Pycaret.ipynb

Pycaret.ipynb

untitled0.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "private_outputs": true,
      "provenance": [],
      "gpuType": "T4",
      "toc_visible": true,
      "name": "Pycaret.ipynb",
      "mount_file_id": "13J41VzJ-ZDT6B2QZeMQ_-5Z6-nvG4Ecy",
      "authorship_tag": "ABX9TyOjFe6djWVWLpqOeBdKKZRC",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/RahulDas-dev/e17486e2c86410c72d7016182320f2e0/untitled0.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ietKSaOfchIj"
      },
      "outputs": [],
      "source": [
        "!pip install pycaret\n",
        "!pip install statsforecast\n",
        "!pip install pycaret[models]"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from pycaret import show_versions\n",
        "\n",
        "show_versions()"
      ],
      "metadata": {
        "id": "gcHleTupyjtv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "from pycaret.time_series import TSForecastingExperiment\n",
        "from pycaret.datasets import get_data\n",
        "import statsforecast"
      ],
      "metadata": {
        "id": "g_rsrTqEdeKx"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#dataframe = pd.read_csv('./drive/MyDrive/datasets/AirQualityUCI.csv')\n",
        "\n",
        "dataframe = get_data(\"airquality\", verbose=False)\n",
        "print(dataframe.shape)\n",
        "\n",
        "dataframe['Date_Time'] = dataframe['Date']+dataframe['Time'].apply(lambda x: str(' ')+x)\n",
        "\n",
        "dataframe['Date_Time'] = pd.to_datetime(dataframe['Date_Time'])\n",
        "\n",
        "dataframe.drop(columns=['Date','Time'], inplace=True)\n",
        "\n",
        "target = \"CO(GT)\"\n",
        "\n",
        "cols_sel = ['Date_Time','CO(GT)','PT08.S1(CO)','NMHC(GT)','C6H6(GT)','PT08.S2(NMHC)','NOx(GT)','T','RH']\n",
        "\n",
        "dataframe = dataframe[cols_sel].copy(deep=True)\n",
        "\n",
        "dataframe.replace(-200, np.nan, inplace=True)\n",
        "\n",
        "dataframe.head()"
      ],
      "metadata": {
        "id": "qdFygqaUeCn9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "global_fig_settings = {\n",
        "    # \"renderer\": \"notebook\",\n",
        "    \"renderer\": \"png\",\n",
        "    \"width\": 1000,\n",
        "    \"height\": 600,\n",
        "}\n",
        "\n",
        "exp_auto = TSForecastingExperiment()\n",
        "\n",
        "FH=48\n",
        "metric = \"mase\"\n",
        "\n",
        "exp_auto.setup(\n",
        "    data=dataframe,\n",
        "    target=target,\n",
        "    index='Date_Time',\n",
        "    engine={\"auto_arima\": \"statsforecast\"},\n",
        "    use_gpu=True,\n",
        "    fh=FH, enforce_exogenous=True,\n",
        "    numeric_imputation_target=\"ffill\", numeric_imputation_exogenous=\"ffill\",\n",
        "    fig_kwargs=global_fig_settings, session_id=42\n",
        ")"
      ],
      "metadata": {
        "id": "5vD0oR9YfvML"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "exclude = [\"auto_arima\", \"bats\", \"tbats\"]\n",
        "best = exp_auto.compare_models(sort=metric, turbo=False, exclude=exclude)"
      ],
      "metadata": {
        "id": "TShIivkogx5n"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "exp_auto.plot_model(best)"
      ],
      "metadata": {
        "id": "oRHCl_0VoqCG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "exp_auto.create_model(\"lasso_cds_dt\")"
      ],
      "metadata": {
        "id": "MIyJXuTmpPY9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "exp_auto.create_model(\"lightgbm_cds_dt\")"
      ],
      "metadata": {
        "id": "iVoJPpHb_fby"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "exp_auto.create_model(\"ridge_cds_dt\")"
      ],
      "metadata": {
        "id": "TXeoZS9V_r1U"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}