2ebf90b0472996efbb3f62d1eef8": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "f303621e07064e5d9859413d61fcfbcd": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "ff7b5502ca6c4aceb4af044bd0ea9527": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "e48469c3a59149f698b79a692c747822": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "7e77c04e79a04111896d5933e2d0a0ca": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_36d298132395444ea7eaaa59f2fa60d9", - "IPY_MODEL_aff02182c01b44a6b1a69f00c855bee1", - "IPY_MODEL_9ad9dcb0a2c64c03af763d976136a264" - ], - "layout": "IPY_MODEL_08f0b54af19a49d482e8a32f7602c1f2" - } - }, - "36d298132395444ea7eaaa59f2fa60d9": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_9566cd7022bb4f19994513807bfd842e", - "placeholder": "​", - "style": "IPY_MODEL_f57549d4cbfb43deb8311a085557a8e9", - "value": "Processing: 100%" - } - }, - "aff02182c01b44a6b1a69f00c855bee1": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_e9a96a94f14148c49d2c782cf71e27aa", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_b1d962540b374fa1b022513b74412ef3", - "value": 81 - } - }, - "9ad9dcb0a2c64c03af763d976136a264": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_3f4666402a4a437eb90c12316dc1d9da", - "placeholder": "​", - "style": "IPY_MODEL_c98f51794e2342b2a9a662407622daf2", - "value": " 81/81 [00:41<00:00, 1.76it/s]" - } - }, - "08f0b54af19a49d482e8a32f7602c1f2": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "9566cd7022bb4f19994513807bfd842e": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "f57549d4cbfb43deb8311a085557a8e9": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "e9a96a94f14148c49d2c782cf71e27aa": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "b1d962540b374fa1b022513b74412ef3": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "3f4666402a4a437eb90c12316dc1d9da": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "c98f51794e2342b2a9a662407622daf2": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "223c0947254e425aa76368d44f7813d3": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_0d52f91171f5465f90e329bcb4588a4b", - "IPY_MODEL_00fb222c2cd0494e808960e3134c1a5c", - "IPY_MODEL_631717151fdd41dd8b6a829265bc5e65" - ], - "layout": "IPY_MODEL_4e28d0d0fda849bdb0887bdcce676a47" - } - }, - "0d52f91171f5465f90e329bcb4588a4b": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_5087cbae72274218ad25334b2551c671", - "placeholder": "​", - "style": "IPY_MODEL_51675431872b4184a9c5c710c39b64a0", - "value": "Processing: 100%" - } - }, - "00fb222c2cd0494e808960e3134c1a5c": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_3cf15a8c384e4541b9f4ae1b8313308e", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_2f198ed728bc44f2ad0a4fafee45d847", - "value": 81 - } - }, - "631717151fdd41dd8b6a829265bc5e65": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_683564dbcd4a407185756d6d139d8525", - "placeholder": "​", - "style": "IPY_MODEL_64b0dafcd97b45a1b80be93fe8c9b668", - "value": " 81/81 [00:30<00:00, 1.78it/s]" - } - }, - "4e28d0d0fda849bdb0887bdcce676a47": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "5087cbae72274218ad25334b2551c671": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "51675431872b4184a9c5c710c39b64a0": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "3cf15a8c384e4541b9f4ae1b8313308e": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "2f198ed728bc44f2ad0a4fafee45d847": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "683564dbcd4a407185756d6d139d8525": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "64b0dafcd97b45a1b80be93fe8c9b668": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "3cec08d71fa949be9b554961313bb327": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_3bc1cb0053a7442f87f7cd13265ec815", - "IPY_MODEL_e05d0af096014f5f833748c413c17dca", - "IPY_MODEL_e1f823c302d1497aa60ee24365fdae0d" - ], - "layout": "IPY_MODEL_0c5ab9e2b5cc45b7b90d3aec1985c4b3" - } - }, - "3bc1cb0053a7442f87f7cd13265ec815": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_3cbc587d0d2e43e1a308d9a6242db746", - "placeholder": "​", - "style": "IPY_MODEL_111fffcb13a5472c88984b3020bf1e69", - "value": "Processing: 100%" - } - }, - "e05d0af096014f5f833748c413c17dca": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_0b6eac0ee5544c97b54c39324036c326", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_12ae166387a74435a61969da3a68293b", - "value": 81 - } - }, - "e1f823c302d1497aa60ee24365fdae0d": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_37fc49a2a49d47f4a0d581cac2df7d6a", - "placeholder": "​", - "style": "IPY_MODEL_c6659b0659a24dec96791173d9e34555", - "value": " 81/81 [00:27<00:00, 2.27it/s]" - } - }, - "0c5ab9e2b5cc45b7b90d3aec1985c4b3": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "3cbc587d0d2e43e1a308d9a6242db746": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "111fffcb13a5472c88984b3020bf1e69": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "0b6eac0ee5544c97b54c39324036c326": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "12ae166387a74435a61969da3a68293b": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "37fc49a2a49d47f4a0d581cac2df7d6a": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "c6659b0659a24dec96791173d9e34555": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "c6f22d4bd1cb4c07850b7e055e617bed": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_bb316aa437a34355b21269d7086a2bdb", - "IPY_MODEL_b623ded86e1f4cc2a16325cd67747075", - "IPY_MODEL_a2d2b0dfe7a849bbb6bf4c7620cbf5c2" - ], - "layout": "IPY_MODEL_60c6b8c107ea458baaf26c2663999320" - } - }, - "bb316aa437a34355b21269d7086a2bdb": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_efd7895fe3de49fbbfe4e954671b2943", - "placeholder": "​", - "style": "IPY_MODEL_95d622af30524225a74644186d76a861", - "value": "Processing: 100%" - } - }, - "b623ded86e1f4cc2a16325cd67747075": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_9ecca98650c34c69b59abc3fd733648f", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_3b3301074c1e4907a7964afc70e1218e", - "value": 81 - } - }, - "a2d2b0dfe7a849bbb6bf4c7620cbf5c2": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_230ef89cdcaf4885bfbb3cffaadb4cc2", - "placeholder": "​", - "style": "IPY_MODEL_4095453c9c4f48ea81570d80060a06c4", - "value": " 81/81 [01:15<00:00, 1.18it/s]" - } - }, - "60c6b8c107ea458baaf26c2663999320": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "efd7895fe3de49fbbfe4e954671b2943": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "95d622af30524225a74644186d76a861": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "9ecca98650c34c69b59abc3fd733648f": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "3b3301074c1e4907a7964afc70e1218e": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "230ef89cdcaf4885bfbb3cffaadb4cc2": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "4095453c9c4f48ea81570d80060a06c4": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "b7a4f5905e504d41ba1124f2c7b363c2": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_70fc58f2565a4401adf41e2ae8d75a89", - "IPY_MODEL_e46d1b1d43a44bbc8945bd9c4ec32519", - "IPY_MODEL_2077179edf684d0783429eedb24cbea6" - ], - "layout": "IPY_MODEL_e2e49676f4f34815af11e380bcb6b27f" - } - }, - "70fc58f2565a4401adf41e2ae8d75a89": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_ece5244e61eb4747bb8375e322c55219", - "placeholder": "​", - "style": "IPY_MODEL_4bcc5940e7b649f1a9883251ee1f95e8", - "value": "Processing: 100%" - } - }, - "e46d1b1d43a44bbc8945bd9c4ec32519": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_53ee9ed83a124e7fb82b6f49d461f116", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_cf7aaacff5aa4eb3bae32a3a25fdab49", - "value": 81 - } - }, - "2077179edf684d0783429eedb24cbea6": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_97c8f63078834324b8a80968e900008c", - "placeholder": "​", - "style": "IPY_MODEL_850e2a8c0e5342b2a63503fc83cacb19", - "value": " 81/81 [00:31<00:00, 2.18it/s]" - } - }, - "e2e49676f4f34815af11e380bcb6b27f": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "ece5244e61eb4747bb8375e322c55219": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "4bcc5940e7b649f1a9883251ee1f95e8": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "53ee9ed83a124e7fb82b6f49d461f116": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "cf7aaacff5aa4eb3bae32a3a25fdab49": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "97c8f63078834324b8a80968e900008c": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "850e2a8c0e5342b2a63503fc83cacb19": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "ce49e4cd8b90424baa289827bbdf5709": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_1cb1af85286c47d3b182d29959ea8588", - "IPY_MODEL_7430bcd44a5248a492d03f902662c548", - "IPY_MODEL_4e21b1a0351341a9809ca5b4d05fd15a" - ], - "layout": "IPY_MODEL_4c21f0f814564cf5a160db87801c2855" - } - }, - "1cb1af85286c47d3b182d29959ea8588": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_57daa23de91a4472aba05a561b010c51", - "placeholder": "​", - "style": "IPY_MODEL_c682dcc24a674038b23bd9624cf905bb", - "value": "Processing: 100%" - } - }, - "7430bcd44a5248a492d03f902662c548": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_0992dbe948894830a1e996a6a3b4edb9", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_93d7b7f241104318b52e69a17c45c3cb", - "value": 81 - } - }, - "4e21b1a0351341a9809ca5b4d05fd15a": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_a8385f7b9c0e4cbb92a35ffd6c5a7621", - "placeholder": "​", - "style": "IPY_MODEL_f5fe066b5bc744379eeb3730a77e8b41", - "value": " 81/81 [00:29<00:00, 2.15it/s]" - } - }, - "4c21f0f814564cf5a160db87801c2855": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "57daa23de91a4472aba05a561b010c51": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "c682dcc24a674038b23bd9624cf905bb": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "0992dbe948894830a1e996a6a3b4edb9": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "93d7b7f241104318b52e69a17c45c3cb": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "a8385f7b9c0e4cbb92a35ffd6c5a7621": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "f5fe066b5bc744379eeb3730a77e8b41": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "00202b43d63b4a56bceb8c3cffcede0b": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_d796a62044644be4a181bbd58ba7fdac", - "IPY_MODEL_15448bdc87934e7c95a72facc3801566", - "IPY_MODEL_c0932334fe134f78a373bee51bd193df" - ], - "layout": "IPY_MODEL_a80999291470457ba7a1eb77893f01e4" - } - }, - "d796a62044644be4a181bbd58ba7fdac": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_fd51ef1e19a24261b2f9616c234fa7c8", - "placeholder": "​", - "style": "IPY_MODEL_33675eb92a284b8a95a677aa098b4f27", - "value": "Processing: 100%" - } - }, - "15448bdc87934e7c95a72facc3801566": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_a0824841b5b54a56a31aaf115822779d", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_e818471e3a8c4cc391131328ea4e89bd", - "value": 81 - } - }, - "c0932334fe134f78a373bee51bd193df": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_d9c42bab2447423b83ab296e467c503a", - "placeholder": "​", - "style": "IPY_MODEL_522745e40ca2425fbee6235a0761f8a0", - "value": " 81/81 [01:13<00:00, 1.19it/s]" - } - }, - "a80999291470457ba7a1eb77893f01e4": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "fd51ef1e19a24261b2f9616c234fa7c8": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "33675eb92a284b8a95a677aa098b4f27": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "a0824841b5b54a56a31aaf115822779d": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "e818471e3a8c4cc391131328ea4e89bd": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "d9c42bab2447423b83ab296e467c503a": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "522745e40ca2425fbee6235a0761f8a0": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "e0ea5230d0384b7482aa787a747de36a": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_e26cb115059243898cd52324cdc9e62e", - "IPY_MODEL_c0cedb777e244c7db748e075f9cd52eb", - "IPY_MODEL_5c5fa00457d8410fb64fa5228a7bc4fa" - ], - "layout": "IPY_MODEL_49cc270350234c2e865223b7605c6906" - } - }, - "e26cb115059243898cd52324cdc9e62e": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_7b2797c8874444c2bd2eaf18ba46e2cd", - "placeholder": "​", - "style": "IPY_MODEL_4b0a8272d56d46acb6f96af5b8c526e0", - "value": "Processing: 100%" - } - }, - "c0cedb777e244c7db748e075f9cd52eb": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_2443b48bb035438584749ad7912fb06e", - "max": 81, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_b106ed6bcaed4c07af5c646ff5ae2935", - "value": 81 - } - }, - "5c5fa00457d8410fb64fa5228a7bc4fa": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_02e851d1d6a54ff69fa898f1ee20c220", - "placeholder": "​", - "style": "IPY_MODEL_d5d2fc5df0b4433e9f6e54a9f288f129", - "value": " 81/81 [01:19<00:00, 1.05it/s]" - } - }, - "49cc270350234c2e865223b7605c6906": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "7b2797c8874444c2bd2eaf18ba46e2cd": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "4b0a8272d56d46acb6f96af5b8c526e0": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "2443b48bb035438584749ad7912fb06e": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "b106ed6bcaed4c07af5c646ff5ae2935": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "02e851d1d6a54ff69fa898f1ee20c220": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "d5d2fc5df0b4433e9f6e54a9f288f129": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "4bc7880483ef46f386959c1c9c23583c": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_a24ac38b47614b47ae89bf813d50ea78", - "IPY_MODEL_caf546ef00f74a87a0b25d0e94de523c", - "IPY_MODEL_1a67db658427407391b7601cb6e482bc" - ], - "layout": "IPY_MODEL_7606dd82dabe4348b1232ddf9e92108d" - } - }, - "a24ac38b47614b47ae89bf813d50ea78": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_2e079d47d0d84613a1d965acab007297", - "placeholder": "​", - "style": "IPY_MODEL_4b0af501fe7444a1aa15f8e110528df7", - "value": "Processing: 100%" - } - }, - "caf546ef00f74a87a0b25d0e94de523c": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_3ad24a154e454af3bd7096a4153dfc30", - "max": 4, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_a715a71eae764c958f3738734be0db32", - "value": 4 - } - }, - "1a67db658427407391b7601cb6e482bc": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_6b065bff14224bdda8e056048fd93c42", - "placeholder": "​", - "style": "IPY_MODEL_797b4411f13743e0a2b77b87005b4882", - "value": " 4/4 [00:03<00:00, 1.32it/s]" - } - }, - "7606dd82dabe4348b1232ddf9e92108d": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "2e079d47d0d84613a1d965acab007297": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "4b0af501fe7444a1aa15f8e110528df7": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "3ad24a154e454af3bd7096a4153dfc30": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "a715a71eae764c958f3738734be0db32": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "6b065bff14224bdda8e056048fd93c42": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "797b4411f13743e0a2b77b87005b4882": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "54be119808c7439796791b7720adc1d3": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_5320115c739147c691b9dafa7b5dbe56", - "IPY_MODEL_8752e62969d84e80944615987cb6eb25", - "IPY_MODEL_6b67ec9f59d341acbc4c97447474e6be" - ], - "layout": "IPY_MODEL_7ad52a834f1040aab17eda9e22509cf8" - } - }, - "5320115c739147c691b9dafa7b5dbe56": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_6e93a33e966645d9bf44b0d80fb4bb8d", - "placeholder": "​", - "style": "IPY_MODEL_6eae832719bc4a2f98ba1b8638cfb365", - "value": "Processing: 100%" - } - }, - "8752e62969d84e80944615987cb6eb25": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_ba1fd5f793d04c89b6f4ac6badbbef30", - "max": 4, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_6f068c43ac6a409084e8278e33c3f6ad", - "value": 4 - } - }, - "6b67ec9f59d341acbc4c97447474e6be": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_a3d5ce55e54644e59a5f954b43e43dd8", - "placeholder": "​", - "style": "IPY_MODEL_6ce36d2e1a12461e8b3ec9d1e234aa79", - "value": " 4/4 [00:03<00:00, 1.30it/s]" - } - }, - "7ad52a834f1040aab17eda9e22509cf8": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "6e93a33e966645d9bf44b0d80fb4bb8d": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "6eae832719bc4a2f98ba1b8638cfb365": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "ba1fd5f793d04c89b6f4ac6badbbef30": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "6f068c43ac6a409084e8278e33c3f6ad": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "a3d5ce55e54644e59a5f954b43e43dd8": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "6ce36d2e1a12461e8b3ec9d1e234aa79": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "98a085e08dac4d29ab7a3f9885d44847": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_47f270b6b4bc4fe79821e90bcd6b502c", - "IPY_MODEL_1d5bfcf3c81f46a7a02dbdb75f0f0438", - "IPY_MODEL_486eb5a29da44627bf11cf5c5b81fa44" - ], - "layout": "IPY_MODEL_59c886d4b12a49beab0e5945eff5e71d" - } - }, - "47f270b6b4bc4fe79821e90bcd6b502c": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_d11ebb2fe03e4f11b15228ce1f29b9d9", - "placeholder": "​", - "style": "IPY_MODEL_e003ee21215d43b686b9aa767c05eb11", - "value": "Processing: 0%" - } - }, - "1d5bfcf3c81f46a7a02dbdb75f0f0438": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_d49dc9db0e774391b15df304042f3897", - "max": 4, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_93b3bf9f0b114c79b72c0eba8f99369e", - "value": 4 - } - }, - "486eb5a29da44627bf11cf5c5b81fa44": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_fb61dd3e3bd2486ba584a94ff5586ea7", - "placeholder": "​", - "style": "IPY_MODEL_cb7da04066c74e09a20aacbe28d049a4", - "value": " 0/4 [00:00<?, ?it/s]" - } - }, - "59c886d4b12a49beab0e5945eff5e71d": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "d11ebb2fe03e4f11b15228ce1f29b9d9": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "e003ee21215d43b686b9aa767c05eb11": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "d49dc9db0e774391b15df304042f3897": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "93b3bf9f0b114c79b72c0eba8f99369e": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "fb61dd3e3bd2486ba584a94ff5586ea7": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "cb7da04066c74e09a20aacbe28d049a4": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "ad62d4b7874b4aaeb04617992ec150b3": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_b13a64f33a9a425b857dca3d86a44c41", - "IPY_MODEL_85c30c8a90184cfc94240ee203c0a97a", - "IPY_MODEL_61518410fb614859a0f2d7e84392139e" - ], - "layout": "IPY_MODEL_794f8392daeb42aab0e51f57b46bbae9" - } - }, - "b13a64f33a9a425b857dca3d86a44c41": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_87e85e0eb5144fe2a53a862e2147e828", - "placeholder": "​", - "style": "IPY_MODEL_967d8b4fe4204da6996d1e375cd2f4ac", - "value": "Processing: 86%" - } - }, - "85c30c8a90184cfc94240ee203c0a97a": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_a92c8a96315c40e1b56b24a3d0eeb40a", - "max": 7, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_785a653ecdd248e1a7d8c79298405e9f", - "value": 7 - } - }, - "61518410fb614859a0f2d7e84392139e": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_09c0b84753084630836d0227b99685a7", - "placeholder": "​", - "style": "IPY_MODEL_66ff23480f3e436da96158a2106dc1e4", - "value": " 6/7 [00:37<00:05, 5.13s/it]" - } - }, - "794f8392daeb42aab0e51f57b46bbae9": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": "hidden", - "width": null - } - }, - "87e85e0eb5144fe2a53a862e2147e828": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "967d8b4fe4204da6996d1e375cd2f4ac": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "a92c8a96315c40e1b56b24a3d0eeb40a": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "785a653ecdd248e1a7d8c79298405e9f": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "09c0b84753084630836d0227b99685a7": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "66ff23480f3e436da96158a2106dc1e4": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - } - } - } - }, - "cells": [ - { - "cell_type": "markdown", - "metadata": { - "id": "view-in-github", - "colab_type": "text" - }, - "source": [ - "\"Open" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "7nQSRI4Dim1Y" - }, - "source": [ - "---\n", - "# **PyCaret for Regression**\n", - "---\n", - "- It is a bundle of many Machine Learning algorithms.\n", - "- Only three lines of code is required to compare 20 ML models.\n", - "- Pycaret is available for:\n", - " - Classification\n", - " - Regression\n", - " - Clustering\n", - "---\n", - "\n", - "### **Self learning resource**\n", - "1. Tutorial on Pycaret ** Click Here**\n", - "\n", - "2. Documentation on Pycaret-Regression: ** Click Here **\n", - "\n", - "---\n", - "\n", - "### **In this tutorial we will learn:**\n", - "\n", - "- Getting Data\n", - "- Setting up Environment\n", - "- Create Model\n", - "- Tune Model\n", - "- Plot Model\n", - "- Finalize Model\n", - "- Predict Model\n", - "- Save / Load Model\n", - "---\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "A30y-VtNim1h" - }, - "source": [ - "### **(a) Install Pycaret**" - ] - }, - { - "cell_type": "code", - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "cF_mSA9Xim1j", - "outputId": "ba01eff2-5618-4ec6-e731-02da5c811ef2" - }, - "source": [ - "!pip install pycaret &> /dev/null\n", - "print (\"Pycaret installed sucessfully!!\")" - ], - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Pycaret installed sucessfully!!\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "UjvBYy0Lim1m" - }, - "source": [ - "### **(b) Get the version of the pycaret**" - ] - }, - { - "cell_type": "code", - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 35 - }, - "id": "NHB2KMhZim1o", - "outputId": "8446a5f0-70d2-4376-8988-c97faccd1c2a" - }, - "source": [ - "from pycaret.utils import version\n", - "version()" - ], - "execution_count": null, - "outputs": [ - { - "output_type": "execute_result", - "data": { - "text/plain": [ - "'3.2.0'" - ], - "application/vnd.google.colaboratory.intrinsic+json": { - "type": "string" - } - }, - "metadata": {}, - "execution_count": 48 - } - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "BTLd3s--im1c" - }, - "source": [ - "---\n", - "# **1. Regression: Basics**\n", - "---\n", - "### **1.1 Get the list of datasets available in pycaret (Total Datasets = 55)**\n", - "\n", - "\n" - ] - }, - { - "cell_type": "code", - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 1000 + "id": "NHB2KMhZim1o", + "outputId": "8c8c0440-3f25-4e4e-a020-e30dc08024b9" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "'3.3.2'" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "string" + } + }, + "metadata": {}, + "execution_count": 2 + } + ], + "source": [ + "from pycaret.utils import version\n", + "version()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BTLd3s--im1c" + }, + "source": [ + "---\n", + "# **1. Regression: Basics**\n", + "---\n", + "### **1.1 Get the list of datasets available in pycaret (Total Datasets = 56)**\n", + "\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 }, "id": "veBU-f21im1x", - "outputId": "61856d7d-ceb0-42ae-ba48-299bac98a169" + "outputId": "30b7d91c-8ab0-4008-d2a0-2f53f6e37f7f" }, - "source": [ - "from pycaret.datasets import get_data\n", - "dataSets = get_data('index')" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -5031,62 +210,62 @@ "55 traffic Multivariate \n", "\n", " Default Task Target Variable 1 Target Variable 2 \\\n", - "0 Anomaly Detection None None \n", + "0 Anomaly Detection NaN NaN \n", "1 Association Rule Mining InvoiceNo Description \n", "2 Association Rule Mining InvoiceNo Description \n", - "3 Classification (Binary) deposit None \n", - "4 Classification (Binary) Class None \n", - "5 Classification (Binary) Class None \n", - "6 Classification (Binary) default None \n", - "7 Classification (Binary) Class variable None \n", - "8 Classification (Binary) stabf None \n", - "9 Classification (Binary) left None \n", - "10 Classification (Binary) DEATH None \n", - "11 Classification (Binary) Disease None \n", - "12 Classification (Binary) Class None \n", - "13 Classification (Binary) income >50K None \n", - "14 Classification (Binary) Purchase None \n", - "15 Classification (Binary) TARGET_5Yrs None \n", - "16 Classification (Binary) type None \n", - "17 Classification (Binary) Class None \n", - "18 Classification (Binary) Survived None \n", - "19 Classification (Binary) party_winner None \n", - "20 Classification (Multiclass) Type None \n", - "21 Classification (Multiclass) species None \n", - "22 Classification (Multiclass) CLASS None \n", - "23 Classification (Multiclass) Next_Question None \n", - "24 Classification (Multiclass) Class None \n", - "25 Classification (Multiclass) NSP None \n", - "26 Clustering None None \n", - "27 Clustering None None \n", - "28 Clustering None None \n", - "29 Clustering None None \n", - "30 Clustering None None \n", - "31 Clustering None None \n", - "32 Clustering None None \n", - "33 Clustering None None \n", - "34 Clustering None None \n", - "35 Clustering None None \n", - "36 Clustering None None \n", - "37 Clustering None None \n", - "38 Clustering None None \n", - "39 NLP tweet None \n", - "40 NLP / Classification reviewText None \n", - "41 NLP / Classification en None \n", - "42 NLP / Regression text None \n", - "43 NLP / Classification Text None \n", - "44 Regression price None \n", - "45 Regression cnt None \n", - "46 Regression medv None \n", - "47 Regression strength None \n", - "48 Regression Price None \n", + "3 Classification (Binary) deposit NaN \n", + "4 Classification (Binary) Class NaN \n", + "5 Classification (Binary) Class NaN \n", + "6 Classification (Binary) default NaN \n", + "7 Classification (Binary) Class variable NaN \n", + "8 Classification (Binary) stabf NaN \n", + "9 Classification (Binary) left NaN \n", + "10 Classification (Binary) DEATH NaN \n", + "11 Classification (Binary) Disease NaN \n", + "12 Classification (Binary) Class NaN \n", + "13 Classification (Binary) income >50K NaN \n", + "14 Classification (Binary) Purchase NaN \n", + "15 Classification (Binary) TARGET_5Yrs NaN \n", + "16 Classification (Binary) type NaN \n", + "17 Classification (Binary) Class NaN \n", + "18 Classification (Binary) Survived NaN \n", + "19 Classification (Binary) party_winner NaN \n", + "20 Classification (Multiclass) Type NaN \n", + "21 Classification (Multiclass) species NaN \n", + "22 Classification (Multiclass) CLASS NaN \n", + "23 Classification (Multiclass) Next_Question NaN \n", + "24 Classification (Multiclass) Class NaN \n", + "25 Classification (Multiclass) NSP NaN \n", + "26 Clustering NaN NaN \n", + "27 Clustering NaN NaN \n", + "28 Clustering NaN NaN \n", + "29 Clustering NaN NaN \n", + "30 Clustering NaN NaN \n", + "31 Clustering NaN NaN \n", + "32 Clustering NaN NaN \n", + "33 Clustering NaN NaN \n", + "34 Clustering NaN NaN \n", + "35 Clustering NaN NaN \n", + "36 Clustering NaN NaN \n", + "37 Clustering NaN NaN \n", + "38 Clustering NaN NaN \n", + "39 NLP tweet NaN \n", + "40 NLP / Classification reviewText NaN \n", + "41 NLP / Classification en NaN \n", + "42 NLP / Regression text NaN \n", + "43 NLP / Classification Text NaN \n", + "44 Regression price NaN \n", + "45 Regression cnt NaN \n", + "46 Regression medv NaN \n", + "47 Regression strength NaN \n", + "48 Regression Price NaN \n", "49 Regression Heating Load Cooling Load \n", - "50 Regression area None \n", - "51 Regression Gold_T+22 None \n", - "52 Regression SalePrice None \n", - "53 Regression charges None \n", - "54 Regression PPE None \n", - "55 Regression traffic_volume None \n", + "50 Regression area NaN \n", + "51 Regression Gold_T+22 NaN \n", + "52 Regression SalePrice NaN \n", + "53 Regression charges NaN \n", + "54 Regression PPE NaN \n", + "55 Regression traffic_volume NaN \n", "\n", " # Instances # Attributes Missing Values \n", "0 1000 10 N \n", @@ -5148,7 +327,7 @@ ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 DescriptionValueDescriptionValue
0Session id61780Session id7985
1Targetmedv1Targetmedv
2Target typeRegression2Target typeRegression
3Original data shape(506, 14)3Original data shape(506, 14)
4Transformed data shape(506, 14)4Transformed data shape(506, 14)
5Transformed train set shape(354, 14)5Transformed train set shape(354, 14)
6Transformed test set shape(152, 14)6Transformed test set shape(152, 14)
7Numeric features137Numeric features13
8PreprocessTrue8PreprocessTrue
9Imputation typesimple9Imputation typesimple
10Numeric imputationmean10Numeric imputationmean
11Categorical imputationmode11Categorical imputationmode
12Fold GeneratorKFold12Fold GeneratorKFold
13Fold Number1013Fold Number10
14CPU Jobs-114CPU Jobs-1
15Use GPUFalse15Use GPUFalse
16Log ExperimentFalse16Log ExperimentFalse
17Experiment Namereg-default-name17Experiment Namereg-default-name
18USI1e0d18USI3e56
\n" @@ -6551,6 +1763,17 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.regression import *\n", + "s = setup(data = myDataSet, target = 'medv')\n", + "\n", + "# Comment the above line-code and Uncomment below line-code and re-run it\n", + "#s = setup(data = myDataSet, target = 'medv', train_size = 0.7, data_split_shuffle = False)\n", + "\n", + "# Other Parameters:\n", + "# train_size = 0.7\n", + "# data_split_shuffle = False" ] }, { @@ -6560,37 +1783,34 @@ }, "source": [ "---\n", - "### **1.4 Run all models (Step-III)**\n", + "### **1.5 Run all models (Step-III)**\n", "---" ] }, { "cell_type": "code", + "execution_count": 7, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 645, "referenced_widgets": [ - "f887b0172fff4f1a86a96e8254a00bdf", - "dc0576828c0a46aca763bbe6bb9da8ed", - "0288ed14b63c402db714c8b1692c0b63", - "6a60004a82fe4abbaee62d412c61dd71", - "5d56734e94004fc9b076231a64aac73b", - "53bab062ba374e0fb8cd8a0b037e2399", - "30a7f33261234f5a953718bb243adbc8", - "a813fc57be61446faee2a510a4f553d2", - "26040ce648604f1192044c7119b67180", - "fdbad02f20ed45ec9262cfa07d135866", - "5b98d73256314de08ed3b2a31aeb813e" + "be2a0d162e804442a0edf19f174179df", + "0cec87531fac407fa41ac68d0c1cd0b5", + "e32694d360e54d419294a3401fa37ddd", + "bc1e50d27fd24daf96adade744a0bedd", + "c114c97f11f24e5d914bda37405519f8", + "5837728f03944861b2a6a6e6779ef21f", + "5205da94e1b044bd9678da2e2f2412c2", + "448ae66e6e41478dbd7987e9a9d1940d", + "622123b39cec44c78412c45308712f39", + "fc105db5c3014f60b5ff622bfdd41575", + "f2c09272223349deb11230fc0d4a59c3" ] }, "id": "XmPJbrOXim19", - "outputId": "babc46ec-3851-43d6-99a6-d571aa31c6bb" + "outputId": "1db8457e-986b-4edb-acf1-5a7a48c1cda9" }, - "source": [ - "cm = compare_models()" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -6606,253 +1826,253 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
gbrGradient Boosting Regressor2.17959.92663.09500.86570.14090.11120.2310
etExtra Trees Regressor2.310111.82323.37510.84740.14530.11310.3450
xgboostExtreme Gradient Boosting2.342312.14033.42580.84180.15050.11640.2760
rfRandom Forest Regressor2.364012.36303.47330.83600.14950.11850.4200
lightgbmLight Gradient Boosting Machine2.438913.13793.56200.83030.15230.11930.2610
adaAdaBoost Regressor2.956818.40654.22400.76250.19180.15710.2930
dtDecision Tree Regressor3.182024.78254.79080.67560.21730.16270.0230
lrLinear Regression3.712427.30345.19400.65460.26100.18200.0220
ridgeRidge Regression3.704827.61095.22070.65050.27310.18270.0360
larLeast Angle Regression3.746627.61075.22600.64940.26280.18520.0220
brBayesian Ridge3.767328.93025.33950.63510.28500.18490.0340
enElastic Net3.930931.30335.53950.61090.28860.18760.0200
lassoLasso Regression4.009932.46295.63760.59720.28280.18980.0200
llarLasso Least Angle Regression4.010032.46285.63760.59720.28290.18980.0230
huberHuber Regressor4.186136.13605.98020.53600.28780.21120.0760
knnK Neighbors Regressor4.959652.33177.09370.35660.27940.23540.0430
ompOrthogonal Matching Pursuit6.096669.85408.29080.13540.32990.29290.0210
dummyDummy Regressor6.797487.08289.2375-0.05960.38810.36180.0370
parPassive Aggressive Regressor7.9059106.026610.1159-0.35160.47940.39000.0380etExtra Trees Regressor2.180811.59113.25160.85500.13720.10640.1990
lightgbmLight Gradient Boosting Machine2.190511.66873.24110.85300.14210.10840.3880
gbrGradient Boosting Regressor2.292611.89663.31660.85180.14730.11530.1520
xgboostExtreme Gradient Boosting2.325512.89993.47330.83950.15080.11530.2250
rfRandom Forest Regressor2.367813.24523.47920.83500.15260.11870.3230
adaAdaBoost Regressor2.829416.14903.89190.79680.17930.14810.1200
dtDecision Tree Regressor3.116321.67824.51550.72300.19090.15090.0450
larLeast Angle Regression3.441922.54054.66580.72260.24310.17610.0260
lrLinear Regression3.412222.78514.69000.71790.24200.17380.6260
ridgeRidge Regression3.414222.95494.70690.71650.25340.17450.0250
brBayesian Ridge3.463723.50154.77110.71110.24810.17590.0370
enElastic Net3.639826.36945.06600.68170.24130.17440.0220
lassoLasso Regression3.645126.61195.08080.67960.24410.17370.0210
llarLasso Least Angle Regression3.645126.61215.08090.67960.24410.17370.0220
huberHuber Regressor3.814930.25615.41400.62570.25910.19520.0760
knnK Neighbors Regressor4.414641.15596.38350.49050.24340.20500.0460
ompOrthogonal Matching Pursuit5.933764.40317.95860.22200.32340.29070.0400
dummyDummy Regressor6.751385.01959.1746-0.02750.39370.36800.0190
parPassive Aggressive Regressor7.159593.05809.3543-0.17180.45190.36440.0350
\n" @@ -6869,7 +2089,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "f887b0172fff4f1a86a96e8254a00bdf" + "model_id": "be2a0d162e804442a0edf19f174179df" } }, "metadata": { @@ -6892,6 +2112,9 @@ }, "metadata": {} } + ], + "source": [ + "cm = compare_models()" ] }, { @@ -6901,44 +2124,387 @@ }, "source": [ "---\n", - "### **1.5 \"Three line of code\" for model comparison for \"Boston\" dataset**\n", + "### **1.6 \"Three line of code\" for model comparison for \"Boston\" dataset**\n", "---\n", "\n" ] }, { "cell_type": "code", + "execution_count": 8, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 645, "referenced_widgets": [ - "a625601ae65c45beab624a2792f17c5e", - "c03e2dc928b04c94b7da9136861b8b37", - "e325e4d9001049bfb453861e084af26a", - "539c563ff63a4ed49fdd37210f7b288b", - "f2acbf7ee7084343bc26083d628bb114", - "551adee0c3d74bab875895a5abbcf63f", - "09c74c3a72b449288192c35e41f3928f", - "8e0a2ebf90b0472996efbb3f62d1eef8", - "f303621e07064e5d9859413d61fcfbcd", - "ff7b5502ca6c4aceb4af044bd0ea9527", - "e48469c3a59149f698b79a692c747822" + "b01b076f341b4870a857260ebc58aaa6", + "dd97ebed0c0d4024816da2e5b520801c", + "defbd1445d7c457e88f40a438e2218ed", + "cbc1cc9aa8164ec0ad08f29ce297977a", + "fc9542622b214eb68c7a2fca1e7b2ac3", + "e91eaa0ca891497899ffa1f57bec0769", + "4114a648cb7c4c2abf93665371276dce", + "32ac0e66a8c14e1ca52d903f703224c0", + "7b49b07c306f4035836a655ebf5b58bf", + "c759b79c49e34a1d8a8cb4f843a4cd8c", + "97d246863e9e4986a8366fbc36ec4a72" ] }, "id": "QPY6bvTsim1_", - "outputId": "2cec537e-36cb-49a7-f6ff-01bcf59b2630" + "outputId": "e3eb1006-4ded-4907-ecf5-048bfe943bb9" }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [] + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)
etExtra Trees Regressor2.36349.82923.02260.71720.11830.10030.1990
rfRandom Forest Regressor2.38579.89123.06050.70630.11930.10070.5760
lightgbmLight Gradient Boosting Machine2.30849.09512.95140.67820.11490.09660.0880
xgboostExtreme Gradient Boosting2.544012.69613.35980.67280.12520.10370.1720
lrLinear Regression2.655012.57293.34020.65390.13480.11090.0270
ridgeRidge Regression2.634612.47863.32520.65220.13390.10970.0230
brBayesian Ridge2.635312.48043.32580.65210.13400.10970.0220
adaAdaBoost Regressor2.639811.18193.26500.60640.12960.11430.1220
llarLasso Least Angle Regression2.930616.84173.90210.57720.15070.11520.0230
lassoLasso Regression2.930616.84233.90220.57720.15070.11520.0230
gbrGradient Boosting Regressor2.538811.03673.23810.57700.12690.10750.1380
larLeast Angle Regression2.739012.59333.40410.55730.15350.11940.0280
huberHuber Regressor3.344121.18444.34100.47440.16240.13240.0490
dtDecision Tree Regressor3.157817.00824.03860.41320.16040.13600.0250
enElastic Net3.529223.89064.66770.39130.19380.13920.0230
ompOrthogonal Matching Pursuit6.257171.29618.0075-0.90180.28650.25290.0260
dummyDummy Regressor6.653476.80538.2838-1.30090.30040.27360.0230
knnK Neighbors Regressor7.128095.62719.4698-2.97820.33490.29170.0280
parPassive Aggressive Regressor9.5028144.963111.0160-4.72530.46290.38390.0270
\n" + ] + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing: 0%| | 0/81 [00:00" + ], + "text/html": [] + }, + "metadata": {} + } + ], "source": [ - "from pickle import TRUE\n", "from pycaret.datasets import get_data\n", "from pycaret.regression import *\n", "\n", - "bostonDataSet = get_data(\"boston\", verbose=False)\n", - "setup(data = bostonDataSet, target='medv', train_size = 0.7, data_split_shuffle = False, verbose=False)\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv',\n", + " train_size = 0.7,\n", + " data_split_shuffle = False,\n", + " verbose = False)\n", + "\n", "cm = compare_models()" - ], - "execution_count": null, + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3Ij2kL4zoMG4" + }, + "source": [ + "---\n", + "### **1.7 \"Three line of code\" for model comparison for \"Insurance\" dataset**\n", + "---\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 645, + "referenced_widgets": [ + "c9a4fa0122794c9285dc7b5eb467f1d4", + "a916a19e3d994e098285ad81adf51d52", + "66bf68d01ef748ff95cff02d5bc8fe1a", + "d76c8aa20ea34a60a8749bfa4540c382", + "2f551e9288e24847a22b3bf2387b1de4", + "d9c996d95d954effb0f74b58bce9bded", + "d25198d73f16475c8ca93fc96bf17d6f", + "56797a2690c34df3b9237ff6237f9206", + "bc482d33717e45eab17b3779e9250964", + "6388b4c8a0c04c0aaf4653fd716f13b5", + "97300eb3c69646aea9d0c77067c49cec" + ] + }, + "id": "MnJrgvdkoMG9", + "outputId": "d1be1e65-b8e5-42a3-e60b-866aab8791ac" + }, "outputs": [ { "output_type": "display_data", @@ -6954,338 +2520,867 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
gbrGradient Boosting Regressor2373.753920023370.20234405.12360.85540.40300.26530.2360
lightgbmLight Gradient Boosting Machine2684.864622189126.67774653.62940.84060.54120.33800.4610
rfRandom Forest Regressor2584.466122828059.18814719.99420.83620.44820.30860.3850
etExtra Trees Regressor2545.123225213731.64784944.16650.81900.45060.29500.3210
adaAdaBoost Regressor3871.294725380185.15545005.39190.81810.59610.65490.0980
xgboostExtreme Gradient Boosting2884.808426410989.40005095.56080.80910.50780.35410.2390
ridgeRidge Regression4044.460135134281.50405908.34340.74890.59820.40970.0850
brBayesian Ridge4037.856835134047.17405908.18780.74880.61650.40860.0820
llarLasso Least Angle Regression4031.936935132513.67785907.95050.74880.62270.40770.0830
larLeast Angle Regression4031.678735132685.40745907.94260.74880.63160.40760.0820
lassoLasso Regression4031.939735132519.86135907.95100.74880.62270.40770.0840
lrLinear Regression4031.678735132685.40745907.94260.74880.63160.40760.0990
dtDecision Tree Regressor2859.040238904715.27556186.89060.71990.49440.32790.1450
huberHuber Regressor3261.938245115822.57576677.87140.67160.43620.21130.1850
parPassive Aggressive Regressor4143.604863101564.07887887.63400.56180.50730.26780.0880
enElastic Net7132.954587975502.11319347.47800.38850.71830.88590.0810
ompOrthogonal Matching Pursuit8800.5636129809259.541511348.83020.09950.85991.10470.0820
knnK Neighbors Regressor7882.2624132560132.000011494.39840.06840.84380.89340.1570
dummyDummy Regressor8968.3332144843431.200011998.1181-0.00651.00571.53550.0810
\n" + ] + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing: 0%| | 0/81 [00:00" + ], + "text/html": [] + }, + "metadata": {} + } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "\n", + "myDataSet = get_data(\"insurance\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'charges',\n", + " train_size = 0.7,\n", + " data_split_shuffle = False,\n", + " verbose = False)\n", + "\n", + "cm = compare_models()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "aIIVJHZC3k4b" + }, + "source": [ + "---\n", + "### **1.8 List all available models in pycaret**\n", + "---\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 865 + }, + "outputId": "9d20240b-6ec6-4b8c-bd22-854e3770c73f", + "id": "2d0IzyPP3k4c" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " Name \\\n", + "ID \n", + "lr Linear Regression \n", + "lasso Lasso Regression \n", + "ridge Ridge Regression \n", + "en Elastic Net \n", + "lar Least Angle Regression \n", + "llar Lasso Least Angle Regression \n", + "omp Orthogonal Matching Pursuit \n", + "br Bayesian Ridge \n", + "ard Automatic Relevance Determination \n", + "par Passive Aggressive Regressor \n", + "ransac Random Sample Consensus \n", + "tr TheilSen Regressor \n", + "huber Huber Regressor \n", + "kr Kernel Ridge \n", + "svm Support Vector Regression \n", + "knn K Neighbors Regressor \n", + "dt Decision Tree Regressor \n", + "rf Random Forest Regressor \n", + "et Extra Trees Regressor \n", + "ada AdaBoost Regressor \n", + "gbr Gradient Boosting Regressor \n", + "mlp MLP Regressor \n", + "xgboost Extreme Gradient Boosting \n", + "lightgbm Light Gradient Boosting Machine \n", + "dummy Dummy Regressor \n", + "\n", + " Reference Turbo \n", + "ID \n", + "lr sklearn.linear_model._base.LinearRegression True \n", + "lasso sklearn.linear_model._coordinate_descent.Lasso True \n", + "ridge sklearn.linear_model._ridge.Ridge True \n", + "en sklearn.linear_model._coordinate_descent.Elast... True \n", + "lar sklearn.linear_model._least_angle.Lars True \n", + "llar sklearn.linear_model._least_angle.LassoLars True \n", + "omp sklearn.linear_model._omp.OrthogonalMatchingPu... True \n", + "br sklearn.linear_model._bayes.BayesianRidge True \n", + "ard sklearn.linear_model._bayes.ARDRegression False \n", + "par sklearn.linear_model._passive_aggressive.Passi... True \n", + "ransac sklearn.linear_model._ransac.RANSACRegressor False \n", + "tr sklearn.linear_model._theil_sen.TheilSenRegressor False \n", + "huber sklearn.linear_model._huber.HuberRegressor True \n", + "kr sklearn.kernel_ridge.KernelRidge False \n", + "svm sklearn.svm._classes.SVR False \n", + "knn sklearn.neighbors._regression.KNeighborsRegressor True \n", + "dt sklearn.tree._classes.DecisionTreeRegressor True \n", + "rf sklearn.ensemble._forest.RandomForestRegressor True \n", + "et sklearn.ensemble._forest.ExtraTreesRegressor True \n", + "ada sklearn.ensemble._weight_boosting.AdaBoostRegr... True \n", + "gbr sklearn.ensemble._gb.GradientBoostingRegressor True \n", + "mlp sklearn.neural_network._multilayer_perceptron.... False \n", + "xgboost xgboost.sklearn.XGBRegressor True \n", + "lightgbm lightgbm.sklearn.LGBMRegressor True \n", + "dummy sklearn.dummy.DummyRegressor True " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", - "
NameReferenceTurbo
ID
rfRandom Forest Regressor2.36799.76873.03710.71780.11820.09960.3210
etExtra Trees Regressor2.35679.63223.01870.71170.11830.10010.2560
lightgbmLight Gradient Boosting Machine2.30849.09512.95140.67820.11490.09660.6730
xgboostExtreme Gradient Boosting2.544012.69613.35980.67280.12520.10370.1570
lrLinear Regression2.655012.57293.34020.65390.13480.11090.0220
ridgeRidge Regression2.634612.47863.32520.65220.13390.10970.0210
brBayesian Ridge2.635312.48043.32580.65210.13400.10970.0240
adaAdaBoost Regressor2.759411.89033.36600.59170.13290.11890.1010
gbrGradient Boosting Regressor2.535511.00253.23350.57960.12670.10730.1160
llarLasso Least Angle Regression2.930616.84173.90210.57720.15070.11520.0220
lassoLasso Regression2.930616.84233.90220.57720.15070.11520.0210
larLeast Angle Regression2.739012.59333.40410.55730.15350.11940.0230
huberHuber Regressor3.344121.18444.34100.47440.16240.13240.0470
enElastic Net3.529223.89064.66770.39130.19380.13920.0230
dtDecision Tree Regressor3.073015.75453.90320.32410.15490.13250.0230
ompOrthogonal Matching Pursuit6.257171.29618.0075-0.90180.28650.25290.0220
dummyDummy Regressor6.653476.80538.2838-1.30090.30040.27360.0190
parPassive Aggressive Regressor7.4055102.75299.0450-1.43040.37520.29510.0210
knnK Neighbors Regressor7.128095.62719.4698-2.97820.33490.29170.0280lrLinear Regressionsklearn.linear_model._base.LinearRegressionTrue
lassoLasso Regressionsklearn.linear_model._coordinate_descent.LassoTrue
ridgeRidge Regressionsklearn.linear_model._ridge.RidgeTrue
enElastic Netsklearn.linear_model._coordinate_descent.Elast...True
larLeast Angle Regressionsklearn.linear_model._least_angle.LarsTrue
llarLasso Least Angle Regressionsklearn.linear_model._least_angle.LassoLarsTrue
ompOrthogonal Matching Pursuitsklearn.linear_model._omp.OrthogonalMatchingPu...True
brBayesian Ridgesklearn.linear_model._bayes.BayesianRidgeTrue
ardAutomatic Relevance Determinationsklearn.linear_model._bayes.ARDRegressionFalse
parPassive Aggressive Regressorsklearn.linear_model._passive_aggressive.Passi...True
ransacRandom Sample Consensussklearn.linear_model._ransac.RANSACRegressorFalse
trTheilSen Regressorsklearn.linear_model._theil_sen.TheilSenRegressorFalse
huberHuber Regressorsklearn.linear_model._huber.HuberRegressorTrue
krKernel Ridgesklearn.kernel_ridge.KernelRidgeFalse
svmSupport Vector Regressionsklearn.svm._classes.SVRFalse
knnK Neighbors Regressorsklearn.neighbors._regression.KNeighborsRegressorTrue
dtDecision Tree Regressorsklearn.tree._classes.DecisionTreeRegressorTrue
rfRandom Forest Regressorsklearn.ensemble._forest.RandomForestRegressorTrue
etExtra Trees Regressorsklearn.ensemble._forest.ExtraTreesRegressorTrue
adaAdaBoost Regressorsklearn.ensemble._weight_boosting.AdaBoostRegr...True
gbrGradient Boosting Regressorsklearn.ensemble._gb.GradientBoostingRegressorTrue
mlpMLP Regressorsklearn.neural_network._multilayer_perceptron....False
xgboostExtreme Gradient Boostingxgboost.sklearn.XGBRegressorTrue
lightgbmLight Gradient Boosting Machinelightgbm.sklearn.LGBMRegressorTrue
dummyDummy Regressorsklearn.dummy.DummyRegressorTrue
\n" - ] - }, - "metadata": {} - }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "Processing: 0%| | 0/81 [00:00\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "\n", + "
\n", + "
\n" ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "a625601ae65c45beab624a2792f17c5e" - } - }, - "metadata": { - "application/vnd.jupyter.widget-view+json": { - "colab": { - "custom_widget_manager": { - "url": "https://ssl.gstatic.com/colaboratory-static/widgets/colab-cdn-widget-manager/2b70e893a8ba7c0f/manager.min.js" - } - } + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "summary": "{\n \"name\": \"models()\",\n \"rows\": 25,\n \"fields\": [\n {\n \"column\": \"ID\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 25,\n \"samples\": [\n \"ard\",\n \"dt\",\n \"lr\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Name\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 25,\n \"samples\": [\n \"Automatic Relevance Determination\",\n \"Decision Tree Regressor\",\n \"Linear Regression\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Reference\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 25,\n \"samples\": [\n \"sklearn.linear_model._bayes.ARDRegression\",\n \"sklearn.tree._classes.DecisionTreeRegressor\",\n \"sklearn.linear_model._base.LinearRegression\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Turbo\",\n \"properties\": {\n \"dtype\": \"boolean\",\n \"num_unique_values\": 2,\n \"samples\": [\n false,\n true\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } - } - }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [] }, - "metadata": {} + "metadata": {}, + "execution_count": 10 } + ], + "source": [ + "models()" ] }, { "cell_type": "markdown", "metadata": { - "id": "3Ij2kL4zoMG4" + "id": "9GWuAHd7xQ8i" + }, + "source": [ + "---\n", + "# **2. Regression: working with user dataset**\n", + "---\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "id": "6ztv-AJdmhcf", + "scrolled": true + }, + "outputs": [], + "source": [ + "from pycaret.regression import *\n", + "import pandas as pd\n", + "\n", + "# First upload the user file (myDataSet.csv) in the colab\n", + "\n", + "#myDataSet = pd.read_csv(\"myDataSet.csv\") # Uncomment and execute\n", + "#s = setup(data = myDataSet, target = 'RMSD', verbose=False) # Uncomment and execute\n", + "#cm = compare_models() # Uncomment and execute" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "zUfHO4wKim1_" }, "source": [ "---\n", - "### **1.6 \"Three line of code\" for model comparison for \"Insurance\" dataset**\n", + "# **3. Regression: Apply \"Data Preprocessing\"**\n", "---\n", - "\n" + "\n", + "### **3.1 Model performance using \"Normalization\"**" ] }, { "cell_type": "code", + "execution_count": 12, "metadata": { + "id": "Atd6jNvmim2A", "colab": { "base_uri": "https://localhost:8080/", "height": 645, "referenced_widgets": [ - "7e77c04e79a04111896d5933e2d0a0ca", - "36d298132395444ea7eaaa59f2fa60d9", - "aff02182c01b44a6b1a69f00c855bee1", - "9ad9dcb0a2c64c03af763d976136a264", - "08f0b54af19a49d482e8a32f7602c1f2", - "9566cd7022bb4f19994513807bfd842e", - "f57549d4cbfb43deb8311a085557a8e9", - "e9a96a94f14148c49d2c782cf71e27aa", - "b1d962540b374fa1b022513b74412ef3", - "3f4666402a4a437eb90c12316dc1d9da", - "c98f51794e2342b2a9a662407622daf2" + "f760f6ffd48f49fab40e9beb4235130a", + "0745110516644d7482695ebdbda30023", + "74798fc0a4044f19921697f2041a84cf", + "45b6c060884047fab46ea787a445afc0", + "45fd289b714d441e8f5aa60f1594280e", + "8cceabe4e11a4a02bea0e73c9c043dd1", + "4c1b70cf7d364e89b2cf559a5b68e5cf", + "31d390c9e7184be497579fdd07e0e529", + "6ec62c1b8b554276bf78e08d6bb684c1", + "0cb0a8bf40af4ad89fd3a5c00ae931c6", + "c592f4b732f14fdca367bb9a9afa4f0f" ] }, - "id": "MnJrgvdkoMG9", - "outputId": "68eab4d2-76bf-44ce-8d49-b0137681fcc2" + "outputId": "3ab4d2f0-2378-46f7-a433-866d58f1ecbe" }, - "source": [ - "from pycaret.datasets import get_data\n", - "from pycaret.regression import *\n", - "\n", - "insuranceDataSet = get_data(\"insurance\", verbose=False)\n", - "setup(data = insuranceDataSet, target='charges', train_size = 0.7, data_split_shuffle = False, verbose=False)\n", - "cm = compare_models()" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -7301,253 +3396,253 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
gbrGradient Boosting Regressor2376.374620007211.37694405.71230.85550.40300.26530.1750
lightgbmLight Gradient Boosting Machine2684.864622189126.67774653.62940.84060.54120.33800.9260
rfRandom Forest Regressor2556.680922361151.05234668.07640.83950.44670.30740.4730
etExtra Trees Regressor2556.612425289921.91934952.64250.81840.45390.29930.3270
adaAdaBoost Regressor3973.235625790726.93015049.35170.81760.61470.69240.1130
xgboostExtreme Gradient Boosting2884.808426410989.40005095.56080.80910.50780.35410.1480
ridgeRidge Regression4044.460135134281.50405908.34340.74890.59820.40970.0960
brBayesian Ridge4037.856835134047.17405908.18780.74880.61650.40860.0970
llarLasso Least Angle Regression4031.936935132513.67785907.95050.74880.62270.40770.1570
larLeast Angle Regression4031.678735132685.40745907.94260.74880.63160.40760.1100
lassoLasso Regression4031.939735132519.86135907.95100.74880.62270.40770.0980
lrLinear Regression4031.678735132685.40745907.94260.74880.63160.40760.1150
dtDecision Tree Regressor2958.805240573378.91326335.33220.70870.50560.34260.0970
huberHuber Regressor3261.938245115822.57576677.87140.67160.43620.21130.1290
parPassive Aggressive Regressor3876.876258873475.68277627.14600.59140.44570.23380.0980
enElastic Net7132.954587975502.11319347.47800.38850.71830.88590.0930
ompOrthogonal Matching Pursuit8800.5636129809259.541511348.83020.09950.85991.10470.0990
knnK Neighbors Regressor7882.2624132560132.000011494.39840.06840.84380.89340.1030
dummyDummy Regressor8968.3332144843431.200011998.1181-0.00651.00571.53550.0920etExtra Trees Regressor2.081610.22893.06220.87630.13200.10000.2090
gbrGradient Boosting Regressor2.176210.78343.10000.87140.14680.11070.1880
lightgbmLight Gradient Boosting Machine2.262011.93113.32380.85580.14580.11060.2170
xgboostExtreme Gradient Boosting2.350513.14043.48710.84160.14820.11260.2750
rfRandom Forest Regressor2.339013.61873.48960.83640.15470.11540.3140
adaAdaBoost Regressor2.796414.79933.72590.82000.17650.14680.1250
brBayesian Ridge3.133820.41174.39350.74890.23550.16010.0540
ridgeRidge Regression3.149620.44774.39790.74830.23670.16110.0290
larLeast Angle Regression3.155020.46564.40000.74800.23710.16140.0440
lrLinear Regression3.155020.46564.40000.74800.23710.16140.0260
huberHuber Regressor3.001520.78394.39550.74550.23600.15240.0370
knnK Neighbors Regressor2.980521.58234.47940.73890.17970.13610.0320
lassoLasso Regression3.481124.58654.88080.69720.24660.18060.0290
llarLasso Least Angle Regression3.481124.58664.88080.69720.24660.18060.0870
enElastic Net3.567326.16095.03320.67860.22900.17990.0290
dtDecision Tree Regressor3.354829.80945.15770.63490.22480.16110.0330
ompOrthogonal Matching Pursuit4.597841.48476.34630.49320.29480.22860.1040
parPassive Aggressive Regressor4.654940.77926.24220.49040.37040.26560.0500
dummyDummy Regressor6.591581.66409.0115-0.01160.38760.35980.0250
\n" @@ -7564,7 +3659,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "7e77c04e79a04111896d5933e2d0a0ca" + "model_id": "f760f6ffd48f49fab40e9beb4235130a" } }, "metadata": { @@ -7587,83 +3682,69 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv',\n", + " normalize = True,\n", + " normalize_method = 'zscore',\n", + " verbose = False)\n", + "\n", + "cm = compare_models()\n", + "\n", + "# Re-run the code again for different parameters\n", + "# normalize_method = {zscore, minmax, maxabs, robust}" ] }, { "cell_type": "markdown", "metadata": { - "id": "9GWuAHd7xQ8i" + "id": "Nf8L2_EnRyKX" }, "source": [ "---\n", - "# **2. Regression: working with user dataset**\n", - "---\n" + "####**Explore more parameters of \"setup()\" on pycaret**\n", + "---\n", + "- Explore setup() paramaeters in **Step 1.3**\n", + "- ** Click Here** for more" ] }, - { - "cell_type": "code", - "metadata": { - "id": "6ztv-AJdmhcf", - "scrolled": true - }, - "source": [ - "from pycaret.regression import *\n", - "import pandas as pd\n", - "\n", - "# First upload the user file (myData.csv) in the colab\n", - "\n", - "# myDataSet = pd.read_csv(\"myDataSet.csv\") # Uncomment and execute\n", - "# s = setup(data = myDataSet, target='target', verbose=False) # Uncomment and execute\n", - "# cm = compare_models() # Uncomment and execute" - ], - "execution_count": null, - "outputs": [] - }, { "cell_type": "markdown", "metadata": { - "id": "zUfHO4wKim1_" + "id": "oWMPU3KMim2B" }, "source": [ "---\n", - "# **3. Regression: Apply \"Data Preprocessing\"**\n", - "---\n", - "\n", - "### **3.1 Model performance using \"Normalization\"**" + "### **3.2 Visualize Feature importance**\n", + "---" ] }, { "cell_type": "code", + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "setup(data = myDataSet, target='medv', verbose = False)\n", + "\n", + "rfModel = create_model('rf', verbose = False)\n", + "\n", + "plot_model(rfModel, plot = 'feature')" + ], "metadata": { - "id": "Atd6jNvmim2A", "colab": { "base_uri": "https://localhost:8080/", - "height": 645, - "referenced_widgets": [ - "223c0947254e425aa76368d44f7813d3", - "0d52f91171f5465f90e329bcb4588a4b", - "00fb222c2cd0494e808960e3134c1a5c", - "631717151fdd41dd8b6a829265bc5e65", - "4e28d0d0fda849bdb0887bdcce676a47", - "5087cbae72274218ad25334b2551c671", - "51675431872b4184a9c5c710c39b64a0", - "3cf15a8c384e4541b9f4ae1b8313308e", - "2f198ed728bc44f2ad0a4fafee45d847", - "683564dbcd4a407185756d6d139d8525", - "64b0dafcd97b45a1b80be93fe8c9b668" - ] + "height": 485 }, - "outputId": "faef2dc2-b89e-4893-9c32-e813e6397e3d" + "id": "oojHWHpEJLjV", + "outputId": "aa888e51-33ec-470a-be5b-8dfd3dcf5a96" }, - "source": [ - "setup(data = bostonDataSet, target = 'medv',\n", - " normalize = True, normalize_method = 'zscore', data_split_shuffle = False, verbose=False)\n", - "cm = compare_models()\n", - "\n", - "# Re-run the code again for different parameters\n", - "# normalize_method = {zscore, minmax, maxabs, robust}" - ], - "execution_count": null, + "execution_count": 13, "outputs": [ { "output_type": "display_data", @@ -7679,354 +3760,49 @@ "output_type": "display_data", "data": { "text/plain": [ - "" - ], - "text/html": [ - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)
rfRandom Forest Regressor2.37749.77853.04290.71160.11910.10030.2870
etExtra Trees Regressor2.39369.99233.05750.71120.11950.10130.2000
huberHuber Regressor2.599812.32553.29540.67770.13040.10670.0330
xgboostExtreme Gradient Boosting2.544012.69613.35980.67280.12520.10370.1640
lightgbmLight Gradient Boosting Machine2.35789.28462.99560.66100.11740.09980.9350
brBayesian Ridge2.635712.42153.32210.66030.13360.10980.0270
ridgeRidge Regression2.645812.49893.33150.65670.13420.11040.0490
lrLinear Regression2.655012.57293.34020.65390.13480.11090.0300
adaAdaBoost Regressor2.614610.91213.22260.65250.12790.11280.1460
larLeast Angle Regression2.961618.17403.69760.60520.16010.12360.0470
gbrGradient Boosting Regressor2.508810.87073.21210.57890.12580.10620.2020
enElastic Net3.217719.65254.16880.51640.14900.12810.0480
lassoLasso Regression2.972515.75533.77520.49670.14500.12470.0390
llarLasso Least Angle Regression2.972515.75513.77510.49670.14500.12470.0280
ompOrthogonal Matching Pursuit3.010215.35123.77520.46040.15700.12970.0280
parPassive Aggressive Regressor3.250417.02183.96600.44570.16410.13420.0290
knnK Neighbors Regressor3.576627.06894.70860.44270.16820.13610.0340
dtDecision Tree Regressor3.173117.64394.15440.28390.16550.13690.0310
dummyDummy Regressor6.653476.80538.2838-1.30090.30040.27360.0350
\n" - ] - }, - "metadata": {} - }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "Processing: 0%| | 0/81 [00:00" + "
" ], - "text/html": [] + "image/png": "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\n" }, "metadata": {} } ] }, - { - "cell_type": "markdown", - "source": [ - "---\n", - "####**Explore more parameters of \"setup()\" on pycaret**\n", - "---\n", - "- Explore setup() paramaeters in **Step 1.3**\n", - "- ** Click Here** for more" - ], - "metadata": { - "id": "Nf8L2_EnRyKX" - } - }, { "cell_type": "markdown", "metadata": { - "id": "oWMPU3KMim2B" + "id": "5y7bTnRcxhy2" }, "source": [ "---\n", - "### **3.2 Model performance using \"Feature Selection\"**\n", + "### **3.3 Model performance using \"Feature Selection\"**\n", "---" ] }, { "cell_type": "code", + "execution_count": 14, "metadata": { - "id": "43l42fj_im2C", "colab": { "base_uri": "https://localhost:8080/", "height": 645, "referenced_widgets": [ - "3cec08d71fa949be9b554961313bb327", - "3bc1cb0053a7442f87f7cd13265ec815", - "e05d0af096014f5f833748c413c17dca", - "e1f823c302d1497aa60ee24365fdae0d", - "0c5ab9e2b5cc45b7b90d3aec1985c4b3", - "3cbc587d0d2e43e1a308d9a6242db746", - "111fffcb13a5472c88984b3020bf1e69", - "0b6eac0ee5544c97b54c39324036c326", - "12ae166387a74435a61969da3a68293b", - "37fc49a2a49d47f4a0d581cac2df7d6a", - "c6659b0659a24dec96791173d9e34555" + "03c1d2a98a7a4e2e9e7989a946dddfb9", + "61bda52f8cb64bb29fea315284fe99d8", + "cb37464c0fb8462fb5a02c2fe25f8991", + "7917db0ac4eb4ac69746459ce2ffb7ab", + "6d720d3c6e1144e69f60be32cf6cde77", + "4c6bd70d900149c893aa5dbc4551318d", + "af15af0ef42740a5b3157e0a674d5225", + "03db168b17764c81901607d2e691cdbb", + "8e408f43e2344348b473442c99d58fa3", + "61103e1036ae41ff97e85e77199d0023", + "0c051190ec684fcbb6e8bb57154da551" ] }, - "outputId": "b5c16ab2-d803-441b-f0ee-bfafc2428442" + "id": "43l42fj_im2C", + "outputId": "0e02bfb9-c4b3-4d86-faa6-2be246461548" }, - "source": [ - "bostonDataSet.columns = bostonDataSet.columns.str.replace(' ', '_')\n", - "setup(data = bostonDataSet, target = 'medv',\n", - " feature_selection = True, feature_selection_method = 'classic',\n", - " n_features_to_select = 0.2, data_split_shuffle = False, verbose=False)\n", - "cm = compare_models()\n", - "\n", - "\n", - "# Re-run the code again for different parameters\n", - "# feature_selection_method = {classic, univariate, sequential}\n", - "# n_features_to_select = {0.1, 0.2, 0.3, 0.4, 0.5, ..... }" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -8042,253 +3818,253 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
lightgbmLight Gradient Boosting Machine2.549311.15833.23380.68820.12880.10760.7570
gbrGradient Boosting Regressor2.524611.10933.19340.68420.12640.10630.1010
knnK Neighbors Regressor2.734913.50033.49930.65760.13630.11220.1020
rfRandom Forest Regressor2.603311.48773.29750.64460.13220.11070.2180
etExtra Trees Regressor2.569211.37703.27700.64180.13320.10980.1750
adaAdaBoost Regressor2.714812.18693.41290.56480.13620.11680.1060
xgboostExtreme Gradient Boosting3.022616.17393.90320.52800.15250.12490.0930
huberHuber Regressor3.016515.51893.77020.51090.16050.12760.0960
ridgeRidge Regression2.999415.11833.75310.48960.16020.12870.0500
brBayesian Ridge2.998915.10323.75220.48870.16040.12880.0490
lrLinear Regression2.998615.09663.75180.48850.16040.12880.0510
larLeast Angle Regression2.998615.09663.75180.48850.16040.12880.0500
lassoLasso Regression3.394620.07604.25110.41640.16900.13870.0510
llarLasso Least Angle Regression3.394720.07684.25120.41630.16900.13870.0510
dtDecision Tree Regressor3.175319.05964.28500.27950.16510.13320.0490
enElastic Net4.179330.71575.25860.15830.20930.16710.0500
ompOrthogonal Matching Pursuit4.958243.03686.2661-0.19990.26630.19790.0510
dummyDummy Regressor6.653476.80538.2838-1.30090.30040.27360.0480
parPassive Aggressive Regressor4.257836.77715.7052-1.38060.29920.19940.0490knnK Neighbors Regressor3.052219.59824.23920.73910.19710.15690.0310
lightgbmLight Gradient Boosting Machine3.095120.25384.33930.73580.20170.16150.4100
rfRandom Forest Regressor3.194622.30784.52910.70770.20980.17010.2070
gbrGradient Boosting Regressor3.206424.11824.68550.68950.21450.16890.1220
etExtra Trees Regressor3.301324.02264.67440.68870.21830.17380.1480
adaAdaBoost Regressor3.439824.76674.79900.68420.21660.18160.1050
xgboostExtreme Gradient Boosting3.432025.84254.88660.65080.23020.18320.1300
ridgeRidge Regression4.045931.77875.53370.56750.30300.20980.0510
brBayesian Ridge4.051431.80825.53770.56740.30230.20970.0280
larLeast Angle Regression4.044031.78205.53340.56720.30340.20990.0520
lrLinear Regression4.044031.78205.53340.56720.30340.20990.0470
huberHuber Regressor3.959832.33165.56120.56010.32180.20370.0370
lassoLasso Regression4.287634.61955.79750.54450.29380.20910.0430
llarLasso Least Angle Regression4.287634.61965.79750.54450.29380.20910.0280
dtDecision Tree Regressor3.906733.51885.65200.54210.26250.21140.0310
enElastic Net4.338335.22765.85050.53770.29510.20970.0510
ompOrthogonal Matching Pursuit4.530138.46686.11630.49780.30490.21410.0280
parPassive Aggressive Regressor5.255446.47016.73260.36070.37260.28770.0280
dummyDummy Regressor6.570582.12068.9488-0.05180.38230.35410.0270
\n" @@ -8305,7 +4081,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "3cec08d71fa949be9b554961313bb327" + "model_id": "03c1d2a98a7a4e2e9e7989a946dddfb9" } }, "metadata": { @@ -8328,6 +4104,24 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv',\n", + " feature_selection = True,\n", + " feature_selection_method = 'univariate',\n", + " n_features_to_select = 0.2,\n", + " verbose = False)\n", + "\n", + "cm = compare_models()\n", + "\n", + "\n", + "# Re-run the code again for different parameters\n", + "# feature_selection_method = {univariate, classic, sequential}\n", + "# n_features_to_select = {0.1, 0.2, 0.3, 0.4, 0.5, ..... }" ] }, { @@ -8343,38 +4137,28 @@ }, { "cell_type": "code", + "execution_count": 15, "metadata": { - "id": "oAaDhJctim2D", "colab": { "base_uri": "https://localhost:8080/", "height": 645, "referenced_widgets": [ - "c6f22d4bd1cb4c07850b7e055e617bed", - "bb316aa437a34355b21269d7086a2bdb", - "b623ded86e1f4cc2a16325cd67747075", - "a2d2b0dfe7a849bbb6bf4c7620cbf5c2", - "60c6b8c107ea458baaf26c2663999320", - "efd7895fe3de49fbbfe4e954671b2943", - "95d622af30524225a74644186d76a861", - "9ecca98650c34c69b59abc3fd733648f", - "3b3301074c1e4907a7964afc70e1218e", - "230ef89cdcaf4885bfbb3cffaadb4cc2", - "4095453c9c4f48ea81570d80060a06c4" + "b8a0909132be410ebbfc358c35eb4902", + "bb762aa5f8af405b9750faa6c9cd62b8", + "df204cbbfca949fca9333849270dd82e", + "e6e860cd71204669b404ff2736e84934", + "ad5ea8d528104237a55c7800a5a6b6fb", + "6703fc6437b14190946cf1a5ee889ecb", + "8636e2a48f8848fe8b8f8f1a7467feda", + "890b0b0a73964fb4a1fbc8bf01478b78", + "0817ac74c2044ade9e97dc5ddc87903d", + "00d968b63250485c9fddb23e7dfd8436", + "8a9da8dfd6824e9d8b65d3667b58b22b" ] }, - "outputId": "89b5e2c7-a08b-48b1-e6c6-51a5f264f989" + "id": "oAaDhJctim2D", + "outputId": "f8a565f7-33a4-4032-e7c9-46aa221e8d0e" }, - "source": [ - "setup(data = bostonDataSet, target = 'medv',\n", - " remove_outliers = True, outliers_method = \"iforest\", outliers_threshold = 0.05,\n", - " data_split_shuffle = False, verbose=False)\n", - "cm = compare_models()\n", - "\n", - "# Re-run the code again for different parameters\n", - "# outliers_threshold = {0.04, 0.05, 0.06, 0.07, 0.08, ....}\n", - "# outliers_method = {iforest, ee, lof}" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -8390,253 +4174,253 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
rfRandom Forest Regressor2.40439.95273.06820.70530.11950.10220.8690
etExtra Trees Regressor2.378110.02243.06910.70490.12020.10130.4060
lightgbmLight Gradient Boosting Machine2.482910.68343.17240.67460.12150.10340.9970
brBayesian Ridge2.679912.97303.37060.65610.13650.11190.2470
ridgeRidge Regression2.680812.97843.37140.65550.13650.11200.2450
lrLinear Regression2.705713.13993.39260.65280.13800.11360.3300
gbrGradient Boosting Regressor2.515310.88783.22240.63940.12500.10670.4570
xgboostExtreme Gradient Boosting2.544911.36423.28650.62270.12550.10690.3720
larLeast Angle Regression2.749713.38613.49100.60860.15440.11950.2570
adaAdaBoost Regressor2.811112.66113.46790.55760.13650.12150.3210
lassoLasso Regression3.051018.51664.06300.55300.15770.11910.2660
llarLasso Least Angle Regression3.051018.51604.06300.55300.15770.11910.3360
huberHuber Regressor3.411122.91524.42090.46080.16680.12990.3920
enElastic Net3.618525.05754.76640.35330.20290.14240.2430
dtDecision Tree Regressor3.274718.97644.28810.18760.16380.14150.2460
ompOrthogonal Matching Pursuit6.290371.58608.0214-0.91570.28740.25480.2420
parPassive Aggressive Regressor7.189775.04948.2005-1.11340.31870.31060.2420
dummyDummy Regressor6.694177.33338.3189-1.32910.30180.27590.2420
knnK Neighbors Regressor7.095294.77319.3746-3.05530.33230.29200.2510gbrGradient Boosting Regressor2.262610.43823.17440.88070.14620.11390.2880
etExtra Trees Regressor2.243210.56793.18350.87570.14050.11190.5410
rfRandom Forest Regressor2.340412.22383.38190.86310.14840.11630.4600
lightgbmLight Gradient Boosting Machine2.560114.76693.70100.83510.15750.12600.6240
xgboostExtreme Gradient Boosting2.590315.27793.79740.82950.15930.12480.3090
adaAdaBoost Regressor2.893715.50633.89990.82220.17860.15140.2870
lrLinear Regression3.643227.41405.03560.71210.26010.18050.1780
ridgeRidge Regression3.645527.75045.06230.70920.25910.18130.2990
larLeast Angle Regression3.666628.36185.10630.70430.27480.18260.1780
brBayesian Ridge3.667628.17235.11440.70400.25920.18130.1780
dtDecision Tree Regressor3.306328.04395.11410.69070.20990.15970.1780
lassoLasso Regression3.899432.08545.54240.65660.27850.18440.1780
enElastic Net3.923432.01645.54070.65660.27810.18580.2880
llarLasso Least Angle Regression3.899432.08545.54240.65660.27850.18440.1780
huberHuber Regressor3.857134.80255.69110.62830.31160.18900.3700
knnK Neighbors Regressor5.007550.60647.02420.43610.26950.23250.1810
ompOrthogonal Matching Pursuit6.398878.15268.71390.15850.34200.30510.1760
dummyDummy Regressor7.197996.68519.7065-0.03320.40340.38440.1730
parPassive Aggressive Regressor8.1312130.817810.5508-0.37160.48650.43740.2320
\n" @@ -8653,7 +4437,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "c6f22d4bd1cb4c07850b7e055e617bed" + "model_id": "b8a0909132be410ebbfc358c35eb4902" } }, "metadata": { @@ -8676,6 +4460,23 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv',\n", + " remove_outliers = True,\n", + " outliers_method = \"iforest\",\n", + " outliers_threshold = 0.05,\n", + " verbose = False)\n", + "\n", + "cm = compare_models()\n", + "\n", + "# Re-run the code again for different parameters\n", + "# outliers_threshold = {0.04, 0.05, 0.06, 0.07, 0.08, ....}\n", + "# outliers_method = {iforest, ee, lof}" ] }, { @@ -8691,34 +4492,28 @@ }, { "cell_type": "code", + "execution_count": 16, "metadata": { - "id": "TAe7byNYim2D", "colab": { "base_uri": "https://localhost:8080/", - "height": 613, + "height": 645, "referenced_widgets": [ - "b7a4f5905e504d41ba1124f2c7b363c2", - "70fc58f2565a4401adf41e2ae8d75a89", - "e46d1b1d43a44bbc8945bd9c4ec32519", - "2077179edf684d0783429eedb24cbea6", - "e2e49676f4f34815af11e380bcb6b27f", - "ece5244e61eb4747bb8375e322c55219", - "4bcc5940e7b649f1a9883251ee1f95e8", - "53ee9ed83a124e7fb82b6f49d461f116", - "cf7aaacff5aa4eb3bae32a3a25fdab49", - "97c8f63078834324b8a80968e900008c", - "850e2a8c0e5342b2a63503fc83cacb19" + "70d608d0f5eb40f9945537575a23b2e9", + "fb151c4abca545dfb576d3988cf56e0f", + "261c920b722c42daaeb12cffa396cc9b", + "8127f0ed020a47218327803e378a2005", + "3e6a2db4fd9b44ae8351d374f5dec5a9", + "0601a3df8100442dae45799c1165fa25", + "6410fbd59bfd4dccb63a54c4147b9dab", + "3b89da1e87644c229cad67a623532d16", + "d72c8d98234940fa8dee835b84fb7ae8", + "cc3cf88a0bf7402aa593f1c7ff52fc46", + "2cc5866bb8fc46fc8d2838932cf6daf2" ] }, - "outputId": "981e2697-2468-44cf-dfcf-fdd9ad09e0bd" + "id": "TAe7byNYim2D", + "outputId": "2dae7264-adbc-4265-b5b5-fb3392004341" }, - "source": [ - "setup(data = bostonDataSet, target = 'medv',\n", - " transformation = True, transformation_method = 'yeo-johnson',\n", - " data_split_shuffle = False, verbose=False)\n", - "cm = compare_models()" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -8734,242 +4529,253 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
lightgbmLight Gradient Boosting Machine2.28238.96222.92880.68080.11380.09540.7610
rfRandom Forest Regressor1.99537.93902.60350.63660.10230.08500.2840
etExtra Trees Regressor2.06408.27792.64790.61860.10410.08760.3320
xgboostExtreme Gradient Boosting2.13068.98472.75060.60510.10670.08900.1700
gbrGradient Boosting Regressor2.16198.95712.77610.49920.10940.09270.1370
adaAdaBoost Regressor2.35869.93342.92430.46610.11690.10250.1300
larLeast Angle Regression3.553722.43404.45740.43090.19810.14910.0560
ridgeRidge Regression3.759223.63904.68980.32640.18440.15470.0900
dtDecision Tree Regressor2.849415.64913.71470.19380.14810.12710.0530
lassoLasso Regression4.450636.39105.66170.14590.20130.18030.0590
llarLasso Least Angle Regression4.450636.39105.66170.14590.20130.18030.0510
enElastic Net5.115949.00316.6418-0.22420.23490.20720.0890
dummyDummy Regressor6.653476.80538.2838-1.30090.30040.27360.0480
lrLinear Regression6.690976.61518.3026-1.33380.30040.27450.0520
ompOrthogonal Matching Pursuit6.690976.61518.3026-1.33380.30040.27450.0500
knnK Neighbors Regressor7.051786.33169.0013-1.80100.32100.28320.0580
parPassive Aggressive Regressor10.1554165.101311.9496-5.24180.67930.43480.0500
brBayesian Ridge98.439711065.8698101.8193-384.06131.40314.25230.0520etExtra Trees Regressor2.237410.73523.20950.87310.14030.11250.2290
gbrGradient Boosting Regressor2.277010.08863.10430.87300.14450.11720.2860
rfRandom Forest Regressor2.307410.99243.24510.86470.14920.11920.3480
lightgbmLight Gradient Boosting Machine2.427613.17793.50890.84370.15100.11990.1120
xgboostExtreme Gradient Boosting2.481514.14323.62980.82410.15350.12270.3190
adaAdaBoost Regressor2.931015.77683.93500.80880.18300.15710.1470
lrLinear Regression3.546923.36674.78210.72740.22580.18160.0530
larLeast Angle Regression3.612524.17584.85750.71860.23300.18610.0650
dtDecision Tree Regressor3.116923.63304.63270.71860.19600.15170.0570
brBayesian Ridge3.637124.59784.89740.71500.23360.18530.0870
ridgeRidge Regression3.641524.68594.90660.71370.23290.18510.0580
lassoLasso Regression3.871130.75175.43270.65310.21980.18620.0580
llarLasso Least Angle Regression3.871130.75175.43270.65310.21980.18620.0810
enElastic Net4.282440.39046.20260.54940.24870.21120.0580
knnK Neighbors Regressor6.366377.58748.74100.09720.34610.31030.0610
ompOrthogonal Matching Pursuit6.384779.87238.82310.09390.36880.33950.0890
dummyDummy Regressor6.837589.01129.3412-0.02040.40060.37720.0550
huberHuber Regressor7.4759105.607010.2141-0.23590.80080.38820.1110
parPassive Aggressive Regressor8.3823132.993811.3777-0.51400.50500.40460.0990
\n" @@ -8986,7 +4792,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "b7a4f5905e504d41ba1124f2c7b363c2" + "model_id": "70d608d0f5eb40f9945537575a23b2e9" } }, "metadata": { @@ -9009,6 +4815,20 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv',\n", + " transformation = True,\n", + " transformation_method = 'yeo-johnson',\n", + " verbose = False)\n", + "\n", + "cm = compare_models()\n", + "\n", + "# transformation_method = {yeo-johnson, quantile}" ] }, { @@ -9024,36 +4844,28 @@ }, { "cell_type": "code", + "execution_count": 17, "metadata": { - "id": "IbRChxn3im2E", "colab": { "base_uri": "https://localhost:8080/", "height": 645, "referenced_widgets": [ - "ce49e4cd8b90424baa289827bbdf5709", - "1cb1af85286c47d3b182d29959ea8588", - "7430bcd44a5248a492d03f902662c548", - "4e21b1a0351341a9809ca5b4d05fd15a", - "4c21f0f814564cf5a160db87801c2855", - "57daa23de91a4472aba05a561b010c51", - "c682dcc24a674038b23bd9624cf905bb", - "0992dbe948894830a1e996a6a3b4edb9", - "93d7b7f241104318b52e69a17c45c3cb", - "a8385f7b9c0e4cbb92a35ffd6c5a7621", - "f5fe066b5bc744379eeb3730a77e8b41" + "a02e5086a0c04562b5124db8ace4878d", + "59b71c97f3ad41b8a7fdb177ab7c656f", + "be330aeb9e0b4f05886b8739f7807704", + "7b5993918f5a48bd9a35ac2afe7f9e3d", + "d8ff5bbe4d4c4acba1189ef61bfaa8f3", + "d87facdcb086437db9bca8af485586a9", + "b78adebb92cb4c5e842b7109be7c41a2", + "8fd6060bf89b4bfa8dc49c20bc788a15", + "491a6871af0e485abf80b012188046be", + "68e670bdc9514df694c2e03ae1617383", + "0448d50aff8441418e9b73daedd5af9e" ] }, - "outputId": "3e1ddcf2-9806-4e60-8757-38689cbd1d85" + "id": "IbRChxn3im2E", + "outputId": "43569e84-ab8b-4bc6-82a2-f804d2c0f988" }, - "source": [ - "setup(data = bostonDataSet, target = 'medv',\n", - " pca = True, pca_method = 'linear', data_split_shuffle = False, verbose=False)\n", - "cm = compare_models()\n", - "\n", - "# Re-run the code again for different parameters\n", - "# pca_method = (linear, kernel, incremental)" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -9069,253 +4881,253 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
lrLinear Regression2.655012.57293.34020.65390.13480.11090.0300
larLeast Angle Regression2.655012.57293.34020.65390.13480.11090.0300
ridgeRidge Regression2.634612.47863.32520.65220.13390.10970.0270
brBayesian Ridge2.635312.48043.32580.65210.13400.10970.0290
huberHuber Regressor2.809615.61793.67150.62530.14600.10990.0520
lassoLasso Regression3.416622.54944.52670.43220.18250.13380.0270
llarLasso Least Angle Regression3.416622.54944.52670.43220.18250.13380.0280
enElastic Net3.636225.21754.80500.33560.20950.14410.0270
etExtra Trees Regressor3.598924.55554.65800.28400.17330.15050.2020
adaAdaBoost Regressor3.895526.00124.83980.21820.18290.16750.1180
lightgbmLight Gradient Boosting Machine3.357522.54074.49890.15310.16630.14070.7270
rfRandom Forest Regressor3.872529.87615.13810.09390.18990.16020.3830
xgboostExtreme Gradient Boosting3.802833.01135.3626-0.03560.19480.15430.3960
gbrGradient Boosting Regressor3.394223.09764.5714-0.04000.16860.14240.1570
ompOrthogonal Matching Pursuit6.202869.90267.9395-0.85850.28330.25000.0280
dummyDummy Regressor6.653476.80538.2838-1.30090.30040.27360.0260
dtDecision Tree Regressor5.005261.96847.3297-2.62140.25220.20170.0560
parPassive Aggressive Regressor4.921450.19425.9490-2.85150.31500.23590.0310
knnK Neighbors Regressor7.128095.62719.4698-2.97820.33490.29170.0470etExtra Trees Regressor2.504413.18893.55590.84320.15570.12450.2140
xgboostExtreme Gradient Boosting2.727714.89943.73120.81880.16300.13390.5350
gbrGradient Boosting Regressor2.680315.16343.76120.81800.16340.13290.2000
rfRandom Forest Regressor2.778016.23893.93680.80550.16370.13420.3770
lightgbmLight Gradient Boosting Machine2.845216.37793.94670.80140.17450.14250.1220
adaAdaBoost Regressor3.384020.59504.48840.75540.20630.18130.1350
dtDecision Tree Regressor3.660725.29564.97250.69950.21590.17400.0540
lrLinear Regression3.548725.50174.95100.69760.26390.17090.0290
larLeast Angle Regression3.548725.50174.95100.69760.26390.17090.0300
ridgeRidge Regression3.525425.72934.96130.69440.25550.17080.0280
brBayesian Ridge3.584326.39245.03550.68550.25740.17320.0390
huberHuber Regressor3.554627.91555.20620.66600.24790.16270.0950
enElastic Net3.735928.72885.31240.65480.26680.17400.0310
lassoLasso Regression3.803829.89425.42470.64050.26900.17610.0310
llarLasso Least Angle Regression3.803829.89425.42470.64050.26900.17610.0320
knnK Neighbors Regressor4.365040.16736.22590.53620.24290.20430.0570
ompOrthogonal Matching Pursuit5.860368.18508.21140.18930.31450.27360.0290
dummyDummy Regressor6.704385.64279.2179-0.01730.37730.34550.0260
parPassive Aggressive Regressor21.0743926.804029.0465-11.03920.93901.23900.0480
\n" @@ -9332,7 +5144,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "ce49e4cd8b90424baa289827bbdf5709" + "model_id": "a02e5086a0c04562b5124db8ace4878d" } }, "metadata": { @@ -9355,6 +5167,21 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv',\n", + " pca = True,\n", + " pca_method = 'linear',\n", + " verbose = False)\n", + "\n", + "cm = compare_models()\n", + "\n", + "# Re-run the code again for different parameters\n", + "# pca_method = (linear, kernel, incremental)" ] }, { @@ -9370,34 +5197,28 @@ }, { "cell_type": "code", + "execution_count": 18, "metadata": { - "id": "9XFmW3nyAFM2", "colab": { "base_uri": "https://localhost:8080/", "height": 645, "referenced_widgets": [ - "00202b43d63b4a56bceb8c3cffcede0b", - "d796a62044644be4a181bbd58ba7fdac", - "15448bdc87934e7c95a72facc3801566", - "c0932334fe134f78a373bee51bd193df", - "a80999291470457ba7a1eb77893f01e4", - "fd51ef1e19a24261b2f9616c234fa7c8", - "33675eb92a284b8a95a677aa098b4f27", - "a0824841b5b54a56a31aaf115822779d", - "e818471e3a8c4cc391131328ea4e89bd", - "d9c42bab2447423b83ab296e467c503a", - "522745e40ca2425fbee6235a0761f8a0" + "a96cb5609fc347268caf7ce4ba6413b7", + "994cc6b76d5a478189b9170b50d904a0", + "05f1c3f32e8d4dedb69e4c1d86990784", + "ee6261b47cd64629ad09b86f2debc2bc", + "b1d88c1e708145748c96ddc527c3b49d", + "9cedeb10db9d4ad9acb9111b101aab54", + "bc3661bb7cf54a33b8e040d986229b6d", + "97a0533478cd47dd8daf000d473af2d5", + "43c559112ce0406e91f8e7fa9e24f2ac", + "d4cb48d9315c4f2c995420e870a7c5ae", + "9fd77786dcac412d8b8775ed08340670" ] }, - "outputId": "b495bd11-ad74-4cae-eaa6-b7eebb21f675" + "id": "9XFmW3nyAFM2", + "outputId": "8aa8269d-9735-4434-f1cb-044a12693e33" }, - "source": [ - "setup(data = bostonDataSet, target = 'medv',\n", - " remove_outliers = True, outliers_threshold = 0.05,\n", - " normalize = True, normalize_method = 'zscore', data_split_shuffle = False, verbose=False)\n", - "cm = compare_models()" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -9413,253 +5234,253 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
rfRandom Forest Regressor2.452610.42573.13750.67990.12260.10470.6220
etExtra Trees Regressor2.449810.77023.17020.67910.12330.10380.4160
huberHuber Regressor2.660813.21623.36350.67570.13550.10980.2540
lightgbmLight Gradient Boosting Machine2.548511.06113.23470.65940.12550.10830.9570
brBayesian Ridge2.744313.72623.43170.65730.14140.11510.2500
ridgeRidge Regression2.756213.85513.44510.65320.14230.11580.2530
lrLinear Regression2.765213.96083.45600.65020.14300.11630.2940
larLeast Angle Regression2.916216.55003.62400.62770.15630.12290.3450
xgboostExtreme Gradient Boosting2.777914.86153.63970.57570.13590.11560.3810
gbrGradient Boosting Regressor2.578511.47433.31040.56220.13010.11060.4240
adaAdaBoost Regressor2.769212.54993.45080.56110.13700.12010.3510
dtDecision Tree Regressor2.970015.87103.87150.54090.15250.12900.2530
enElastic Net3.215019.71384.16720.52020.14900.12780.2650
lassoLasso Regression2.967715.85343.77940.49740.14430.12420.2520
llarLasso Least Angle Regression2.967715.85323.77930.49740.14430.12420.2490
ompOrthogonal Matching Pursuit3.021315.47113.78130.46370.15700.12980.2480
knnK Neighbors Regressor3.704229.52374.90230.37270.17630.14220.2560
parPassive Aggressive Regressor4.770939.58615.8589-0.60990.27570.20720.3730
dummyDummy Regressor6.700577.16528.3198-1.35170.30180.27660.2470huberHuber Regressor2.789914.42083.79740.75020.14830.11540.3300
brBayesian Ridge2.779914.69913.81470.73560.15750.11980.3150
ompOrthogonal Matching Pursuit2.979615.39703.92390.73380.16220.12870.2700
ridgeRidge Regression2.790214.81303.82550.73250.15880.12090.1800
etExtra Trees Regressor3.003515.51083.93640.72860.15620.13390.3600
lrLinear Regression2.808415.03993.85140.72760.16030.12210.1850
larLeast Angle Regression2.808415.03993.85140.72760.16030.12210.1950
llarLasso Least Angle Regression3.121717.82214.21460.69970.15960.13070.1800
lassoLasso Regression3.121717.82194.21460.69970.15960.13070.1900
lightgbmLight Gradient Boosting Machine3.209118.57004.26560.69830.15930.13470.2100
gbrGradient Boosting Regressor3.154119.07234.35710.68010.16300.13400.2650
rfRandom Forest Regressor3.337419.53744.41870.66630.17180.14670.4000
enElastic Net3.267121.44384.56540.65540.15710.12640.2000
adaAdaBoost Regressor3.459321.11274.58350.64620.17630.15090.2900
xgboostExtreme Gradient Boosting3.409821.54174.63940.62370.17370.14700.2700
dtDecision Tree Regressor4.237933.90275.75240.45230.20580.17550.3300
parPassive Aggressive Regressor4.438638.66196.07700.27450.34380.20500.3100
knnK Neighbors Regressor4.953845.63636.73080.24020.24000.20330.3200
dummyDummy Regressor7.9109101.045010.0313-0.69750.36400.33030.1900
\n" @@ -9676,7 +5497,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "00202b43d63b4a56bceb8c3cffcede0b" + "model_id": "a96cb5609fc347268caf7ce4ba6413b7" } }, "metadata": { @@ -9699,6 +5520,25 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv',\n", + " remove_outliers = True,\n", + " outliers_threshold = 0.05,\n", + " normalize = True,\n", + " normalize_method = 'zscore',\n", + " data_split_shuffle = False,\n", + " fold = 2,\n", + " verbose = False)\n", + "\n", + "cm = compare_models()\n", + "\n", + "\n", + "# Question: Does order of the data preprocessing matters ???" ] }, { @@ -9714,36 +5554,28 @@ }, { "cell_type": "code", + "execution_count": 19, "metadata": { - "id": "qoSbD33ktEvw", "colab": { "base_uri": "https://localhost:8080/", "height": 645, "referenced_widgets": [ - "e0ea5230d0384b7482aa787a747de36a", - "e26cb115059243898cd52324cdc9e62e", - "c0cedb777e244c7db748e075f9cd52eb", - "5c5fa00457d8410fb64fa5228a7bc4fa", - "49cc270350234c2e865223b7605c6906", - "7b2797c8874444c2bd2eaf18ba46e2cd", - "4b0a8272d56d46acb6f96af5b8c526e0", - "2443b48bb035438584749ad7912fb06e", - "b106ed6bcaed4c07af5c646ff5ae2935", - "02e851d1d6a54ff69fa898f1ee20c220", - "d5d2fc5df0b4433e9f6e54a9f288f129" + "ab40f21c43a143f5a7d05e8440af6e7a", + "aff0e209e73b43ac85e1b65afc0cf106", + "acad53b6be224f64adfbdf623457e9ef", + "ab34d8b8b5f4424a93aa28b94fcd7e3c", + "8d23f1397abd4a9696cf33ebcfb26b1a", + "f5acbdebc161496ea348e9fe48447951", + "30852c51f74044458e4cf5fc84da6d76", + "bbf1b7158eb749cfb058e1a5a09e5c2b", + "7257cadaf01b490aa4b28143e1f4cd0e", + "f06f49e680e7429ea07d0317410484aa", + "c47016c75b3d4f2d854659b51eb8ba6e" ] }, - "outputId": "372869b4-e8c9-4c4a-e3c8-9752077d3738" + "id": "qoSbD33ktEvw", + "outputId": "dab78fa3-6a9c-4e82-beb7-9f83ebdf9e5d" }, - "source": [ - "setup(data = bostonDataSet, target = 'medv',\n", - " remove_outliers = True, outliers_threshold = 0.05,\n", - " normalize = True, normalize_method = 'zscore',\n", - " transformation = True, transformation_method = 'yeo-johnson',\n", - " data_split_shuffle = False, verbose=False)\n", - "cm = compare_models()" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -9759,253 +5591,253 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)ModelMAEMSERMSER2RMSLEMAPETT (Sec)
etExtra Trees Regressor2.428810.43943.11490.70470.12220.10360.4390
rfRandom Forest Regressor2.410610.15163.10240.69310.12080.10250.5560
lightgbmLight Gradient Boosting Machine2.460510.46663.15620.65660.12170.10370.9950
gbrGradient Boosting Regressor2.466610.46933.14860.64250.12310.10490.3680
xgboostExtreme Gradient Boosting2.628712.66753.41970.63880.12960.10960.4070
huberHuber Regressor2.961916.40143.82970.57140.16240.11910.2870
brBayesian Ridge3.009715.95143.82610.55070.17950.12500.3140
dtDecision Tree Regressor3.068116.68343.95530.54600.15530.13250.3530
ridgeRidge Regression3.027116.06193.84050.54540.18230.12590.2760
larLeast Angle Regression3.033816.10503.84590.54380.18310.12620.3660
lrLinear Regression3.033816.10503.84590.54380.18310.12620.3820
enElastic Net3.401322.70904.45620.45540.15870.13380.3000
adaAdaBoost Regressor2.814513.29573.53130.44620.13830.12230.4830
knnK Neighbors Regressor3.516923.54604.52840.44040.16550.13910.2860
lassoLasso Regression3.407922.30454.45700.38910.18670.13880.2810
llarLasso Least Angle Regression3.407922.30454.45700.38910.18670.13880.2790
parPassive Aggressive Regressor4.032730.47445.19950.20690.22330.16130.3430
ompOrthogonal Matching Pursuit3.807326.96964.91910.11050.26730.16630.2780
dummyDummy Regressor6.709176.99198.3150-1.35570.30170.27740.3080etExtra Trees Regressor3.029915.66533.95510.72520.15880.13590.3450
brBayesian Ridge3.124918.28134.27290.68910.21760.12920.2000
rfRandom Forest Regressor3.273518.40824.28930.68520.16850.14450.4050
ridgeRidge Regression3.159118.34624.28310.68450.22420.13190.3500
lrLinear Regression3.182518.44934.29530.68140.22820.13340.3450
larLeast Angle Regression3.182518.44934.29530.68140.22820.13340.3850
gbrGradient Boosting Regressor3.240519.45324.39280.67670.16740.13940.2900
lightgbmLight Gradient Boosting Machine3.295020.33234.43740.67470.16340.13690.3350
huberHuber Regressor3.306121.98494.63790.64360.21120.12940.2300
adaAdaBoost Regressor3.584322.59744.74350.62050.18260.15730.2950
enElastic Net3.647227.44175.07280.57430.17270.13880.3750
xgboostExtreme Gradient Boosting3.680925.93775.09010.54590.18670.15800.2950
lassoLasso Regression3.862728.66585.28200.53860.20650.15500.3450
llarLasso Least Angle Regression3.862628.66515.28190.53860.20650.15500.2650
dtDecision Tree Regressor4.273732.20035.66390.45830.21770.19050.2100
parPassive Aggressive Regressor4.417934.30135.85140.41840.22130.18850.2050
knnK Neighbors Regressor4.320638.06046.00730.40410.21050.16790.2350
ompOrthogonal Matching Pursuit4.499737.93756.04800.39610.29050.18230.2300
dummyDummy Regressor7.9937102.808710.1163-0.72480.36770.33330.3600
\n" @@ -10022,7 +5854,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "e0ea5230d0384b7482aa787a747de36a" + "model_id": "ab40f21c43a143f5a7d05e8440af6e7a" } }, "metadata": { @@ -10045,6 +5877,24 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv',\n", + " remove_outliers = True,\n", + " outliers_threshold = 0.05,\n", + " normalize = True,\n", + " normalize_method = 'zscore',\n", + " transformation = True,\n", + " transformation_method = 'yeo-johnson',\n", + " data_split_shuffle = False,\n", + " fold = 2,\n", + " verbose = False)\n", + "\n", + "cm = compare_models()" ] }, { @@ -10057,7 +5907,7 @@ "### **3.8 Explore more parameters of \"setup()\" on pycaret**\n", "---\n", "- Explore setup() paramaeters in **Step 1.3**\n", - "- ** Click Here** for more" + "- ** Click Here** for more (Right Click on the link and click on \"Open Link in New Window\")" ] }, { @@ -10074,38 +5924,28 @@ }, { "cell_type": "code", + "execution_count": 20, "metadata": { - "id": "DxkGGGthim2G", "colab": { "base_uri": "https://localhost:8080/", "height": 457, "referenced_widgets": [ - "4bc7880483ef46f386959c1c9c23583c", - "a24ac38b47614b47ae89bf813d50ea78", - "caf546ef00f74a87a0b25d0e94de523c", - "1a67db658427407391b7601cb6e482bc", - "7606dd82dabe4348b1232ddf9e92108d", - "2e079d47d0d84613a1d965acab007297", - "4b0af501fe7444a1aa15f8e110528df7", - "3ad24a154e454af3bd7096a4153dfc30", - "a715a71eae764c958f3738734be0db32", - "6b065bff14224bdda8e056048fd93c42", - "797b4411f13743e0a2b77b87005b4882" + "e0437a7860284a098058a7d74449b7ef", + "7217b03b1eb7436f9090367b382527a7", + "95d977914c834115aa54236a21ff55ed", + "ccf5278f491244febde543e236e86394", + "cf7a6161741a4f44ae9c2d54ab9b4ec7", + "b9f4671df39049ab9d9da1b1e1bce2d5", + "ea5898e313814cc0a96d0c475528c575", + "cad0919c4d2d4c38aec25bea355bb5a0", + "1908d641a3fc4740aadf819d71c2a29e", + "16c9e98ea5d14d24bc57de44f26960d5", + "c023d012dd8a4b54b7433d7e0f13bf78" ] }, - "outputId": "f98cc5a7-af0c-45a8-e47d-345d3a4ce80f" + "id": "DxkGGGthim2G", + "outputId": "182a1958-bd34-4940-87b4-c0708b2bf263" }, - "source": [ - "from pycaret.datasets import get_data\n", - "from pycaret.regression import *\n", - "\n", - "bostonDataSet = get_data(\"boston\", verbose=False) # SN is 46\n", - "setup(data = bostonDataSet, target='medv', verbose=False)\n", - "\n", - "rfModel = create_model('rf')\n", - "# Explore more parameters" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -10121,24 +5961,24 @@ "output_type": "display_data", "data": { "text/plain": [ - "" + "" ], "text/html": [ "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", @@ -10152,112 +5992,112 @@ " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 MAEMSERMSER2RMSLEMAPEMAEMSERMSER2RMSLEMAPE
Fold
03.169932.88885.73490.70030.17980.1236
12.511410.09583.17740.90090.18610.1579
22.18539.62983.10320.89540.13430.0991
32.08536.82582.61260.92580.12940.1099
43.157839.20496.26140.51950.20880.1425
52.408110.43093.22970.75400.18080.1491
61.57443.76661.94080.90790.12360.0978
72.170810.94023.30760.93260.11380.0913
82.673116.00004.00000.73320.17800.1364
92.322213.48233.67180.85660.18350.1325
Mean2.425815.32653.70390.81260.16180.1240
Std0.461810.91501.26780.12700.03130.022202.141210.44313.23160.84790.15720.1166
12.705519.39334.40380.64420.17520.1359
22.462412.16343.48760.87560.13110.1090
32.549213.01723.60790.87780.14580.1194
42.12749.92923.15110.85560.12490.0973
52.269610.95883.31040.79820.16110.1280
61.96106.54852.55900.92180.09960.0857
72.28568.53342.92120.82480.18190.1560
82.363510.14173.18460.91060.14920.1203
92.359810.32703.21360.89160.15220.1236
Mean2.322511.14563.30710.84480.14780.1192
Std0.20653.22810.45700.07580.02310.0186
\n" @@ -10274,7 +6114,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "4bc7880483ef46f386959c1c9c23583c" + "model_id": "e0437a7860284a098058a7d74449b7ef" } }, "metadata": { @@ -10297,6 +6137,16 @@ }, "metadata": {} } + ], + "source": [ + "from pycaret.datasets import get_data\n", + "from pycaret.regression import *\n", + "myDataSet = get_data(\"boston\", verbose = False)\n", + "\n", + "setup(data = myDataSet, target = 'medv', verbose = False)\n", + "\n", + "rfModel = create_model('rf')\n", + "# Explore more parameters" ] }, { @@ -10355,19 +6205,15 @@ }, { "cell_type": "code", + "execution_count": 21, "metadata": { "id": "ukwXEWlAim2O", "colab": { "base_uri": "https://localhost:8080/", "height": 206 }, - "outputId": "a20aeecb-5c9c-4a42-91af-8421424cb242" + "outputId": "508180e1-6fd2-42f9-c58b-78b5e0bf599b" }, - "source": [ - "# Select top 10 rows from boston dataset\n", - "newDataSet = get_data(\"boston\").iloc[:10]" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -10389,7 +6235,7 @@ ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPEModelMAEMSERMSER2RMSLEMAPE
0Random Forest Regressor1.32462.58401.60750.93770.06470.05480Random Forest Regressor1.45502.81811.67870.93210.05800.0535
\n" @@ -10809,20 +6659,20 @@ "9 0.17004 12.5 7.87 0 0.524 6.004 85.900002 6.5921 5 311 \n", "\n", " ptratio black lstat medv prediction_label \n", - "0 15.300000 396.899994 4.98 24.000000 25.272000 \n", - "1 17.799999 396.899994 9.14 21.600000 21.724000 \n", - "2 17.799999 392.829987 4.03 34.700001 34.843000 \n", - "3 18.700001 394.630005 2.94 33.400002 35.128001 \n", - "4 18.700001 396.899994 5.33 36.200001 35.689001 \n", - "5 18.700001 394.119995 5.21 28.700001 26.582001 \n", - "6 15.200000 395.600006 12.43 22.900000 22.066000 \n", - "7 15.200000 396.899994 19.15 27.100000 23.936000 \n", - "8 15.200000 386.630005 29.93 16.500000 18.430000 \n", - "9 15.200000 386.709991 17.10 18.900000 20.322000 " + "0 15.300000 396.899994 4.98 24.000000 24.294000 \n", + "1 17.799999 396.899994 9.14 21.600000 23.302000 \n", + "2 17.799999 392.829987 4.03 34.700001 36.013001 \n", + "3 18.700001 394.630005 2.94 33.400002 35.302001 \n", + "4 18.700001 396.899994 5.33 36.200001 33.330000 \n", + "5 18.700001 394.119995 5.21 28.700001 26.951001 \n", + "6 15.200000 395.600006 12.43 22.900000 22.310000 \n", + "7 15.200000 396.899994 19.15 27.100000 24.443000 \n", + "8 15.200000 386.630005 29.93 16.500000 17.420000 \n", + "9 15.200000 386.709991 17.10 18.900000 19.453000 " ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", @@ -11625,112 +7445,112 @@ " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 MAEMSERMSER2RMSLEMAPEMAEMSERMSER2RMSLEMAPE
Fold
03.169932.88885.73490.70030.17980.1236
12.511410.09583.17740.90090.18610.1579
22.18539.62983.10320.89540.13430.0991
32.08536.82582.61260.92580.12940.1099
43.157839.20496.26140.51950.20880.1425
52.408110.43093.22970.75400.18080.1491
61.57443.76661.94080.90790.12360.0978
72.170810.94023.30760.93260.11380.0913
82.673116.00004.00000.73320.17800.1364
92.322213.48233.67180.85660.18350.1325
Mean2.425815.32653.70390.81260.16180.1240
Std0.461810.91501.26780.12700.03130.022202.141210.44313.23160.84790.15720.1166
12.705519.39334.40380.64420.17520.1359
22.462412.16343.48760.87560.13110.1090
32.549213.01723.60790.87780.14580.1194
42.12749.92923.15110.85560.12490.0973
52.269610.95883.31040.79820.16110.1280
61.96106.54852.55900.92180.09960.0857
72.28568.53342.92120.82480.18190.1560
82.363510.14173.18460.91060.14920.1203
92.359810.32703.21360.89160.15220.1236
Mean2.322511.14563.30710.84480.14780.1192
Std0.20653.22810.45700.07580.02310.0186
\n" @@ -11747,7 +7567,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "54be119808c7439796791b7720adc1d3" + "model_id": "29dad85fc3fe4e098cbe2df2994ba8c0" } }, "metadata": { @@ -11770,6 +7590,9 @@ }, "metadata": {} } + ], + "source": [ + "rfModel = create_model('rf')" ] }, { @@ -11785,18 +7608,15 @@ }, { "cell_type": "code", + "execution_count": 30, "metadata": { "id": "cDcFbCCGim2T", "colab": { "base_uri": "https://localhost:8080/", "height": 526 }, - "outputId": "694a9b92-29d0-410e-9bd8-177a99e10d85" + "outputId": "fa860f28-f561-4631-a5ea-402ea444b19b" }, - "source": [ - "plot_model(rfModel, plot='error')" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -11814,10 +7634,13 @@ "text/plain": [ "
" ], - "image/png": "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\n" + "image/png": "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\n" }, "metadata": {} } + ], + "source": [ + "plot_model(rfModel, plot='error')" ] }, { @@ -11833,18 +7656,15 @@ }, { "cell_type": "code", + "execution_count": 31, "metadata": { "id": "AArgjedYim2U", "colab": { "base_uri": "https://localhost:8080/", "height": 524 }, - "outputId": "028528a8-8ebe-4c4e-8988-80d48b4e084c" + "outputId": "4beb7f3c-b294-4900-8604-0dc74d4a2b9e" }, - "source": [ - "plot_model(rfModel, plot='learning')" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -11862,10 +7682,13 @@ "text/plain": [ "
" ], - "image/png": "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\n" + "image/png": "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\n" }, "metadata": {} } + ], + "source": [ + "plot_model(rfModel, plot='learning')" ] }, { @@ -11881,18 +7704,15 @@ }, { "cell_type": "code", + "execution_count": 32, "metadata": { "id": "Xicye76Bim2V", "colab": { "base_uri": "https://localhost:8080/", "height": 524 }, - "outputId": "97260fdb-8b64-4c9c-f1f8-e353f0ad6e3d" + "outputId": "3ccedf11-578a-4ce1-f897-1eaa265d721c" }, - "source": [ - "plot_model(rfModel, plot='vc')" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -11910,10 +7730,13 @@ "text/plain": [ "
" ], - "image/png": "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\n" + "image/png": "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\n" }, "metadata": {} } + ], + "source": [ + "plot_model(rfModel, plot='vc')" ] }, { @@ -11929,18 +7752,15 @@ }, { "cell_type": "code", + "execution_count": 33, "metadata": { "id": "GpfVxNKA4ml3", "colab": { "base_uri": "https://localhost:8080/", - "height": 582 + "height": 614 }, - "outputId": "c5e974fa-10b4-4ccf-cf99-70a951d3fefb" + "outputId": "d01164e8-88dc-485d-e8ff-fd42f9de96e7" }, - "source": [ - "plot_model(rfModel, plot='parameter')" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -11968,16 +7788,17 @@ "min_samples_leaf 1\n", "min_samples_split 2\n", "min_weight_fraction_leaf 0.0\n", + "monotonic_cst None\n", "n_estimators 100\n", "n_jobs -1\n", "oob_score False\n", - "random_state 3304\n", + "random_state 3981\n", "verbose 0\n", "warm_start False" ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", @@ -12531,112 +8360,112 @@ " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 MAEMSERMSER2RMSLEMAPEMAEMSERMSER2RMSLEMAPE
Fold
04.002843.95646.63000.59940.21180.1560
12.991717.57864.19270.82750.16690.1415
22.791717.89694.23050.80570.16670.1238
33.750040.86226.39240.55600.21730.1633
43.060032.26375.68010.60460.20670.1560
53.377124.15204.91450.43040.24830.1944
62.500010.38493.22260.74610.20920.1562
73.400037.50866.12440.76900.25060.1747
83.380045.27866.72890.24490.24510.1790
93.697159.07267.68590.37160.29700.2349
Mean3.295032.89545.58020.59550.22200.1680
Std0.439014.45271.32550.18780.03780.029103.638924.99724.99970.63600.19290.1630
14.525053.95867.34570.01010.25740.2131
23.200018.46944.29760.81120.19330.1455
33.550020.11334.48480.81110.19660.1650
43.080024.97264.99730.63690.19390.1471
52.477111.58203.40320.78670.15530.1298
63.160027.20465.21580.67500.17190.1424
73.637127.62945.25640.43270.28260.2429
83.551430.18095.49370.73390.23170.1775
92.800013.72113.70420.85600.19230.1521
Mean3.362025.28294.91980.63900.20680.1678
Std0.531211.18121.03830.24010.03700.0333
\n" @@ -12653,7 +8482,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "98a085e08dac4d29ab7a3f9885d44847" + "model_id": "3f513e85e2d34137b64d0441436eeb8c" } }, "metadata": { @@ -12676,6 +8505,9 @@ }, "metadata": {} } + ], + "source": [ + "dtModel = create_model('dt')" ] }, { @@ -12689,18 +8521,15 @@ }, { "cell_type": "code", + "execution_count": 38, "metadata": { "id": "Je3ABt9cLDyC", "colab": { "base_uri": "https://localhost:8080/", - "height": 394 + "height": 425 }, - "outputId": "a756d360-2182-40b9-e854-a648e6d1a711" + "outputId": "5c8ed0b6-5ea0-4187-86e7-6ee26fb1aa3a" }, - "source": [ - "plot_model(dtModel, plot='parameter')" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -12726,12 +8555,13 @@ "min_samples_leaf 1\n", "min_samples_split 2\n", "min_weight_fraction_leaf 0.0\n", - "random_state 3304\n", + "monotonic_cst None\n", + "random_state 3981\n", "splitter best" ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", - "\n", + "
\n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", @@ -13100,112 +8938,112 @@ " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", "
 MAEMSERMSER2RMSLEMAPEMAEMSERMSER2RMSLEMAPE
Fold
03.063930.28395.50310.72400.18010.1224
13.573619.92534.46380.80450.27540.2440
22.431912.78303.57530.86120.18100.1194
32.09447.39922.72010.91960.12940.1083
44.217154.30197.36900.33450.29500.2138
52.984317.14264.14040.59570.21800.1809
61.99717.94942.81950.80570.16830.1283
72.507115.99453.99930.90150.14100.1104
83.100019.01014.36010.68300.18250.1523
92.085711.50233.39150.87760.17700.1263
Mean2.805519.62924.23420.75070.19480.1506
Std0.683713.17001.30410.16970.05080.044603.317219.38844.40320.71770.17870.1534
13.867729.41065.42320.46050.21430.1886
23.188818.53724.30550.81050.17980.1448
32.996318.33904.28240.82780.18340.1439
43.157727.66705.25990.59770.19030.1421
52.428410.76153.28050.80180.16240.1382
62.773917.48384.18140.79110.12930.1069
73.620727.73265.26620.43060.23140.1974
82.887415.32663.91490.86490.20470.1656
93.045016.08684.01080.83120.21040.1715
Mean3.128320.07344.43280.71340.18850.1552
Std0.39085.84110.65090.15180.02760.0251
\n" @@ -13222,7 +9060,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "ad62d4b7874b4aaeb04617992ec150b3" + "model_id": "3aaf4087f4c0472c967eb3ae5033ee6b" } }, "metadata": { @@ -13239,7 +9077,7 @@ "output_type": "stream", "name": "stdout", "text": [ - "Fitting 10 folds for each of 200 candidates, totalling 2000 fits\n" + "Fitting 10 folds for each of 50 candidates, totalling 500 fits\n" ] }, { @@ -13252,6 +9090,9 @@ }, "metadata": {} } + ], + "source": [ + "dtModelTuned = tune_model(dtModel, n_iter = 50)" ] }, { @@ -13265,18 +9106,15 @@ }, { "cell_type": "code", + "execution_count": 40, "metadata": { "id": "1x3FC2eeQc0L", "colab": { "base_uri": "https://localhost:8080/", - "height": 394 + "height": 425 }, - "outputId": "d4384f0a-0a49-4fd8-e3a9-6858fc24b298" + "outputId": "c91354bf-a9f5-4b00-e8f7-0758e0e27b38" }, - "source": [ - "plot_model(dtModelTuned, plot='parameter')" - ], - "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -13292,22 +9130,23 @@ "output_type": "display_data", "data": { "text/plain": [ - " Parameters\n", - "ccp_alpha 0.0\n", - "criterion absolute_error\n", - "max_depth 13\n", - "max_features 1.0\n", - "max_leaf_nodes None\n", - "min_impurity_decrease 0.05\n", - "min_samples_leaf 4\n", - "min_samples_split 10\n", - "min_weight_fraction_leaf 0.0\n", - "random_state 3304\n", - "splitter best" + " Parameters\n", + "ccp_alpha 0.0\n", + "criterion friedman_mse\n", + "max_depth 11\n", + "max_features 1.0\n", + "max_leaf_nodes None\n", + "min_impurity_decrease 0.01\n", + "min_samples_leaf 2\n", + "min_samples_split 9\n", + "min_weight_fraction_leaf 0.0\n", + "monotonic_cst None\n", + "random_state 3981\n", + "splitter best" ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "