Machine_Learning_MEET_EU_Group1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyMBGMf+YUe3RYzd9KdV1ZEh",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/agerexx/meeteu_24-25/blob/main/Machine_Learning_MEET_EU_Group1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Intro\n",
        "We are going to use **Python** to build our machine learning model. Python is a programming language and we are using Google Colab to run (execute) our Python code. Google Colab allow us to share a notebook (like this one, titled \"Machine Learning - MEET-EU Group1\") were we can write in Python (code cells) but we can also write normal text (text cells), like this section. All the cells can be executed one by one. To execute one cell you can either click the triangle on the left of the cell or use Shift+Enter.\n",
        "\n",
        "We will use some very popular **Python libraries**. Normally, a library is a collection of books or is a room or place where many books are stored to be used later. Similarly, in the programming world, a library is a collection of precompiled codes that can be used later on in a program for some specific well-defined operations. Libraries need to be imported before use. This is tipically done at the very beginning of the code, in the `import` statements. With `as` we can specify with which abbreviation we are referring to that library later in the code.\n",
        "\n",
        "**Pandas**: import .csv, .xlsx and other file formats, deal with tabular data and perform data analysis (usual abbreviation: pd)\n",
        "\n",
        "**Scikit-Learn**: machine learning\n",
        "\n",
        "**Seaborn**: we can draw some graphs if you get interested in use it (usual abbreviation: sns)\n",
        "\n",
        "For example, we will use `pd.read_csv` to read a *.csv* file. This formula means that we are using the function `read_csv` from the library `pandas` (abbreviated as `pd`).\n",
        "\n",
        "The symbol `#` is used to add comments. This means that what comes after the `#` is not code, but just a line of text we are adding to the code, usually as an explanation of what we are doing or as the title of a section of code.\n",
        "\n",
        "With the symbol `=` we are assigning the value to a variable. What comes before the `=` is the name of the variable (in our case, you'll be free to choose all the names of the variables, but it is useful to assign names that help you remind what these variable represents)."
      ],
      "metadata": {
        "id": "yQyzJ1G7-B3Z"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Input file\n",
        "We need to pass to the model our input data. Let's start with the BEST approach: for each molecule we use only the scores of the best pose (*_conf_01* in VEGA ZZ).\n",
        "The data is formatted as a table:\n",
        "\n",
        "\n",
        "*   each row is a molecule (best pose). In our cases, we have 220 **samples** (or **instances**).\n",
        "*   in the columns we have the scores of that pose calculated with RESCORE+ after the docking. These scores represents the **features**.\n",
        "\n",
        "The output of RESCORE+ is *.csv* file, with semicolon (;) as separator between values and comma (,) as decimal separator. You can easily visualize the semicolon as separator by opening the *.csv* file with Notepad++ or WordPad etc. (text editors).\n",
        "We will use **Pandas** to read the *.csv* file. Pandas is a very popular Python library for manipulating tabular data and time series."
      ],
      "metadata": {
        "id": "uI43TJMnmrK1"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "HvpYsitxmpow"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "df = pd.read_csv(r'/content/TrmD-Ligand_dataset_Rescore_MEAN.csv', sep=';', decimal=',')\n",
        "# if .xlsx file, uncomment the cell below:\n",
        "# df = pd.read_excel(r'/content/TrmD-Ligand_dataset_Rescore_MEAN.xlsx')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df #\"df\" is a common abbreviation for \"dataframe\""
      ],
      "metadata": {
        "id": "vJb4LCEbEnM8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 444
        },
        "outputId": "845e75e9-c358-49a1-e7be-1bb49e1112d2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  Name  Activity        CHARMM          CVFF       ELECT  \\\n",
              "0    135399632_conf_01         1  1.941301e+02  3.178161e+02   -3.498840   \n",
              "1    135680893_conf_01         1  2.417934e+02  3.373358e+02   -1.620516   \n",
              "2    135741302_conf_01         0  4.090772e+01  4.381574e+01   -3.603769   \n",
              "3    135992006_conf_01         1  7.166365e+01  1.125269e+02    0.945706   \n",
              "4       215639_conf_01         1  1.203303e+02  1.599074e+02   -6.180972   \n",
              "..                 ...       ...           ...           ...         ...   \n",
              "215  P90271950_conf_01         0  9.726771e+01  7.488078e+01  -13.732842   \n",
              "216  P92054619_conf_01         0  1.386542e+04  1.869711e+04  -10.274024   \n",
              "217  P92081331_conf_01         0  2.643720e+07  4.678889e+07  -29.381078   \n",
              "218  P96241614_conf_01         0  1.400440e+01 -1.094068e+01  120.042229   \n",
              "219  P97990151_conf_01         0  1.803129e+03  2.263881e+03  -12.982710   \n",
              "\n",
              "       ELECTDD     MLPINS  MLPINS_2  MLPINS_3  MLPINS_F  ...  \\\n",
              "0    -1.768797  11.432599  1.497692  0.223678  1.494388  ...   \n",
              "1    -0.783618   6.153829  0.801446  0.117008  0.727015  ...   \n",
              "2    -0.935819  15.083629  1.991056  0.290506  1.991479  ...   \n",
              "3    -0.087274   6.372239  0.831118  0.118428  0.728735  ...   \n",
              "4    -0.593197  18.990729  2.479175  0.365199  2.257870  ...   \n",
              "..         ...        ...       ...       ...       ...  ...   \n",
              "215  -2.563642   2.394215  0.169957 -0.012641 -0.985839  ...   \n",
              "216  -0.739233  20.262608  2.707630  0.398525  2.467520  ...   \n",
              "217  -7.345710 -30.579139 -4.092250 -0.612330 -3.772317  ...   \n",
              "218  10.646903   4.612017  0.554234  0.068404  0.300188  ...   \n",
              "219  -7.429924   3.626619  0.693547  0.153793  2.405467  ...   \n",
              "\n",
              "     PLANTS_PLP_NORM_HEVATMS  PLANTS_PLP_NORM_WEIGHT  PLANTS_PLP95  \\\n",
              "0                  -4.097188               -0.310365   -100.639000   \n",
              "1                  -4.011187               -0.301038   -101.889000   \n",
              "2                  -3.741529               -0.286202    -99.318000   \n",
              "3                  -3.941062               -0.297148   -100.370000   \n",
              "4                  -4.094875               -0.287020    -97.660000   \n",
              "..                       ...                     ...           ...   \n",
              "215                -3.446857               -0.243444    -80.057000   \n",
              "216                -3.406000               -0.185466   -152.829997   \n",
              "217                 0.783455                0.056514    -33.980000   \n",
              "218                -4.067250               -0.255894   -102.108001   \n",
              "219                -2.672586               -0.199033   -141.663998   \n",
              "\n",
              "     PLANTS_PLP95_NORM_HEVATMS  PLANTS_PLP95_NORM_WEIGHT  XS_HPScore  \\\n",
              "0                    -6.289937                 -0.476467       5.730   \n",
              "1                    -6.368062                 -0.477921       5.753   \n",
              "2                    -5.842235                 -0.446891       5.873   \n",
              "3                    -6.273125                 -0.472981       5.646   \n",
              "4                    -6.103750                 -0.427827       5.822   \n",
              "..                         ...                       ...         ...   \n",
              "215                  -5.718357                 -0.403875       5.080   \n",
              "216                  -5.458214                 -0.297215       6.850   \n",
              "217                  -1.029697                 -0.074277       6.162   \n",
              "218                  -6.381750                 -0.401512       5.479   \n",
              "219                  -4.884965                 -0.363794       7.130   \n",
              "\n",
              "     XS_HMScore  XS_HSScore  XS_Average  XS_Binding  \n",
              "0         5.896       5.691       5.773      -7.872  \n",
              "1         5.957       5.752       5.821      -7.938  \n",
              "2         5.940       5.988       5.934      -8.095  \n",
              "3         5.836       5.713       5.734      -7.819  \n",
              "4         5.996       5.922       5.914      -8.065  \n",
              "..          ...         ...         ...         ...  \n",
              "215       5.502       5.382       5.321      -7.259  \n",
              "216       7.043       7.005       6.966      -9.502  \n",
              "217       6.018       5.944       6.042      -8.241  \n",
              "218       5.645       5.551       5.560      -7.581  \n",
              "219       6.997       7.306       7.143      -9.745  \n",
              "\n",
              "[220 rows x 27 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-db013f7f-a249-4b6f-b46c-0b3a69f968d6\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Name</th>\n",
              "      <th>Activity</th>\n",
              "      <th>CHARMM</th>\n",
              "      <th>CVFF</th>\n",
              "      <th>ELECT</th>\n",
              "      <th>ELECTDD</th>\n",
              "      <th>MLPINS</th>\n",
              "      <th>MLPINS_2</th>\n",
              "      <th>MLPINS_3</th>\n",
              "      <th>MLPINS_F</th>\n",
              "      <th>...</th>\n",
              "      <th>PLANTS_PLP_NORM_HEVATMS</th>\n",
              "      <th>PLANTS_PLP_NORM_WEIGHT</th>\n",
              "      <th>PLANTS_PLP95</th>\n",
              "      <th>PLANTS_PLP95_NORM_HEVATMS</th>\n",
              "      <th>PLANTS_PLP95_NORM_WEIGHT</th>\n",
              "      <th>XS_HPScore</th>\n",
              "      <th>XS_HMScore</th>\n",
              "      <th>XS_HSScore</th>\n",
              "      <th>XS_Average</th>\n",
              "      <th>XS_Binding</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>135399632_conf_01</td>\n",
              "      <td>1</td>\n",
              "      <td>1.941301e+02</td>\n",
              "      <td>3.178161e+02</td>\n",
              "      <td>-3.498840</td>\n",
              "      <td>-1.768797</td>\n",
              "      <td>11.432599</td>\n",
              "      <td>1.497692</td>\n",
              "      <td>0.223678</td>\n",
              "      <td>1.494388</td>\n",
              "      <td>...</td>\n",
              "      <td>-4.097188</td>\n",
              "      <td>-0.310365</td>\n",
              "      <td>-100.639000</td>\n",
              "      <td>-6.289937</td>\n",
              "      <td>-0.476467</td>\n",
              "      <td>5.730</td>\n",
              "      <td>5.896</td>\n",
              "      <td>5.691</td>\n",
              "      <td>5.773</td>\n",
              "      <td>-7.872</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>135680893_conf_01</td>\n",
              "      <td>1</td>\n",
              "      <td>2.417934e+02</td>\n",
              "      <td>3.373358e+02</td>\n",
              "      <td>-1.620516</td>\n",
              "      <td>-0.783618</td>\n",
              "      <td>6.153829</td>\n",
              "      <td>0.801446</td>\n",
              "      <td>0.117008</td>\n",
              "      <td>0.727015</td>\n",
              "      <td>...</td>\n",
              "      <td>-4.011187</td>\n",
              "      <td>-0.301038</td>\n",
              "      <td>-101.889000</td>\n",
              "      <td>-6.368062</td>\n",
              "      <td>-0.477921</td>\n",
              "      <td>5.753</td>\n",
              "      <td>5.957</td>\n",
              "      <td>5.752</td>\n",
              "      <td>5.821</td>\n",
              "      <td>-7.938</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>135741302_conf_01</td>\n",
              "      <td>0</td>\n",
              "      <td>4.090772e+01</td>\n",
              "      <td>4.381574e+01</td>\n",
              "      <td>-3.603769</td>\n",
              "      <td>-0.935819</td>\n",
              "      <td>15.083629</td>\n",
              "      <td>1.991056</td>\n",
              "      <td>0.290506</td>\n",
              "      <td>1.991479</td>\n",
              "      <td>...</td>\n",
              "      <td>-3.741529</td>\n",
              "      <td>-0.286202</td>\n",
              "      <td>-99.318000</td>\n",
              "      <td>-5.842235</td>\n",
              "      <td>-0.446891</td>\n",
              "      <td>5.873</td>\n",
              "      <td>5.940</td>\n",
              "      <td>5.988</td>\n",
              "      <td>5.934</td>\n",
              "      <td>-8.095</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>135992006_conf_01</td>\n",
              "      <td>1</td>\n",
              "      <td>7.166365e+01</td>\n",
              "      <td>1.125269e+02</td>\n",
              "      <td>0.945706</td>\n",
              "      <td>-0.087274</td>\n",
              "      <td>6.372239</td>\n",
              "      <td>0.831118</td>\n",
              "      <td>0.118428</td>\n",
              "      <td>0.728735</td>\n",
              "      <td>...</td>\n",
              "      <td>-3.941062</td>\n",
              "      <td>-0.297148</td>\n",
              "      <td>-100.370000</td>\n",
              "      <td>-6.273125</td>\n",
              "      <td>-0.472981</td>\n",
              "      <td>5.646</td>\n",
              "      <td>5.836</td>\n",
              "      <td>5.713</td>\n",
              "      <td>5.734</td>\n",
              "      <td>-7.819</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>215639_conf_01</td>\n",
              "      <td>1</td>\n",
              "      <td>1.203303e+02</td>\n",
              "      <td>1.599074e+02</td>\n",
              "      <td>-6.180972</td>\n",
              "      <td>-0.593197</td>\n",
              "      <td>18.990729</td>\n",
              "      <td>2.479175</td>\n",
              "      <td>0.365199</td>\n",
              "      <td>2.257870</td>\n",
              "      <td>...</td>\n",
              "      <td>-4.094875</td>\n",
              "      <td>-0.287020</td>\n",
              "      <td>-97.660000</td>\n",
              "      <td>-6.103750</td>\n",
              "      <td>-0.427827</td>\n",
              "      <td>5.822</td>\n",
              "      <td>5.996</td>\n",
              "      <td>5.922</td>\n",
              "      <td>5.914</td>\n",
              "      <td>-8.065</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>215</th>\n",
              "      <td>P90271950_conf_01</td>\n",
              "      <td>0</td>\n",
              "      <td>9.726771e+01</td>\n",
              "      <td>7.488078e+01</td>\n",
              "      <td>-13.732842</td>\n",
              "      <td>-2.563642</td>\n",
              "      <td>2.394215</td>\n",
              "      <td>0.169957</td>\n",
              "      <td>-0.012641</td>\n",
              "      <td>-0.985839</td>\n",
              "      <td>...</td>\n",
              "      <td>-3.446857</td>\n",
              "      <td>-0.243444</td>\n",
              "      <td>-80.057000</td>\n",
              "      <td>-5.718357</td>\n",
              "      <td>-0.403875</td>\n",
              "      <td>5.080</td>\n",
              "      <td>5.502</td>\n",
              "      <td>5.382</td>\n",
              "      <td>5.321</td>\n",
              "      <td>-7.259</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>216</th>\n",
              "      <td>P92054619_conf_01</td>\n",
              "      <td>0</td>\n",
              "      <td>1.386542e+04</td>\n",
              "      <td>1.869711e+04</td>\n",
              "      <td>-10.274024</td>\n",
              "      <td>-0.739233</td>\n",
              "      <td>20.262608</td>\n",
              "      <td>2.707630</td>\n",
              "      <td>0.398525</td>\n",
              "      <td>2.467520</td>\n",
              "      <td>...</td>\n",
              "      <td>-3.406000</td>\n",
              "      <td>-0.185466</td>\n",
              "      <td>-152.829997</td>\n",
              "      <td>-5.458214</td>\n",
              "      <td>-0.297215</td>\n",
              "      <td>6.850</td>\n",
              "      <td>7.043</td>\n",
              "      <td>7.005</td>\n",
              "      <td>6.966</td>\n",
              "      <td>-9.502</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>217</th>\n",
              "      <td>P92081331_conf_01</td>\n",
              "      <td>0</td>\n",
              "      <td>2.643720e+07</td>\n",
              "      <td>4.678889e+07</td>\n",
              "      <td>-29.381078</td>\n",
              "      <td>-7.345710</td>\n",
              "      <td>-30.579139</td>\n",
              "      <td>-4.092250</td>\n",
              "      <td>-0.612330</td>\n",
              "      <td>-3.772317</td>\n",
              "      <td>...</td>\n",
              "      <td>0.783455</td>\n",
              "      <td>0.056514</td>\n",
              "      <td>-33.980000</td>\n",
              "      <td>-1.029697</td>\n",
              "      <td>-0.074277</td>\n",
              "      <td>6.162</td>\n",
              "      <td>6.018</td>\n",
              "      <td>5.944</td>\n",
              "      <td>6.042</td>\n",
              "      <td>-8.241</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>218</th>\n",
              "      <td>P96241614_conf_01</td>\n",
              "      <td>0</td>\n",
              "      <td>1.400440e+01</td>\n",
              "      <td>-1.094068e+01</td>\n",
              "      <td>120.042229</td>\n",
              "      <td>10.646903</td>\n",
              "      <td>4.612017</td>\n",
              "      <td>0.554234</td>\n",
              "      <td>0.068404</td>\n",
              "      <td>0.300188</td>\n",
              "      <td>...</td>\n",
              "      <td>-4.067250</td>\n",
              "      <td>-0.255894</td>\n",
              "      <td>-102.108001</td>\n",
              "      <td>-6.381750</td>\n",
              "      <td>-0.401512</td>\n",
              "      <td>5.479</td>\n",
              "      <td>5.645</td>\n",
              "      <td>5.551</td>\n",
              "      <td>5.560</td>\n",
              "      <td>-7.581</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>219</th>\n",
              "      <td>P97990151_conf_01</td>\n",
              "      <td>0</td>\n",
              "      <td>1.803129e+03</td>\n",
              "      <td>2.263881e+03</td>\n",
              "      <td>-12.982710</td>\n",
              "      <td>-7.429924</td>\n",
              "      <td>3.626619</td>\n",
              "      <td>0.693547</td>\n",
              "      <td>0.153793</td>\n",
              "      <td>2.405467</td>\n",
              "      <td>...</td>\n",
              "      <td>-2.672586</td>\n",
              "      <td>-0.199033</td>\n",
              "      <td>-141.663998</td>\n",
              "      <td>-4.884965</td>\n",
              "      <td>-0.363794</td>\n",
              "      <td>7.130</td>\n",
              "      <td>6.997</td>\n",
              "      <td>7.306</td>\n",
              "      <td>7.143</td>\n",
              "      <td>-9.745</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>220 rows × 27 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-db013f7f-a249-4b6f-b46c-0b3a69f968d6')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-db013f7f-a249-4b6f-b46c-0b3a69f968d6 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-db013f7f-a249-4b6f-b46c-0b3a69f968d6');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-90469dd8-3239-4352-991b-bd2d9eb36d9e\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-90469dd8-3239-4352-991b-bd2d9eb36d9e')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-90469dd8-3239-4352-991b-bd2d9eb36d9e button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_94d790c1-4618-498f-8ab1-7720f13f2907\">\n",
              "    <style>\n",
              "      .colab-df-generate {\n",
              "        background-color: #E8F0FE;\n",
              "        border: none;\n",
              "        border-radius: 50%;\n",
              "        cursor: pointer;\n",
              "        display: none;\n",
              "        fill: #1967D2;\n",
              "        height: 32px;\n",
              "        padding: 0 0 0 0;\n",
              "        width: 32px;\n",
              "      }\n",
              "\n",
              "      .colab-df-generate:hover {\n",
              "        background-color: #E2EBFA;\n",
              "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "        fill: #174EA6;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate:hover {\n",
              "        background-color: #434B5C;\n",
              "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "        fill: #FFFFFF;\n",
              "      }\n",
              "    </style>\n",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('df')\"\n",
              "            title=\"Generate code using this dataframe.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_94d790c1-4618-498f-8ab1-7720f13f2907 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('df');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df"
            }
          },
          "metadata": {},
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Index, features, class\n",
        "The machine learning model doesn't need the names of the molecules to learn about them, but we need to keep these names to be later able to identitfy the molecules (to which molecule does each row of data correspond?). We can set the names as the **index** of the table. The index is not considered as a normal column, so we won't need to exclude it from the features. In the below section of code, we are setting the column \"Name\" as the index of the dataframe.\n",
        "\n",
        "A **feature** is a measurable property of some data-sample that is used as input for a ML model for training and predicting. In our case, the features are the scores of the rescoring. The machine learning model will be trained to predict if a molecule is active or not based on these features. When building a machine learning model, the features are the X variables, so we are defining X as our set of features (our scores from the rescoring).\n",
        "\n",
        "The **class** is the label assigned to each sample. The machine learning model has to predict the label of each sample. In our case, the classes are two: active molecules and inactive molecules. Since we have just two classes, our task is defined as binary classification (the model has to predict to which class each sample belongs and the classes are only two). When building a machine learning model, the class to be predicted is the y variable, so we are defining y the \"Activity\" column."
      ],
      "metadata": {
        "id": "vPQzC0zrG_3s"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df.set_index(\"Name\", inplace=True)"
      ],
      "metadata": {
        "id": "77ubk94GEns-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X = df.loc[:, \"CHARMM\":\"XS_Binding\"]"
      ],
      "metadata": {
        "id": "XeTS-N0eJbht"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y = df['Activity']"
      ],
      "metadata": {
        "id": "WST1wJbCLSHu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Training set and test set\n",
        "We need to divide our entire set (220 samples=110 actives + 110 inactives) to obtain a training set and a test set.\n",
        "\n",
        "*   **Training set**: it is the data your machine learning model learns from. These data are labelled and during training we pass to the model both the features (X) and the class (y) of these samples. In this way, the model knows if each sample is active or not and learn to recognize patterns in the features to discriminate between active and inactive molecules.\n",
        "*   **Test set**: based on what it has learned in the training phase, the model will predict the class (y) of these samples knowing only their features (X). Since we know the class (y) of these samples too, after the model prediction we can use the prediction itself to evaluate model's performance: was the model able to predict the correct label or not?\n",
        "\n",
        "We use the function `train_test_split` from scikit-learn to divide the training set and the test set. We will split the data so that the 80% of it made up the training set, while the remaining 20% represents the test set. We will use the option `stratify=y`; a **stratified split** ensures that the proportion of each class (active/inactive) in the original dataset is preserved in both the training and testing sets. In our case, the proportion is 50:50 because our dataset is balanced.\n",
        "\n",
        "`shuffle=True` is used to scramble the order of the samples before performing the splitting and this is useful to ensure the randomness in data distribution.\n",
        "\n",
        "`random_state=42` ensures that the same randomization is used each time you run the code, resulting in the same splits of the data. This is in general the purpose of random state: while this number remains the same, the output of a random operation remains the same. The random state can be any number and you can choose the number you want. The only thing that you have to remember is to use the same number if you want to reproduce the exact same output. If you have a lucky number is your time to speak up and we will use it as random state to bring luck to our project ;)\n"
      ],
      "metadata": {
        "id": "wsmxVychc-5l"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split"
      ],
      "metadata": {
        "id": "l8igBRgYiIU2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.8, stratify=y, shuffle=True, random_state=42)"
      ],
      "metadata": {
        "id": "qxJK5YPsdBJT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Accuracy\n",
        "With sklearn we can use different metrics to evaluate the performance of our model. Accuracy is the default scoring metric for Random Forest in scikit-learn and it is calculated according to the formula:\n",
        "![Accuracy.PNG](data:image/png;base64,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)\n",
        "\n",
        "We are working with binary classification, so we just have two classes, that are generalized as True/False in the formula. Let's say that True is class 1 (Active) and False is class 0 (inactive).\n",
        "To add some clarity:\n",
        "*   True positive: the molecules that are active (1) and are correctly predicted as active (1)\n",
        "*   True negative: the molecules that are inactive (0) and are correctly predicted as inactive (0)\n",
        "*   False positive: the molecules that are inactive (0) but are wrongly predicted as active (1)\n",
        "*   False negative: the molecules that are active (1) but are wrongly predicted as inactive (0).\n",
        "\n",
        "The accuracy can be bewteen 0 and 1. Accuracy of 0 means means the model got all the predictions wrong. Accuracy of 0.5 is the equivalent of random guessing, if the classes are balanced, as in our case. Accuracy of 1 means the model got all the predictions right."
      ],
      "metadata": {
        "id": "MT65vk9GKuyW"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Model initialization\n",
        "We need to initialize a machine learning model. We'll try to use Random Forest as first approach, with default parameters.\n",
        "\n",
        "We are performing a classification: the model need to predict the class (label) of each sample. So, we need the `RandomForestClassifier` from sklearn."
      ],
      "metadata": {
        "id": "jT_WvwT9Th1u"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "model = RandomForestClassifier(random_state=42)"
      ],
      "metadata": {
        "id": "q_NMR6JhTu7i"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Cross validation\n",
        "**We will perform the training in cross-validation.** This is a way to combine the training and the validation step using the entire training set. The validation is different from testing.\n",
        "\n",
        "*   **Validation set**: used iteratively during model training for performance monitoring and hyperparameter tuning. It is used to asses the model's performance during training and choose the hyperparameters if the hyperparameter tuning is performed.\n",
        "*   **Test set**: typically used only once, after model development. It is used to evaluate the final model's performance.\n",
        "\n",
        "So, a part from the training set and the test set, we would need only a validation set. But divide our entire dataset (220 sampels) in three sets would reduce too much the data to train on. As a solution: we use cross-validation.\n",
        "\n",
        "1.   Split the data (the data in this case is the training set, which is 80% of the entire dataset): divide the training set into folds (eg. 5 parts)\n",
        "2.  Training + Validations: the model is trained on 4 folds, while the 5th left-out fold is used for validation (the model is trained on 4 folds and then based on what it has learned it predicts the class (y) of the 5th fold).\n",
        "3.  Repeat: rotate which fold is used for validation until every fold has been used once.\n",
        "4.  Average the results: combine the validation results to get an overall performance score.\n",
        "\n",
        "**Cross-validation** is a way to check how well a machine learning model will perform on unseen data while still developing it. It splits your data into parts to train the model on one part and validate it on another.\n",
        "\n",
        "![5-K-Fold-Cross-Validation.png](data:image/png;base64,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)\n",
        "\n",
        "To create the folds, we use `StratifiedKFold` from `sklearn.model_selection`."
      ],
      "metadata": {
        "id": "m7REbA7NjFax"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import StratifiedKFold"
      ],
      "metadata": {
        "id": "YIAV9Ja5jEId"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "cv = StratifiedKFold(n_splits=5)"
      ],
      "metadata": {
        "id": "C2QRaMWmTDp-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We then use `cross_val_score` to evaluate the performance of the base model in cross-validation, passing the division in folds we just made (cv)."
      ],
      "metadata": {
        "id": "nzIBKltLWRun"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import cross_val_score\n",
        "cross_val_score(model, X_train, y_train, scoring='accuracy', cv=cv)"
      ],
      "metadata": {
        "id": "ddSkiJhNWN1b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0eec7478-e899-4e6d-ee04-ab08eb020ff5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.88888889, 0.82857143, 0.77142857, 0.82857143, 0.8       ])"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cross_val_score(model, X_train, y_train, scoring='accuracy', cv=cv).mean()"
      ],
      "metadata": {
        "id": "_kzZrJ6Vgvrv",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4cee31ee-4891-4184-d802-5bce449d963f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8234920634920636"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print('accuracy: ', cross_val_score(model, X_train, y_train, scoring='accuracy', cv=cv).mean())\n",
        "print('precision: ', cross_val_score(model, X_train, y_train, scoring='precision', cv=cv).mean())\n",
        "print('recall: ', cross_val_score(model, X_train, y_train, scoring='recall', cv=cv).mean())\n",
        "print('f1: ', cross_val_score(model, X_train, y_train, scoring='f1', cv=cv).mean())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-l-vfObFSnt5",
        "outputId": "7c283c28-64db-4aaa-d25b-ef26c06ba321"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "accuracy:  0.8234920634920636\n",
            "precision:  0.8486676721970839\n",
            "recall:  0.7947712418300653\n",
            "f1:  0.8164787165573643\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Feature selection\n",
        "We checked the performance of a base Random Forest model when trained on our training set using cross-validation.\n",
        "We will now try to increment the performance of the model. The first step to do that is **feature selection**. With feature selection why are trying to keep only the most informative features (columns=rescoring scores=the info the model uses to guess if a molecules is active or not) and this is useful because:\n",
        "\n",
        "*   Decreases over-fitting: **overfitting** is an undesirable machine learning behavior that occurs when the machine learning model gives accurate predictions for training data but not for new data. In overfitting, a model memorizes the training set so closely that it fails to make prediction on new unseen data. In this case, the model it's like a student that memorizes the lesson by-heart: they are very good if the question was written in the book they studied from, but they can't generalize to answer a new question on the subject.\n",
        "*   Improves Accuracy: less misleading data means better modeling accuracy. With feature selection we are trying to reduce the \"noise\" that can confuse the model.\n",
        "*   Reduces Training Time: less data means quicker algorithms.\n",
        "\n",
        "Usually feature selection is a trade-off between accuracy of the selection and time needed for the selection. Since we have quite a few features in total, we can choose a quite extensive feature selection: `SequentialFeatureSelector` from sklearn.\n",
        "\n",
        "**Sequential feature selection** can be performed in two directions: forward or backward. Let's explain forward feature selection. Concretely, we initially start with zero features and find the one feature that maximizes a cross-validated score when an estimator is trained on this single feature. Once that first feature is selected, we repeat the procedure by adding a new feature to the set of selected features. The procedure stops when the desired number of selected features is reached, as determined by the `n_features_to_select` parameter.\n",
        "\n",
        "Backward-SFS follows the same idea but works in the opposite direction: instead of starting with no features and greedily adding features, we start with all the features and greedily remove features from the set.\n",
        "\n",
        "In general, forward and backward selection do not yield equivalent results. Also, one may be much faster than the other depending on the requested number of selected features: if we have 10 features and ask for 7 selected features, forward selection would need to perform 7 iterations while backward selection would only need to perform 3.\n",
        "\n",
        "**How to choose how many features to keep?**\n",
        "Here I used `SequentialFeatureSelector` to figure out how many features our model should use for the best performance. The process starts with no features and adds them one at a time, choosing the most useful features at each step. For every number of features (from 1 to 25 which is the total number of features) the method evaluates different feature combinations and finds the set that works best for that specific number. For that best set of features, the average cross-validation score is plotted.\n",
        "\n",
        "In our case, we see in the plot that a plateau is reached quite soon: with 3 features the accuracy has already reach the plateau. However, we will try with 7 features, since there seems to be a peak.\n",
        "\n",
        "![Feature_selectionpng.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "Ogzj1ELlZ2hO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "## This is the code I used to perform the Sequential Feature Selection by setting n_features_to_select on all the possible values between 1 and 25.\n",
        "## Below I plotted the results using matplotlib\n",
        "## We don't run the code during the meeting because it takes quite a lot of time, but you can try to run it on you pc!\n",
        "import numpy as np\n",
        "from sklearn.feature_selection import SequentialFeatureSelector\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "scores = []\n",
        "num_features = range(1, 26)\n",
        "\n",
        "for n in num_features:\n",
        "    sfs = SequentialFeatureSelector(\n",
        "        model, n_features_to_select=n, direction='forward', cv=5, n_jobs=-1\n",
        "    )\n",
        "    sfs.fit(X_train, y_train)\n",
        "    selected_features = sfs.get_support()\n",
        "    X_train_selected = X_train.loc[:, selected_features]\n",
        "    score = cross_val_score(model, X_train_selected, y_train, cv=5).mean()\n",
        "    scores.append(score)\n",
        "\n",
        "# Plotting the results\n",
        "plt.figure(figsize=(10, 6))\n",
        "plt.plot(num_features, scores, marker='o', label='Cross-validated Accuracy')\n",
        "plt.title('Performance vs. Number of Selected Features')\n",
        "plt.xlabel('Number of Features Selected')\n",
        "plt.ylabel('Cross-Validated Accuracy')\n",
        "plt.ylim([0, 1])\n",
        "plt.grid()\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "KKYVb5I5YFcB",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 356
        },
        "outputId": "51d95213-9dde-4602-8759-b1b33a4696d6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "n_features_to_select must be < n_features.",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-17-633cc8a548e1>\u001b[0m in \u001b[0;36m<cell line: 11>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     13\u001b[0m         \u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_features_to_select\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdirection\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'forward'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcv\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     14\u001b[0m     )\n\u001b[0;32m---> 15\u001b[0;31m     \u001b[0msfs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     16\u001b[0m     \u001b[0mselected_features\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msfs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_support\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     17\u001b[0m     \u001b[0mX_train_selected\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mX_train\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mselected_features\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/sklearn/base.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(estimator, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1387\u001b[0m                 )\n\u001b[1;32m   1388\u001b[0m             ):\n\u001b[0;32m-> 1389\u001b[0;31m                 \u001b[0;32mreturn\u001b[0m \u001b[0mfit_method\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mestimator\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1390\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1391\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/sklearn/feature_selection/_sequential.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, **params)\u001b[0m\n\u001b[1;32m    247\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_features_to_select\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mIntegral\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    248\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_features_to_select\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mn_features\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 249\u001b[0;31m                 \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"n_features_to_select must be < n_features.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    250\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_features_to_select_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_features_to_select\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    251\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_features_to_select\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mReal\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: n_features_to_select must be < n_features."
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.feature_selection import SequentialFeatureSelector\n",
        "sfs = SequentialFeatureSelector(model, n_features_to_select=7, direction='forward', scoring='accuracy', cv=cv)\n",
        "sfs.fit(X_train, y_train)"
      ],
      "metadata": {
        "id": "R135-eO5V3rQ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 166
        },
        "outputId": "f14cb52d-d3be-4143-b352-0e2a51098e6f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "SequentialFeatureSelector(cv=StratifiedKFold(n_splits=5, random_state=None, shuffle=False),\n",
              "                          estimator=RandomForestClassifier(random_state=42),\n",
              "                          n_features_to_select=7, scoring='accuracy')"
            ],
            "text/html": [
              "<style>#sk-container-id-1 {\n",
              "  /* Definition of color scheme common for light and dark mode */\n",
              "  --sklearn-color-text: #000;\n",
              "  --sklearn-color-text-muted: #666;\n",
              "  --sklearn-color-line: gray;\n",
              "  /* Definition of color scheme for unfitted estimators */\n",
              "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
              "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
              "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
              "  --sklearn-color-unfitted-level-3: chocolate;\n",
              "  /* Definition of color scheme for fitted estimators */\n",
              "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
              "  --sklearn-color-fitted-level-1: #d4ebff;\n",
              "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
              "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
              "\n",
              "  /* Specific color for light theme */\n",
              "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
              "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
              "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
              "  --sklearn-color-icon: #696969;\n",
              "\n",
              "  @media (prefers-color-scheme: dark) {\n",
              "    /* Redefinition of color scheme for dark theme */\n",
              "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
              "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
              "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
              "    --sklearn-color-icon: #878787;\n",
              "  }\n",
              "}\n",
              "\n",
              "#sk-container-id-1 {\n",
              "  color: var(--sklearn-color-text);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 pre {\n",
              "  padding: 0;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 input.sk-hidden--visually {\n",
              "  border: 0;\n",
              "  clip: rect(1px 1px 1px 1px);\n",
              "  clip: rect(1px, 1px, 1px, 1px);\n",
              "  height: 1px;\n",
              "  margin: -1px;\n",
              "  overflow: hidden;\n",
              "  padding: 0;\n",
              "  position: absolute;\n",
              "  width: 1px;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-dashed-wrapped {\n",
              "  border: 1px dashed var(--sklearn-color-line);\n",
              "  margin: 0 0.4em 0.5em 0.4em;\n",
              "  box-sizing: border-box;\n",
              "  padding-bottom: 0.4em;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-container {\n",
              "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
              "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
              "     so we also need the `!important` here to be able to override the\n",
              "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
              "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
              "  display: inline-block !important;\n",
              "  position: relative;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-text-repr-fallback {\n",
              "  display: none;\n",
              "}\n",
              "\n",
              "div.sk-parallel-item,\n",
              "div.sk-serial,\n",
              "div.sk-item {\n",
              "  /* draw centered vertical line to link estimators */\n",
              "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
              "  background-size: 2px 100%;\n",
              "  background-repeat: no-repeat;\n",
              "  background-position: center center;\n",
              "}\n",
              "\n",
              "/* Parallel-specific style estimator block */\n",
              "\n",
              "#sk-container-id-1 div.sk-parallel-item::after {\n",
              "  content: \"\";\n",
              "  width: 100%;\n",
              "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
              "  flex-grow: 1;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-parallel {\n",
              "  display: flex;\n",
              "  align-items: stretch;\n",
              "  justify-content: center;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  position: relative;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-parallel-item {\n",
              "  display: flex;\n",
              "  flex-direction: column;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
              "  align-self: flex-end;\n",
              "  width: 50%;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
              "  align-self: flex-start;\n",
              "  width: 50%;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
              "  width: 0;\n",
              "}\n",
              "\n",
              "/* Serial-specific style estimator block */\n",
              "\n",
              "#sk-container-id-1 div.sk-serial {\n",
              "  display: flex;\n",
              "  flex-direction: column;\n",
              "  align-items: center;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  padding-right: 1em;\n",
              "  padding-left: 1em;\n",
              "}\n",
              "\n",
              "\n",
              "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
              "clickable and can be expanded/collapsed.\n",
              "- Pipeline and ColumnTransformer use this feature and define the default style\n",
              "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
              "*/\n",
              "\n",
              "/* Pipeline and ColumnTransformer style (default) */\n",
              "\n",
              "#sk-container-id-1 div.sk-toggleable {\n",
              "  /* Default theme specific background. It is overwritten whether we have a\n",
              "  specific estimator or a Pipeline/ColumnTransformer */\n",
              "  background-color: var(--sklearn-color-background);\n",
              "}\n",
              "\n",
              "/* Toggleable label */\n",
              "#sk-container-id-1 label.sk-toggleable__label {\n",
              "  cursor: pointer;\n",
              "  display: flex;\n",
              "  width: 100%;\n",
              "  margin-bottom: 0;\n",
              "  padding: 0.5em;\n",
              "  box-sizing: border-box;\n",
              "  text-align: center;\n",
              "  align-items: start;\n",
              "  justify-content: space-between;\n",
              "  gap: 0.5em;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 label.sk-toggleable__label .caption {\n",
              "  font-size: 0.6rem;\n",
              "  font-weight: lighter;\n",
              "  color: var(--sklearn-color-text-muted);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
              "  /* Arrow on the left of the label */\n",
              "  content: \"▸\";\n",
              "  float: left;\n",
              "  margin-right: 0.25em;\n",
              "  color: var(--sklearn-color-icon);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
              "  color: var(--sklearn-color-text);\n",
              "}\n",
              "\n",
              "/* Toggleable content - dropdown */\n",
              "\n",
              "#sk-container-id-1 div.sk-toggleable__content {\n",
              "  max-height: 0;\n",
              "  max-width: 0;\n",
              "  overflow: hidden;\n",
              "  text-align: left;\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-toggleable__content pre {\n",
              "  margin: 0.2em;\n",
              "  border-radius: 0.25em;\n",
              "  color: var(--sklearn-color-text);\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
              "  /* Expand drop-down */\n",
              "  max-height: 200px;\n",
              "  max-width: 100%;\n",
              "  overflow: auto;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
              "  content: \"▾\";\n",
              "}\n",
              "\n",
              "/* Pipeline/ColumnTransformer-specific style */\n",
              "\n",
              "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Estimator-specific style */\n",
              "\n",
              "/* Colorize estimator box */\n",
              "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
              "#sk-container-id-1 div.sk-label label {\n",
              "  /* The background is the default theme color */\n",
              "  color: var(--sklearn-color-text-on-default-background);\n",
              "}\n",
              "\n",
              "/* On hover, darken the color of the background */\n",
              "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "/* Label box, darken color on hover, fitted */\n",
              "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Estimator label */\n",
              "\n",
              "#sk-container-id-1 div.sk-label label {\n",
              "  font-family: monospace;\n",
              "  font-weight: bold;\n",
              "  display: inline-block;\n",
              "  line-height: 1.2em;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-label-container {\n",
              "  text-align: center;\n",
              "}\n",
              "\n",
              "/* Estimator-specific */\n",
              "#sk-container-id-1 div.sk-estimator {\n",
              "  font-family: monospace;\n",
              "  border: 1px dotted var(--sklearn-color-border-box);\n",
              "  border-radius: 0.25em;\n",
              "  box-sizing: border-box;\n",
              "  margin-bottom: 0.5em;\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-estimator.fitted {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "/* on hover */\n",
              "#sk-container-id-1 div.sk-estimator:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
              "\n",
              "/* Common style for \"i\" and \"?\" */\n",
              "\n",
              ".sk-estimator-doc-link,\n",
              "a:link.sk-estimator-doc-link,\n",
              "a:visited.sk-estimator-doc-link {\n",
              "  float: right;\n",
              "  font-size: smaller;\n",
              "  line-height: 1em;\n",
              "  font-family: monospace;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  border-radius: 1em;\n",
              "  height: 1em;\n",
              "  width: 1em;\n",
              "  text-decoration: none !important;\n",
              "  margin-left: 0.5em;\n",
              "  text-align: center;\n",
              "  /* unfitted */\n",
              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-unfitted-level-1);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link.fitted,\n",
              "a:link.sk-estimator-doc-link.fitted,\n",
              "a:visited.sk-estimator-doc-link.fitted {\n",
              "  /* fitted */\n",
              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-fitted-level-1);\n",
              "}\n",
              "\n",
              "/* On hover */\n",
              "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
              ".sk-estimator-doc-link:hover,\n",
              "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
              ".sk-estimator-doc-link:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
              ".sk-estimator-doc-link.fitted:hover,\n",
              "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
              ".sk-estimator-doc-link.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "/* Span, style for the box shown on hovering the info icon */\n",
              ".sk-estimator-doc-link span {\n",
              "  display: none;\n",
              "  z-index: 9999;\n",
              "  position: relative;\n",
              "  font-weight: normal;\n",
              "  right: .2ex;\n",
              "  padding: .5ex;\n",
              "  margin: .5ex;\n",
              "  width: min-content;\n",
              "  min-width: 20ex;\n",
              "  max-width: 50ex;\n",
              "  color: var(--sklearn-color-text);\n",
              "  box-shadow: 2pt 2pt 4pt #999;\n",
              "  /* unfitted */\n",
              "  background: var(--sklearn-color-unfitted-level-0);\n",
              "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link.fitted span {\n",
              "  /* fitted */\n",
              "  background: var(--sklearn-color-fitted-level-0);\n",
              "  border: var(--sklearn-color-fitted-level-3);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link:hover span {\n",
              "  display: block;\n",
              "}\n",
              "\n",
              "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
              "\n",
              "#sk-container-id-1 a.estimator_doc_link {\n",
              "  float: right;\n",
              "  font-size: 1rem;\n",
              "  line-height: 1em;\n",
              "  font-family: monospace;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  border-radius: 1rem;\n",
              "  height: 1rem;\n",
              "  width: 1rem;\n",
              "  text-decoration: none;\n",
              "  /* unfitted */\n",
              "  color: var(--sklearn-color-unfitted-level-1);\n",
              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
              "  /* fitted */\n",
              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-fitted-level-1);\n",
              "}\n",
              "\n",
              "/* On hover */\n",
              "#sk-container-id-1 a.estimator_doc_link:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-3);\n",
              "}\n",
              "</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>SequentialFeatureSelector(cv=StratifiedKFold(n_splits=5, random_state=None, shuffle=False),\n",
              "                          estimator=RandomForestClassifier(random_state=42),\n",
              "                          n_features_to_select=7, scoring=&#x27;accuracy&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>SequentialFeatureSelector</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.6/modules/generated/sklearn.feature_selection.SequentialFeatureSelector.html\">?<span>Documentation for SequentialFeatureSelector</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\"><pre>SequentialFeatureSelector(cv=StratifiedKFold(n_splits=5, random_state=None, shuffle=False),\n",
              "                          estimator=RandomForestClassifier(random_state=42),\n",
              "                          n_features_to_select=7, scoring=&#x27;accuracy&#x27;)</pre></div> </div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>estimator: RandomForestClassifier</div></div></label><div class=\"sk-toggleable__content fitted\"><pre>RandomForestClassifier(random_state=42)</pre></div> </div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>RandomForestClassifier</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.6/modules/generated/sklearn.ensemble.RandomForestClassifier.html\">?<span>Documentation for RandomForestClassifier</span></a></div></label><div class=\"sk-toggleable__content fitted\"><pre>RandomForestClassifier(random_state=42)</pre></div> </div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Transform X_train and X_test to include only the selected features\n",
        "selected_features_mask = sfs.get_support()\n",
        "selected_features = X_train.columns[selected_features_mask]\n",
        "\n",
        "X_train_selected = X_train.loc[:, selected_features]  # Transform X_train\n",
        "X_test_selected = X_test.loc[:, selected_features]    # Transform X_test\n",
        "\n",
        "# Print the selected feature names\n",
        "print(\"Selected features:\", list(selected_features))"
      ],
      "metadata": {
        "id": "QBQ2rIjveFMN",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1706eedb-8150-4388-891c-c71444b529e4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Selected features: ['Contacts', 'Contacts_NORM_HEVATMS', 'PLANTS_CHEMPLP', 'PLANTS_PLP', 'PLANTS_PLP95_NORM_HEVATMS', 'PLANTS_PLP95_NORM_WEIGHT', 'XS_HPScore']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cross_val_score(model, X_train_selected, y_train, cv=cv)"
      ],
      "metadata": {
        "id": "5Qx96TCLgotM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2f260e8e-d1f6-4c27-81c2-51af3036397f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.91666667, 0.82857143, 0.82857143, 0.82857143, 0.8       ])"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cross_val_score(model, X_train_selected, y_train, cv=cv).mean()"
      ],
      "metadata": {
        "id": "QTimnUFig7j1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7c922485-e5d9-4c15-86bb-2914971c4b70"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8404761904761905"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print('accuracy: ', cross_val_score(model, X_train_selected, y_train, scoring='accuracy', cv=cv).mean())\n",
        "print('precision: ', cross_val_score(model, X_train_selected, y_train, scoring='precision', cv=cv).mean())\n",
        "print('recall: ', cross_val_score(model, X_train_selected, y_train, scoring='recall', cv=cv).mean())\n",
        "print('f1: ', cross_val_score(model, X_train_selected, y_train, scoring='f1', cv=cv).mean())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8Ur1NRPGTkb3",
        "outputId": "0e67c173-1e49-4b71-c058-77dfbeef2935"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "accuracy:  0.8404761904761905\n",
            "precision:  0.8461633722314836\n",
            "recall:  0.8411764705882353\n",
            "f1:  0.8384160951123899\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Hyperparameter tuning\n",
        "\n",
        "After feature selection, another operation we can perform to enhance the model performance is **hyperparameter tuning**.\n",
        "\n",
        "Random Forest is like a group of decision-making experts (decision trees). Each expert giver their opinion (a prediction), and the Random Forest combines these opinions to make a final decision. This teamwork helps make more reliable predictions.\n",
        "\n",
        "**Hyperparameters** are like settings or rules you set before the forest starts growing. These rules control how the experts (trees) behave and work together. Examples of these settings include:\n",
        "\n",
        "*   How many trees you want in your forest: `n_estimators`\n",
        "*   How deep each tree is allowed to grow: `max_depth`\n",
        "*   How many features each tree should look at when making decisions: `max_features`\n",
        "\n",
        "Hyperparameter tuning is the process of trying different settings to find the best combination. You try different settings (different combination of hyperparameters) and pick the ones that make the model perform the best; in other words, we train the model with different combinations of hyperparameter and validate it (using the training set in cross-validation). We will pick the model with the best cross-validation score.\n",
        "\n",
        "If you want to learn more about Random Forest parameters: https://medium.com/@taplapinger/tuning-a-random-forest-classifier-1b252d1dde92\n",
        "\n",
        "`GridSearchCV` is a tool of scikit-learn that can automate this process by testing many combinations and finding the best one. Using the option `refit=True`, the function refit the model using the best found parameters on the whole dataset. So at the end of th grid search we have the model ready to make predictions.\n"
      ],
      "metadata": {
        "id": "M3CIuS0WZ6Je"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# default params: n_estimators=100, max_depth=None, max_features='sqrt'\n",
        "param_grid = {'n_estimators': [50, 100, 150, 200],\n",
        "              'max_depth': [None, 3, 5, 7],\n",
        "              'max_features': ['sqrt', 'log2']}"
      ],
      "metadata": {
        "id": "5SO--YVw0JUP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import GridSearchCV\n",
        "grid_search = GridSearchCV(model, param_grid, scoring='accuracy', cv=cv, refit=True)"
      ],
      "metadata": {
        "id": "Ii-_1EHP3Bvz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search.fit(X_train_selected, y_train)"
      ],
      "metadata": {
        "id": "u0qpwT_E4xyk",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 166
        },
        "outputId": "1c722742-af3d-4d71-90bb-92e3b4fcaf57"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=StratifiedKFold(n_splits=5, random_state=None, shuffle=False),\n",
              "             estimator=RandomForestClassifier(random_state=42),\n",
              "             param_grid={'max_depth': [None, 3, 5, 7],\n",
              "                         'max_features': ['sqrt', 'log2'],\n",
              "                         'n_estimators': [50, 100, 150, 200]},\n",
              "             scoring='accuracy')"
            ],
            "text/html": [
              "<style>#sk-container-id-2 {\n",
              "  /* Definition of color scheme common for light and dark mode */\n",
              "  --sklearn-color-text: #000;\n",
              "  --sklearn-color-text-muted: #666;\n",
              "  --sklearn-color-line: gray;\n",
              "  /* Definition of color scheme for unfitted estimators */\n",
              "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
              "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
              "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
              "  --sklearn-color-unfitted-level-3: chocolate;\n",
              "  /* Definition of color scheme for fitted estimators */\n",
              "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
              "  --sklearn-color-fitted-level-1: #d4ebff;\n",
              "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
              "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
              "\n",
              "  /* Specific color for light theme */\n",
              "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
              "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
              "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
              "  --sklearn-color-icon: #696969;\n",
              "\n",
              "  @media (prefers-color-scheme: dark) {\n",
              "    /* Redefinition of color scheme for dark theme */\n",
              "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
              "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
              "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
              "    --sklearn-color-icon: #878787;\n",
              "  }\n",
              "}\n",
              "\n",
              "#sk-container-id-2 {\n",
              "  color: var(--sklearn-color-text);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 pre {\n",
              "  padding: 0;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 input.sk-hidden--visually {\n",
              "  border: 0;\n",
              "  clip: rect(1px 1px 1px 1px);\n",
              "  clip: rect(1px, 1px, 1px, 1px);\n",
              "  height: 1px;\n",
              "  margin: -1px;\n",
              "  overflow: hidden;\n",
              "  padding: 0;\n",
              "  position: absolute;\n",
              "  width: 1px;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-dashed-wrapped {\n",
              "  border: 1px dashed var(--sklearn-color-line);\n",
              "  margin: 0 0.4em 0.5em 0.4em;\n",
              "  box-sizing: border-box;\n",
              "  padding-bottom: 0.4em;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-container {\n",
              "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
              "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
              "     so we also need the `!important` here to be able to override the\n",
              "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
              "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
              "  display: inline-block !important;\n",
              "  position: relative;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-text-repr-fallback {\n",
              "  display: none;\n",
              "}\n",
              "\n",
              "div.sk-parallel-item,\n",
              "div.sk-serial,\n",
              "div.sk-item {\n",
              "  /* draw centered vertical line to link estimators */\n",
              "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
              "  background-size: 2px 100%;\n",
              "  background-repeat: no-repeat;\n",
              "  background-position: center center;\n",
              "}\n",
              "\n",
              "/* Parallel-specific style estimator block */\n",
              "\n",
              "#sk-container-id-2 div.sk-parallel-item::after {\n",
              "  content: \"\";\n",
              "  width: 100%;\n",
              "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
              "  flex-grow: 1;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-parallel {\n",
              "  display: flex;\n",
              "  align-items: stretch;\n",
              "  justify-content: center;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  position: relative;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-parallel-item {\n",
              "  display: flex;\n",
              "  flex-direction: column;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-parallel-item:first-child::after {\n",
              "  align-self: flex-end;\n",
              "  width: 50%;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-parallel-item:last-child::after {\n",
              "  align-self: flex-start;\n",
              "  width: 50%;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-parallel-item:only-child::after {\n",
              "  width: 0;\n",
              "}\n",
              "\n",
              "/* Serial-specific style estimator block */\n",
              "\n",
              "#sk-container-id-2 div.sk-serial {\n",
              "  display: flex;\n",
              "  flex-direction: column;\n",
              "  align-items: center;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  padding-right: 1em;\n",
              "  padding-left: 1em;\n",
              "}\n",
              "\n",
              "\n",
              "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
              "clickable and can be expanded/collapsed.\n",
              "- Pipeline and ColumnTransformer use this feature and define the default style\n",
              "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
              "*/\n",
              "\n",
              "/* Pipeline and ColumnTransformer style (default) */\n",
              "\n",
              "#sk-container-id-2 div.sk-toggleable {\n",
              "  /* Default theme specific background. It is overwritten whether we have a\n",
              "  specific estimator or a Pipeline/ColumnTransformer */\n",
              "  background-color: var(--sklearn-color-background);\n",
              "}\n",
              "\n",
              "/* Toggleable label */\n",
              "#sk-container-id-2 label.sk-toggleable__label {\n",
              "  cursor: pointer;\n",
              "  display: flex;\n",
              "  width: 100%;\n",
              "  margin-bottom: 0;\n",
              "  padding: 0.5em;\n",
              "  box-sizing: border-box;\n",
              "  text-align: center;\n",
              "  align-items: start;\n",
              "  justify-content: space-between;\n",
              "  gap: 0.5em;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 label.sk-toggleable__label .caption {\n",
              "  font-size: 0.6rem;\n",
              "  font-weight: lighter;\n",
              "  color: var(--sklearn-color-text-muted);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 label.sk-toggleable__label-arrow:before {\n",
              "  /* Arrow on the left of the label */\n",
              "  content: \"▸\";\n",
              "  float: left;\n",
              "  margin-right: 0.25em;\n",
              "  color: var(--sklearn-color-icon);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {\n",
              "  color: var(--sklearn-color-text);\n",
              "}\n",
              "\n",
              "/* Toggleable content - dropdown */\n",
              "\n",
              "#sk-container-id-2 div.sk-toggleable__content {\n",
              "  max-height: 0;\n",
              "  max-width: 0;\n",
              "  overflow: hidden;\n",
              "  text-align: left;\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-toggleable__content.fitted {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-toggleable__content pre {\n",
              "  margin: 0.2em;\n",
              "  border-radius: 0.25em;\n",
              "  color: var(--sklearn-color-text);\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-toggleable__content.fitted pre {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
              "  /* Expand drop-down */\n",
              "  max-height: 200px;\n",
              "  max-width: 100%;\n",
              "  overflow: auto;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
              "  content: \"▾\";\n",
              "}\n",
              "\n",
              "/* Pipeline/ColumnTransformer-specific style */\n",
              "\n",
              "#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Estimator-specific style */\n",
              "\n",
              "/* Colorize estimator box */\n",
              "#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-label label.sk-toggleable__label,\n",
              "#sk-container-id-2 div.sk-label label {\n",
              "  /* The background is the default theme color */\n",
              "  color: var(--sklearn-color-text-on-default-background);\n",
              "}\n",
              "\n",
              "/* On hover, darken the color of the background */\n",
              "#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "/* Label box, darken color on hover, fitted */\n",
              "#sk-container-id-2 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Estimator label */\n",
              "\n",
              "#sk-container-id-2 div.sk-label label {\n",
              "  font-family: monospace;\n",
              "  font-weight: bold;\n",
              "  display: inline-block;\n",
              "  line-height: 1.2em;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-label-container {\n",
              "  text-align: center;\n",
              "}\n",
              "\n",
              "/* Estimator-specific */\n",
              "#sk-container-id-2 div.sk-estimator {\n",
              "  font-family: monospace;\n",
              "  border: 1px dotted var(--sklearn-color-border-box);\n",
              "  border-radius: 0.25em;\n",
              "  box-sizing: border-box;\n",
              "  margin-bottom: 0.5em;\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-estimator.fitted {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "/* on hover */\n",
              "#sk-container-id-2 div.sk-estimator:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-2 div.sk-estimator.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
              "\n",
              "/* Common style for \"i\" and \"?\" */\n",
              "\n",
              ".sk-estimator-doc-link,\n",
              "a:link.sk-estimator-doc-link,\n",
              "a:visited.sk-estimator-doc-link {\n",
              "  float: right;\n",
              "  font-size: smaller;\n",
              "  line-height: 1em;\n",
              "  font-family: monospace;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  border-radius: 1em;\n",
              "  height: 1em;\n",
              "  width: 1em;\n",
              "  text-decoration: none !important;\n",
              "  margin-left: 0.5em;\n",
              "  text-align: center;\n",
              "  /* unfitted */\n",
              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-unfitted-level-1);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link.fitted,\n",
              "a:link.sk-estimator-doc-link.fitted,\n",
              "a:visited.sk-estimator-doc-link.fitted {\n",
              "  /* fitted */\n",
              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-fitted-level-1);\n",
              "}\n",
              "\n",
              "/* On hover */\n",
              "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
              ".sk-estimator-doc-link:hover,\n",
              "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
              ".sk-estimator-doc-link:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
              ".sk-estimator-doc-link.fitted:hover,\n",
              "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
              ".sk-estimator-doc-link.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "/* Span, style for the box shown on hovering the info icon */\n",
              ".sk-estimator-doc-link span {\n",
              "  display: none;\n",
              "  z-index: 9999;\n",
              "  position: relative;\n",
              "  font-weight: normal;\n",
              "  right: .2ex;\n",
              "  padding: .5ex;\n",
              "  margin: .5ex;\n",
              "  width: min-content;\n",
              "  min-width: 20ex;\n",
              "  max-width: 50ex;\n",
              "  color: var(--sklearn-color-text);\n",
              "  box-shadow: 2pt 2pt 4pt #999;\n",
              "  /* unfitted */\n",
              "  background: var(--sklearn-color-unfitted-level-0);\n",
              "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link.fitted span {\n",
              "  /* fitted */\n",
              "  background: var(--sklearn-color-fitted-level-0);\n",
              "  border: var(--sklearn-color-fitted-level-3);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link:hover span {\n",
              "  display: block;\n",
              "}\n",
              "\n",
              "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
              "\n",
              "#sk-container-id-2 a.estimator_doc_link {\n",
              "  float: right;\n",
              "  font-size: 1rem;\n",
              "  line-height: 1em;\n",
              "  font-family: monospace;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  border-radius: 1rem;\n",
              "  height: 1rem;\n",
              "  width: 1rem;\n",
              "  text-decoration: none;\n",
              "  /* unfitted */\n",
              "  color: var(--sklearn-color-unfitted-level-1);\n",
              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 a.estimator_doc_link.fitted {\n",
              "  /* fitted */\n",
              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-fitted-level-1);\n",
              "}\n",
              "\n",
              "/* On hover */\n",
              "#sk-container-id-2 a.estimator_doc_link:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "#sk-container-id-2 a.estimator_doc_link.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-3);\n",
              "}\n",
              "</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=StratifiedKFold(n_splits=5, random_state=None, shuffle=False),\n",
              "             estimator=RandomForestClassifier(random_state=42),\n",
              "             param_grid={&#x27;max_depth&#x27;: [None, 3, 5, 7],\n",
              "                         &#x27;max_features&#x27;: [&#x27;sqrt&#x27;, &#x27;log2&#x27;],\n",
              "                         &#x27;n_estimators&#x27;: [50, 100, 150, 200]},\n",
              "             scoring=&#x27;accuracy&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>GridSearchCV</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.6/modules/generated/sklearn.model_selection.GridSearchCV.html\">?<span>Documentation for GridSearchCV</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\"><pre>GridSearchCV(cv=StratifiedKFold(n_splits=5, random_state=None, shuffle=False),\n",
              "             estimator=RandomForestClassifier(random_state=42),\n",
              "             param_grid={&#x27;max_depth&#x27;: [None, 3, 5, 7],\n",
              "                         &#x27;max_features&#x27;: [&#x27;sqrt&#x27;, &#x27;log2&#x27;],\n",
              "                         &#x27;n_estimators&#x27;: [50, 100, 150, 200]},\n",
              "             scoring=&#x27;accuracy&#x27;)</pre></div> </div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>best_estimator_: RandomForestClassifier</div></div></label><div class=\"sk-toggleable__content fitted\"><pre>RandomForestClassifier(random_state=42)</pre></div> </div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>RandomForestClassifier</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.6/modules/generated/sklearn.ensemble.RandomForestClassifier.html\">?<span>Documentation for RandomForestClassifier</span></a></div></label><div class=\"sk-toggleable__content fitted\"><pre>RandomForestClassifier(random_state=42)</pre></div> </div></div></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search.best_score_  # mean cross-validated score of the best_estimator"
      ],
      "metadata": {
        "id": "iCYkaUdq5Rf8",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "115a0e74-427c-428a-85d9-8ae504a31069"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8404761904761905"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "grid_search.best_params_ # parameter setting that gave the best results"
      ],
      "metadata": {
        "id": "rXPNSecT5aZ2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ef72e20d-e15b-4f68-9f45-52c737da7eed"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'max_depth': None, 'max_features': 'sqrt', 'n_estimators': 100}"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "best_model = grid_search.best_estimator_"
      ],
      "metadata": {
        "id": "amj0YeA65iTC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Test\n",
        "\n",
        "We finally use the developed model to predict the activity of the molecules in the test set. We will thus evaluate the performance of the model on unseen data (this data have not been used during feature selection, hypesparameter tuning and training). Remember that also for the test set we consider only the features we selected with feature selection (otherwise you receive an error because the model have been trained with some features and then has to predict using other features)."
      ],
      "metadata": {
        "id": "gywBJbKpZ94V"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = best_model.predict(X_test_selected)"
      ],
      "metadata": {
        "id": "EDudTni7UdWh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "accuracy_score(y_test, y_pred)"
      ],
      "metadata": {
        "id": "ETiEUkcYU6Bn",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a9a97dad-d087-439e-b735-0930b293c64d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8409090909090909"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import precision_score, recall_score, f1_score\n",
        "print('precision: ', precision_score(y_test, y_pred))\n",
        "print('recall: ', recall_score(y_test, y_pred))\n",
        "print('f1: ', f1_score(y_test, y_pred))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EtBx5wFzT3hf",
        "outputId": "00e1567b-15af-4596-be3b-c59a7b6d66a7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "precision:  0.7777777777777778\n",
            "recall:  0.9545454545454546\n",
            "f1:  0.8571428571428571\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Virtual screening\n",
        "Now we are going to use the best-performing model (in our case, the MEAN model) to predict the activity or inactivity of the molecules of the screening library."
      ],
      "metadata": {
        "id": "cVDbTfCL6VcI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "screening_df = pd.read_csv(r'/content/Antibacterial_focused_library_RESCORE_MEAN.csv', sep=';', decimal=',')\n",
        "# if .xlsx file, uncomment the cell below:\n",
        "# df = pd.read_excel(r'/content/Antibacterial_focused_library_RESCORE_MEAN.xlsx')"
      ],
      "metadata": {
        "id": "qZwZnhg66muN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "screening_df.set_index('Name', inplace=True)"
      ],
      "metadata": {
        "id": "whiVo9Ns66Ae"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "screening_filtered = screening_df.loc[:, selected_features]"
      ],
      "metadata": {
        "id": "waZN-vtS7g7u"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y_screening = best_model.predict(screening_filtered)"
      ],
      "metadata": {
        "id": "Eihuko5k6gzu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "screening_df['Prediction'] = y_screening"
      ],
      "metadata": {
        "id": "IjewMHs67H6y"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "screening_df['Prediction'].sum()"
      ],
      "metadata": {
        "id": "GsOGWnu49Jv6",
        "outputId": "fe8fa1d0-9bbd-4ddc-cb2a-7a6a3deb897d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "612"
            ]
          },
          "metadata": {},
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "screening_df['Prediction'].to_csv(r'/content/Antibacterial_focused_library_PREDICTIONS.csv', sep=';', decimal=',')"
      ],
      "metadata": {
        "id": "heHSAkMd99IB"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}