Feature_Selection_Wrapper_Methods.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/saleemhamo/ecg-data-feature-engineering/blob/main/Feature_Selection_Wrapper_Methods.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8nHSOFZWNTNv"
      },
      "source": [
        "### Imports"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Kqhad1DbNODg"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "from sklearn.feature_selection import SelectKBest\n",
        "from sklearn.feature_selection import chi2, f_classif, mutual_info_classif\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.preprocessing import KBinsDiscretizer\n",
        "from scipy.stats import pearsonr\n",
        "from sklearn.model_selection import LeaveOneGroupOut\n",
        "import json\n",
        "import matplotlib.pyplot as plt"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Mz-mppveFfZH"
      },
      "source": [
        "### Prepare Data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "juY6_HvUFl5u",
        "outputId": "d0d25af4-0e5f-495d-b694-d8ef023a38f2"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n",
            "/content/drive/MyDrive/University/AI & ML - Case Study 3\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "%cd /content/drive/MyDrive/University/AI & ML - Case Study 3/"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9C1O5nb1s7Ov"
      },
      "outputs": [],
      "source": [
        "feature_names = pd.read_csv('feature_names.csv').columns.tolist()\n",
        "labels = pd.read_csv('labels.csv', names=['label'])\n",
        "df_data = pd.read_csv('data.csv', header=None, names=feature_names)\n",
        "data = df_data.values.tolist()\n",
        "\n",
        "X = df_data\n",
        "y = labels"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I8ycaCWNYytw"
      },
      "source": [
        "#Wrapper Methods"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OAe0TfP8JXzz"
      },
      "source": [
        "##Backward Elimination"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4ruLCslJJ3dQ"
      },
      "source": [
        "###Random Forest"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q3IfQhhGMe36",
        "outputId": "677764ef-f210-4200-a8b4-15c4f2fa87e0"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "### Backward Feature Elimination for Random Forest ###\n",
            "Initial Performance with all features: 0.8222222222222223\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "ERROR:root:Internal Python error in the inspect module.\n",
            "Below is the traceback from this internal error.\n",
            "\n",
            "ERROR:root:Internal Python error in the inspect module.\n",
            "Below is the traceback from this internal error.\n",
            "\n",
            "ERROR:root:Internal Python error in the inspect module.\n",
            "Below is the traceback from this internal error.\n",
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\n",
            "    exec(code_obj, self.user_global_ns, self.user_ns)\n",
            "  File \"<ipython-input-37-0c3b8f71388c>\", line 28, in <cell line: 26>\n",
            "    results = Parallel(n_jobs=-1)(delayed(evaluate_feature_set)([f for f in features if f != feature]) for feature in features)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1952, in __call__\n",
            "    return output if self.return_generator else list(output)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1595, in _get_outputs\n",
            "    yield from self._retrieve()\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1707, in _retrieve\n",
            "    time.sleep(0.01)\n",
            "KeyboardInterrupt\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'KeyboardInterrupt' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1101, in get_records\n",
            "    return _fixed_getinnerframes(etb, number_of_lines_of_context, tb_offset)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 248, in wrapped\n",
            "    return f(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 281, in _fixed_getinnerframes\n",
            "    records = fix_frame_records_filenames(inspect.getinnerframes(etb, context))\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1662, in getinnerframes\n",
            "    frameinfo = (tb.tb_frame,) + getframeinfo(tb, context)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1620, in getframeinfo\n",
            "    filename = getsourcefile(frame) or getfile(frame)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 829, in getsourcefile\n",
            "    module = getmodule(object, filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 861, in getmodule\n",
            "    file = getabsfile(object, _filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 845, in getabsfile\n",
            "    return os.path.normcase(os.path.abspath(_filename))\n",
            "  File \"/usr/lib/python3.10/posixpath.py\", line 384, in abspath\n",
            "    cwd = os.getcwd()\n",
            "OSError: [Errno 107] Transport endpoint is not connected\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\n",
            "    exec(code_obj, self.user_global_ns, self.user_ns)\n",
            "  File \"<ipython-input-37-0c3b8f71388c>\", line 28, in <cell line: 26>\n",
            "    results = Parallel(n_jobs=-1)(delayed(evaluate_feature_set)([f for f in features if f != feature]) for feature in features)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1952, in __call__\n",
            "    return output if self.return_generator else list(output)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1595, in _get_outputs\n",
            "    yield from self._retrieve()\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1707, in _retrieve\n",
            "    time.sleep(0.01)\n",
            "KeyboardInterrupt\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'KeyboardInterrupt' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3473, in run_ast_nodes\n",
            "    if (await self.run_code(code, result,  async_=asy)):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3575, in run_code\n",
            "    self.showtraceback(running_compiled_code=True)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2101, in showtraceback\n",
            "    stb = self.InteractiveTB.structured_traceback(etype,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1367, in structured_traceback\n",
            "    return FormattedTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1267, in structured_traceback\n",
            "    return VerboseTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1124, in structured_traceback\n",
            "    formatted_exception = self.format_exception_as_a_whole(etype, evalue, etb, number_of_lines_of_context,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1082, in format_exception_as_a_whole\n",
            "    last_unique, recursion_repeat = find_recursion(orig_etype, evalue, records)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 382, in find_recursion\n",
            "    return len(records), 0\n",
            "TypeError: object of type 'NoneType' has no len()\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'TypeError' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1101, in get_records\n",
            "    return _fixed_getinnerframes(etb, number_of_lines_of_context, tb_offset)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 248, in wrapped\n",
            "    return f(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 281, in _fixed_getinnerframes\n",
            "    records = fix_frame_records_filenames(inspect.getinnerframes(etb, context))\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1662, in getinnerframes\n",
            "    frameinfo = (tb.tb_frame,) + getframeinfo(tb, context)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1620, in getframeinfo\n",
            "    filename = getsourcefile(frame) or getfile(frame)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 829, in getsourcefile\n",
            "    module = getmodule(object, filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 861, in getmodule\n",
            "    file = getabsfile(object, _filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 845, in getabsfile\n",
            "    return os.path.normcase(os.path.abspath(_filename))\n",
            "  File \"/usr/lib/python3.10/posixpath.py\", line 384, in abspath\n",
            "    cwd = os.getcwd()\n",
            "OSError: [Errno 107] Transport endpoint is not connected\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\n",
            "    exec(code_obj, self.user_global_ns, self.user_ns)\n",
            "  File \"<ipython-input-37-0c3b8f71388c>\", line 28, in <cell line: 26>\n",
            "    results = Parallel(n_jobs=-1)(delayed(evaluate_feature_set)([f for f in features if f != feature]) for feature in features)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1952, in __call__\n",
            "    return output if self.return_generator else list(output)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1595, in _get_outputs\n",
            "    yield from self._retrieve()\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1707, in _retrieve\n",
            "    time.sleep(0.01)\n",
            "KeyboardInterrupt\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'KeyboardInterrupt' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3473, in run_ast_nodes\n",
            "    if (await self.run_code(code, result,  async_=asy)):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3575, in run_code\n",
            "    self.showtraceback(running_compiled_code=True)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2101, in showtraceback\n",
            "    stb = self.InteractiveTB.structured_traceback(etype,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1367, in structured_traceback\n",
            "    return FormattedTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1267, in structured_traceback\n",
            "    return VerboseTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1124, in structured_traceback\n",
            "    formatted_exception = self.format_exception_as_a_whole(etype, evalue, etb, number_of_lines_of_context,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1082, in format_exception_as_a_whole\n",
            "    last_unique, recursion_repeat = find_recursion(orig_etype, evalue, records)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 382, in find_recursion\n",
            "    return len(records), 0\n",
            "TypeError: object of type 'NoneType' has no len()\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'TypeError' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3030, in _run_cell\n",
            "    return runner(coro)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/async_helpers.py\", line 78, in _pseudo_sync_runner\n",
            "    coro.send(None)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3257, in run_cell_async\n",
            "    has_raised = await self.run_ast_nodes(code_ast.body, cell_name,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3492, in run_ast_nodes\n",
            "    self.showtraceback()\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2101, in showtraceback\n",
            "    stb = self.InteractiveTB.structured_traceback(etype,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1367, in structured_traceback\n",
            "    return FormattedTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1267, in structured_traceback\n",
            "    return VerboseTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1142, in structured_traceback\n",
            "    formatted_exceptions += self.format_exception_as_a_whole(etype, evalue, etb, lines_of_context,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1082, in format_exception_as_a_whole\n",
            "    last_unique, recursion_repeat = find_recursion(orig_etype, evalue, records)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 382, in find_recursion\n",
            "    return len(records), 0\n",
            "TypeError: object of type 'NoneType' has no len()\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'TypeError' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1101, in get_records\n",
            "    return _fixed_getinnerframes(etb, number_of_lines_of_context, tb_offset)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 248, in wrapped\n",
            "    return f(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 281, in _fixed_getinnerframes\n",
            "    records = fix_frame_records_filenames(inspect.getinnerframes(etb, context))\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1662, in getinnerframes\n",
            "    frameinfo = (tb.tb_frame,) + getframeinfo(tb, context)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1620, in getframeinfo\n",
            "    filename = getsourcefile(frame) or getfile(frame)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 829, in getsourcefile\n",
            "    module = getmodule(object, filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 861, in getmodule\n",
            "    file = getabsfile(object, _filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 845, in getabsfile\n",
            "    return os.path.normcase(os.path.abspath(_filename))\n",
            "  File \"/usr/lib/python3.10/posixpath.py\", line 384, in abspath\n",
            "    cwd = os.getcwd()\n",
            "OSError: [Errno 107] Transport endpoint is not connected\n"
          ]
        }
      ],
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.model_selection import cross_val_score, LeaveOneGroupOut\n",
        "from joblib import Parallel, delayed\n",
        "import numpy as np\n",
        "\n",
        "# Define the groups based on your sample grouping (every 10 rows)\n",
        "groups = np.repeat(np.arange(len(X)//10), 10)\n",
        "\n",
        "# Initialize feature set F to include all features\n",
        "features = X.columns.tolist()\n",
        "\n",
        "# Initialize Random Forest classifier\n",
        "rf_clf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "\n",
        "print(\"\\n### Backward Feature Elimination for Random Forest ###\")\n",
        "\n",
        "# Evaluate performance with the initial feature set F\n",
        "initial_score = np.mean(cross_val_score(rf_clf, X[features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "print(f\"Initial Performance with all features: {initial_score}\")\n",
        "\n",
        "def evaluate_feature_set(feature_set):\n",
        "    score = np.mean(cross_val_score(rf_clf, X[feature_set], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "    return feature_set, score\n",
        "\n",
        "# Backward feature elimination\n",
        "while len(features) > 1:  # Continue until only one feature is left\n",
        "    # Evaluate performance for each feature set in parallel\n",
        "    results = Parallel(n_jobs=-1)(delayed(evaluate_feature_set)([f for f in features if f != feature]) for feature in features)\n",
        "\n",
        "    # Find the feature that maximally maintains or increases performance\n",
        "    best_feature, best_score = max(results, key=lambda x: x[1])\n",
        "\n",
        "    if best_score >= initial_score:\n",
        "        # Update feature set to exclude the feature that maximally maintains or increases performance\n",
        "        features.remove(best_feature)\n",
        "        initial_score = best_score\n",
        "        print(f\"Removed feature: {best_feature}, Updated Performance: {initial_score}\")\n",
        "    else:\n",
        "        # Break if performance degrades\n",
        "        print(\"Performance degraded. Stopping.\")\n",
        "        break\n",
        "\n",
        "# Resulting features after backward feature elimination for Random Forest\n",
        "print(\"Final selected features for Random Forest:\", features)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1DtYKJUyKO_B"
      },
      "source": [
        "###KNN"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "IQuUU1OpKQYP"
      },
      "outputs": [],
      "source": [
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.model_selection import cross_val_score, LeaveOneGroupOut\n",
        "import numpy as np\n",
        "\n",
        "# Define the groups based on your sample grouping (every 10 rows)\n",
        "groups = np.repeat(np.arange(len(X)//10), 10)\n",
        "\n",
        "# Initialize feature set F to include all features\n",
        "features = X.columns.tolist()\n",
        "\n",
        "# Initialize KNN classifier\n",
        "knn_clf = KNeighborsClassifier(n_neighbors=5)\n",
        "\n",
        "print(\"\\n### Backward Feature Elimination for KNN ###\")\n",
        "\n",
        "# Evaluate performance with the initial feature set F\n",
        "initial_score = np.mean(cross_val_score(knn_clf, X[features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "print(f\"Initial Performance with all features: {initial_score}\")\n",
        "\n",
        "# Backward feature elimination\n",
        "while len(features) > 1:  # Continue until only one feature is left\n",
        "    scores = []\n",
        "    for feature in features:\n",
        "        # Evaluate performance with feature sets F\\fi\n",
        "        current_features = [f for f in features if f != feature]\n",
        "        score = np.mean(cross_val_score(knn_clf, X[current_features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "        scores.append((feature, score))\n",
        "\n",
        "    # Find the feature that maximally maintains or increases performance\n",
        "    best_feature, best_score = max(scores, key=lambda x: x[1])\n",
        "\n",
        "    if best_score >= initial_score:\n",
        "        # Update feature set to exclude the feature that maximally maintains or increases performance\n",
        "        features.remove(best_feature)\n",
        "        initial_score = best_score\n",
        "        print(f\"Removed feature: {best_feature}, Updated Performance: {initial_score}\")\n",
        "    else:\n",
        "        # Break if performance degrades\n",
        "        print(\"Performance degraded. Stopping.\")\n",
        "        break\n",
        "\n",
        "# Resulting features after backward feature elimination for KNN\n",
        "print(\"Final selected features for KNN:\", features)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bD3-3ZJsKUxu"
      },
      "source": [
        "###SVM"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZO7G2CiRKX64",
        "outputId": "44fc397d-ed7b-42ce-8922-f64aba8a4e08"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "### Backward Feature Elimination for SVM ###\n",
            "Initial Performance with all features: 0.8722222222222222\n",
            "Removed feature: ratio_theta_31, Updated Performance: 0.8888888888888888\n",
            "Removed feature: alpha_ec_0, Updated Performance: 0.8888888888888888\n",
            "Removed feature: alpha_ec_1, Updated Performance: 0.8888888888888888\n",
            "Removed feature: alpha_ec_4, Updated Performance: 0.8888888888888888\n",
            "Removed feature: ratio_alpha_11, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_2, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_5, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_3, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_8, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_9, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_7, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_10, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_12, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_13, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_15, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_16, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_17, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_6, Updated Performance: 0.8944444444444445\n",
            "Removed feature: alpha_ec_20, Updated Performance: 0.8944444444444445\n",
            "Removed feature: ratio_theta_32, Updated Performance: 0.9000000000000001\n",
            "Removed feature: ratio_theta_38, Updated Performance: 0.9055555555555554\n",
            "Removed feature: alpha_ec_14, Updated Performance: 0.9055555555555554\n",
            "Removed feature: alpha_ec_18, Updated Performance: 0.9055555555555554\n",
            "Removed feature: ratio_alpha_47, Updated Performance: 0.911111111111111\n",
            "Removed feature: alpha_ec_23, Updated Performance: 0.911111111111111\n",
            "Removed feature: alpha_ec_24, Updated Performance: 0.911111111111111\n",
            "Removed feature: alpha_ec_26, Updated Performance: 0.911111111111111\n",
            "Removed feature: ratio_beta_24, Updated Performance: 0.9166666666666666\n",
            "Removed feature: alpha_ec_11, Updated Performance: 0.9166666666666666\n",
            "Removed feature: alpha_ec_27, Updated Performance: 0.9166666666666666\n",
            "Removed feature: alpha_ec_28, Updated Performance: 0.9166666666666666\n",
            "Removed feature: alpha_ec_29, Updated Performance: 0.9166666666666666\n",
            "Removed feature: alpha_ec_32, Updated Performance: 0.9166666666666666\n",
            "Removed feature: alpha_ec_33, Updated Performance: 0.9166666666666666\n",
            "Removed feature: alpha_ec_34, Updated Performance: 0.9166666666666666\n",
            "Removed feature: alpha_ec_35, Updated Performance: 0.9166666666666666\n"
          ]
        }
      ],
      "source": [
        "from sklearn.svm import SVC\n",
        "from sklearn.model_selection import cross_val_score, LeaveOneGroupOut\n",
        "import numpy as np\n",
        "\n",
        "# Define the groups based on your sample grouping (every 10 rows)\n",
        "groups = np.repeat(np.arange(len(X)//10), 10)\n",
        "\n",
        "# Initialize feature set F to include all features\n",
        "features = X.columns.tolist()\n",
        "\n",
        "# Initialize SVM classifier\n",
        "svm_clf = SVC(kernel='linear', C=1)\n",
        "\n",
        "print(\"\\n### Backward Feature Elimination for SVM ###\")\n",
        "\n",
        "# Evaluate performance with the initial feature set F\n",
        "initial_score = np.mean(cross_val_score(svm_clf, X[features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "print(f\"Initial Performance with all features: {initial_score}\")\n",
        "\n",
        "# Backward feature elimination\n",
        "while len(features) > 1:  # Continue until only one feature is left\n",
        "    scores = []\n",
        "    for feature in features:\n",
        "        # Evaluate performance with feature sets F\\fi\n",
        "        current_features = [f for f in features if f != feature]\n",
        "        score = np.mean(cross_val_score(svm_clf, X[current_features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "        scores.append((feature, score))\n",
        "\n",
        "    # Find the feature that maximally maintains or increases performance\n",
        "    best_feature, best_score = max(scores, key=lambda x: x[1])\n",
        "\n",
        "    if best_score >= initial_score:\n",
        "        # Update feature set to exclude the feature that maximally maintains or increases performance\n",
        "        features.remove(best_feature)\n",
        "        initial_score = best_score\n",
        "        print(f\"Removed feature: {best_feature}, Updated Performance: {initial_score}\")\n",
        "    else:\n",
        "        # Break if performance degrades\n",
        "        print(\"Performance degraded. Stopping.\")\n",
        "        break\n",
        "\n",
        "# Resulting features after backward feature elimination for SVM\n",
        "print(\"Final selected features for SVM:\", features)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "00fIYY1pKj3M"
      },
      "source": [
        "## Forward Feature Selection"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MbRWNzq_K1V2"
      },
      "source": [
        "###Random Forest"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "4eLhr7B-Kyq-"
      },
      "outputs": [],
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.model_selection import cross_val_score, LeaveOneGroupOut\n",
        "import numpy as np\n",
        "\n",
        "# Define the groups based on your sample grouping (every 10 rows)\n",
        "groups = np.repeat(np.arange(len(X)//10), 10)\n",
        "\n",
        "# Initialize feature set F to include all features\n",
        "features = []\n",
        "\n",
        "# Initialize Random Forest classifier\n",
        "rf_clf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "\n",
        "print(\"\\n### Forward Feature Selection for Random Forest ###\")\n",
        "\n",
        "# Iterate over features\n",
        "while len(features) < len(X.columns):  # Continue until all features are selected\n",
        "    scores = []\n",
        "    for feature in X.columns:\n",
        "        if feature not in features:\n",
        "            # Evaluate performance with the current feature set F\n",
        "            current_features = features + [feature]\n",
        "            score = np.mean(cross_val_score(rf_clf, X[current_features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "            scores.append((feature, score))\n",
        "\n",
        "    # Find the feature that maximally improves performance\n",
        "    best_feature, best_score = max(scores, key=lambda x: x[1])\n",
        "\n",
        "    # Update feature set to include the feature that maximally improves performance\n",
        "    features.append(best_feature)\n",
        "    print(f\"Added feature: {best_feature}, Updated Performance: {best_score}\")\n",
        "\n",
        "# Resulting features after forward feature selection for Random Forest\n",
        "print(\"Final selected features for Random Forest:\", features)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R_AK_Kd5K2cY"
      },
      "source": [
        "###KNN"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "nZ1xsl7FK4BK"
      },
      "outputs": [],
      "source": [
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.model_selection import cross_val_score, LeaveOneGroupOut\n",
        "import numpy as np\n",
        "\n",
        "# Define the groups based on your sample grouping (every 10 rows)\n",
        "groups = np.repeat(np.arange(len(X)//10), 10)\n",
        "\n",
        "# Initialize feature set F to include all features\n",
        "features = []\n",
        "\n",
        "# Initialize KNN classifier\n",
        "knn_clf = KNeighborsClassifier(n_neighbors=5)\n",
        "\n",
        "print(\"\\n### Forward Feature Selection for KNN ###\")\n",
        "\n",
        "# Iterate over features\n",
        "while len(features) < len(X.columns):  # Continue until all features are selected\n",
        "    scores = []\n",
        "    for feature in X.columns:\n",
        "        if feature not in features:\n",
        "            # Evaluate performance with the current feature set F\n",
        "            current_features = features + [feature]\n",
        "            score = np.mean(cross_val_score(knn_clf, X[current_features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "            scores.append((feature, score))\n",
        "\n",
        "    # Find the feature that maximally improves performance\n",
        "    best_feature, best_score = max(scores, key=lambda x: x[1])\n",
        "\n",
        "    # Update feature set to include the feature that maximally improves performance\n",
        "    features.append(best_feature)\n",
        "    print(f\"Added feature: {best_feature}, Updated Performance: {best_score}\")\n",
        "\n",
        "# Resulting features after forward feature selection for KNN\n",
        "print(\"Final selected features for KNN:\", features)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tzcIyKhxK6Az"
      },
      "source": [
        "###SVM"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "n9-_DIfYK8lg",
        "outputId": "72439b37-4381-4602-f01a-8a351236930d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "### Forward Feature Selection for SVM ###\n",
            "Added feature: ratio_beta_37, Updated Performance: 0.6611111111111112\n",
            "Added feature: ratio_beta_34, Updated Performance: 0.711111111111111\n",
            "Added feature: ratio_beta_44, Updated Performance: 0.7611111111111111\n",
            "Added feature: alpha_eo_9, Updated Performance: 0.7888888888888889\n",
            "Added feature: alpha_ec_11, Updated Performance: 0.8\n",
            "Added feature: beta_ec_6, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_0, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_10, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_14, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_12, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_13, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_32, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_43, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_1, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_2, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_27, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_3, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_8, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_15, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_17, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_21, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_29, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_33, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_34, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_24, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_35, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_36, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_41, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_44, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_42, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_47, Updated Performance: 0.8111111111111112\n",
            "Added feature: theta_ec_6, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_37, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_43, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_23, Updated Performance: 0.8111111111111112\n",
            "Added feature: theta_ec_13, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_eo_2, Updated Performance: 0.8111111111111112\n",
            "Added feature: theta_ec_7, Updated Performance: 0.8111111111111112\n",
            "Added feature: theta_ec_10, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_26, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_10, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_16, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_22, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_13, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_4, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_12, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_20, Updated Performance: 0.8111111111111112\n",
            "Added feature: beta_ec_9, Updated Performance: 0.8111111111111112\n",
            "Added feature: alpha_ec_24, Updated Performance: 0.8111111111111112\n",
            "Added feature: ratio_beta_21, Updated Performance: 0.8166666666666668\n",
            "Added feature: ratio_alpha_6, Updated Performance: 0.8333333333333335\n",
            "Added feature: alpha_eo_31, Updated Performance: 0.8500000000000001\n",
            "Added feature: beta_eo_26, Updated Performance: 0.8722222222222222\n",
            "Added feature: ratio_alpha_35, Updated Performance: 0.8777777777777778\n",
            "Added feature: theta_ec_32, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_0, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_18, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_8, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_37, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_9, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_15, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_19, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_36, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_47, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_40, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_20, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_14, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_27, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_32, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_39, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_28, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_38, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_45, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_46, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_eo_1, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_7, Updated Performance: 0.8944444444444445\n",
            "Added feature: theta_ec_36, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_eo_20, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_eo_15, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_38, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_25, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_40, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_5, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_11, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_1, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_19, Updated Performance: 0.8944444444444445\n",
            "Added feature: theta_ec_8, Updated Performance: 0.8944444444444445\n",
            "Added feature: theta_ec_17, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_39, Updated Performance: 0.8944444444444445\n",
            "Added feature: beta_ec_30, Updated Performance: 0.8944444444444445\n",
            "Added feature: alpha_ec_35, Updated Performance: 0.8944444444444445\n",
            "Added feature: theta_ec_4, Updated Performance: 0.8944444444444445\n",
            "Added feature: theta_ec_1, Updated Performance: 0.8944444444444445\n",
            "Added feature: theta_ec_3, Updated Performance: 0.8944444444444445\n",
            "Added feature: theta_eo_8, Updated Performance: 0.9000000000000001\n",
            "Added feature: ratio_beta_47, Updated Performance: 0.9055555555555554\n",
            "Added feature: alpha_ec_33, Updated Performance: 0.9166666666666666\n",
            "Added feature: alpha_ec_3, Updated Performance: 0.9166666666666666\n",
            "Added feature: theta_ec_5, Updated Performance: 0.9166666666666666\n",
            "Added feature: alpha_ec_2, Updated Performance: 0.9166666666666666\n",
            "Added feature: alpha_ec_44, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_21, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_26, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_46, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_ec_31, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_11, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_12, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_14, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_16, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_41, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_18, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_28, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_4, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_34, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_25, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_2, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_9, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_15, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_8, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_19, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_22, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_28, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_35, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_27, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_0, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_41, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_9, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_20, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_4, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_37, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_29, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_44, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_17, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_30, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_42, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_13, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_5, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_4, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_47, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_8, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_45, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_10, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_7, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_38, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_31, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_44, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_40, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_6, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_2, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_17, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_25, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_33, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_27, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_13, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_21, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_45, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_28, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_eo_16, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_0, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_25, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_12, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_43, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_30, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_42, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_23, Updated Performance: 0.9222222222222223\n",
            "Added feature: ratio_beta_12, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_46, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_24, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_18, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_25, Updated Performance: 0.9166666666666666\n",
            "Added feature: beta_eo_16, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_39, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_7, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_41, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_26, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_16, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_ec_45, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_18, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_36, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_6, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_27, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_46, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_12, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_42, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_ec_22, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_10, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_37, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_47, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_33, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_0, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_14, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_22, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_21, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_29, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_37, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_43, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_38, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_39, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_38, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_43, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_47, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_41, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_eo_6, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_33, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_3, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_35, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_36, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_5, Updated Performance: 0.9166666666666666\n",
            "Added feature: ratio_beta_20, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_44, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_28, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_11, Updated Performance: 0.9222222222222223\n",
            "Added feature: alpha_eo_23, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_15, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_45, Updated Performance: 0.9222222222222223\n",
            "Added feature: theta_ec_34, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_23, Updated Performance: 0.9222222222222223\n",
            "Added feature: beta_eo_31, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_ec_42, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_ec_5, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_eo_20, Updated Performance: 0.9277777777777779\n",
            "Added feature: beta_eo_11, Updated Performance: 0.9277777777777779\n",
            "Added feature: beta_eo_34, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_eo_19, Updated Performance: 0.9277777777777779\n",
            "Added feature: beta_ec_18, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_ec_23, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_ec_6, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_4, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_ec_29, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_11, Updated Performance: 0.9277777777777779\n",
            "Added feature: beta_eo_1, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_10, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_eo_24, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_ec_30, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_3, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_ec_7, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_eo_30, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_17, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_eo_32, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_18, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_15, Updated Performance: 0.9277777777777779\n",
            "Added feature: ratio_beta_1, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_46, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_47, Updated Performance: 0.9277777777777779\n",
            "Added feature: beta_eo_32, Updated Performance: 0.9277777777777779\n",
            "Added feature: ratio_alpha_2, Updated Performance: 0.9277777777777779\n",
            "Added feature: theta_eo_5, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_ec_31, Updated Performance: 0.9222222222222223\n",
            "Added feature: ratio_theta_42, Updated Performance: 0.9277777777777779\n",
            "Added feature: alpha_eo_3, Updated Performance: 0.9277777777777779\n",
            "Added feature: ratio_alpha_23, Updated Performance: 0.9333333333333333\n",
            "Added feature: ratio_alpha_18, Updated Performance: 0.9388888888888888\n",
            "Added feature: alpha_eo_26, Updated Performance: 0.9388888888888888\n",
            "Added feature: beta_eo_19, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_22, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_42, Updated Performance: 0.9388888888888888\n",
            "Added feature: ratio_theta_15, Updated Performance: 0.9388888888888888\n",
            "Added feature: alpha_eo_35, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_23, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_24, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_25, Updated Performance: 0.9388888888888888\n",
            "Added feature: alpha_ec_17, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_7, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_44, Updated Performance: 0.9388888888888888\n",
            "Added feature: ratio_beta_6, Updated Performance: 0.9333333333333333\n",
            "Added feature: theta_eo_31, Updated Performance: 0.9333333333333333\n",
            "Added feature: ratio_alpha_20, Updated Performance: 0.9388888888888888\n",
            "Added feature: beta_eo_46, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_2, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_20, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_0, Updated Performance: 0.9388888888888888\n",
            "Added feature: alpha_eo_29, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_9, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_38, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_21, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_30, Updated Performance: 0.9388888888888888\n",
            "Added feature: ratio_beta_7, Updated Performance: 0.9388888888888888\n",
            "Added feature: beta_eo_39, Updated Performance: 0.9388888888888888\n",
            "Added feature: ratio_beta_3, Updated Performance: 0.9388888888888888\n",
            "Added feature: ratio_beta_11, Updated Performance: 0.9388888888888888\n",
            "Added feature: ratio_alpha_11, Updated Performance: 0.9388888888888888\n",
            "Added feature: beta_eo_24, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_26, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_33, Updated Performance: 0.9388888888888888\n",
            "Added feature: alpha_eo_21, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_37, Updated Performance: 0.9388888888888888\n",
            "Added feature: theta_eo_39, Updated Performance: 0.9388888888888888\n",
            "Added feature: ratio_beta_45, Updated Performance: 0.9333333333333333\n",
            "Added feature: ratio_alpha_31, Updated Performance: 0.9388888888888888\n",
            "Added feature: ratio_beta_18, Updated Performance: 0.9444444444444444\n",
            "Added feature: ratio_alpha_8, Updated Performance: 0.9499999999999998\n",
            "Added feature: alpha_eo_14, Updated Performance: 0.9499999999999998\n",
            "Added feature: alpha_eo_34, Updated Performance: 0.9499999999999998\n",
            "Added feature: theta_eo_1, Updated Performance: 0.9499999999999998\n",
            "Added feature: theta_eo_12, Updated Performance: 0.9499999999999998\n",
            "Added feature: theta_eo_43, Updated Performance: 0.9499999999999998\n",
            "Added feature: ratio_alpha_4, Updated Performance: 0.9499999999999998\n",
            "Added feature: beta_eo_40, Updated Performance: 0.9499999999999998\n",
            "Added feature: theta_eo_13, Updated Performance: 0.9499999999999998\n",
            "Added feature: theta_eo_14, Updated Performance: 0.9499999999999998\n",
            "Added feature: theta_eo_35, Updated Performance: 0.9499999999999998\n",
            "Added feature: theta_eo_40, Updated Performance: 0.9499999999999998\n",
            "Added feature: ratio_alpha_0, Updated Performance: 0.9500000000000001\n",
            "Added feature: theta_eo_32, Updated Performance: 0.9500000000000001\n",
            "Added feature: theta_eo_19, Updated Performance: 0.9500000000000001\n",
            "Added feature: ratio_beta_15, Updated Performance: 0.9555555555555555\n",
            "Added feature: ratio_beta_17, Updated Performance: 0.9611111111111111\n",
            "Added feature: alpha_eo_22, Updated Performance: 0.9611111111111111\n",
            "Added feature: theta_eo_27, Updated Performance: 0.9611111111111111\n",
            "Added feature: alpha_ec_16, Updated Performance: 0.9611111111111111\n",
            "Added feature: theta_eo_34, Updated Performance: 0.9611111111111111\n",
            "Added feature: theta_eo_36, Updated Performance: 0.9611111111111111\n",
            "Added feature: theta_eo_41, Updated Performance: 0.9611111111111111\n",
            "Added feature: theta_eo_45, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_alpha_14, Updated Performance: 0.9611111111111111\n",
            "Added feature: alpha_eo_40, Updated Performance: 0.9611111111111111\n",
            "Added feature: theta_eo_28, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_alpha_16, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_beta_13, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_beta_26, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_beta_29, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_alpha_29, Updated Performance: 0.9611111111111111\n",
            "Added feature: theta_eo_29, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_beta_16, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_theta_34, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_alpha_13, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_beta_36, Updated Performance: 0.9666666666666666\n",
            "Added feature: ratio_beta_10, Updated Performance: 0.9666666666666666\n",
            "Added feature: ratio_beta_0, Updated Performance: 0.9666666666666666\n",
            "Added feature: ratio_beta_22, Updated Performance: 0.9666666666666666\n",
            "Added feature: ratio_beta_28, Updated Performance: 0.9666666666666666\n",
            "Added feature: ratio_beta_27, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_beta_23, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_theta_23, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_alpha_17, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_theta_20, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_beta_30, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_alpha_15, Updated Performance: 0.9611111111111111\n",
            "Added feature: ratio_alpha_22, Updated Performance: 0.9555555555555557\n",
            "Added feature: ratio_alpha_7, Updated Performance: 0.9444444444444444\n",
            "Added feature: ratio_theta_7, Updated Performance: 0.9444444444444444\n",
            "Added feature: ratio_theta_9, Updated Performance: 0.9444444444444444\n",
            "Added feature: ratio_theta_16, Updated Performance: 0.9444444444444444\n",
            "Added feature: ratio_theta_6, Updated Performance: 0.9444444444444444\n",
            "Added feature: ratio_beta_25, Updated Performance: 0.9444444444444444\n",
            "Added feature: ratio_theta_4, Updated Performance: 0.9555555555555555\n",
            "Added feature: ratio_alpha_1, Updated Performance: 0.9555555555555555\n",
            "Added feature: ratio_theta_13, Updated Performance: 0.9500000000000001\n",
            "Added feature: ratio_beta_14, Updated Performance: 0.9444444444444444\n",
            "Added feature: ratio_beta_31, Updated Performance: 0.9499999999999998\n",
            "Added feature: ratio_beta_8, Updated Performance: 0.9499999999999998\n",
            "Added feature: ratio_theta_11, Updated Performance: 0.9499999999999998\n",
            "Added feature: ratio_theta_33, Updated Performance: 0.9499999999999998\n",
            "Added feature: ratio_alpha_10, Updated Performance: 0.9499999999999998\n",
            "Added feature: ratio_beta_2, Updated Performance: 0.9499999999999998\n",
            "Added feature: ratio_beta_9, Updated Performance: 0.9499999999999998\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "ERROR:root:Internal Python error in the inspect module.\n",
            "Below is the traceback from this internal error.\n",
            "\n",
            "ERROR:root:Internal Python error in the inspect module.\n",
            "Below is the traceback from this internal error.\n",
            "\n",
            "ERROR:root:Internal Python error in the inspect module.\n",
            "Below is the traceback from this internal error.\n",
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\n",
            "    exec(code_obj, self.user_global_ns, self.user_ns)\n",
            "  File \"<ipython-input-9-0deb7089d17a>\", line 23, in <cell line: 17>\n",
            "    score = np.mean(cross_val_score(svm_clf, X[current_features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 515, in cross_val_score\n",
            "    cv_results = cross_validate(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 266, in cross_validate\n",
            "    results = parallel(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/utils/parallel.py\", line 63, in __call__\n",
            "    return super().__call__(iterable_with_config)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1863, in __call__\n",
            "    return output if self.return_generator else list(output)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1792, in _get_sequential_output\n",
            "    res = func(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/utils/parallel.py\", line 123, in __call__\n",
            "    return self.function(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 686, in _fit_and_score\n",
            "    estimator.fit(X_train, y_train, **fit_params)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/svm/_base.py\", line 252, in fit\n",
            "    fit(X, y, sample_weight, solver_type, kernel, random_seed=seed)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/svm/_base.py\", line 331, in _dense_fit\n",
            "    ) = libsvm.fit(\n",
            "KeyboardInterrupt\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'KeyboardInterrupt' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1101, in get_records\n",
            "    return _fixed_getinnerframes(etb, number_of_lines_of_context, tb_offset)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 248, in wrapped\n",
            "    return f(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 281, in _fixed_getinnerframes\n",
            "    records = fix_frame_records_filenames(inspect.getinnerframes(etb, context))\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1662, in getinnerframes\n",
            "    frameinfo = (tb.tb_frame,) + getframeinfo(tb, context)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1620, in getframeinfo\n",
            "    filename = getsourcefile(frame) or getfile(frame)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 829, in getsourcefile\n",
            "    module = getmodule(object, filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 861, in getmodule\n",
            "    file = getabsfile(object, _filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 845, in getabsfile\n",
            "    return os.path.normcase(os.path.abspath(_filename))\n",
            "  File \"/usr/lib/python3.10/posixpath.py\", line 384, in abspath\n",
            "    cwd = os.getcwd()\n",
            "OSError: [Errno 107] Transport endpoint is not connected\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\n",
            "    exec(code_obj, self.user_global_ns, self.user_ns)\n",
            "  File \"<ipython-input-9-0deb7089d17a>\", line 23, in <cell line: 17>\n",
            "    score = np.mean(cross_val_score(svm_clf, X[current_features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 515, in cross_val_score\n",
            "    cv_results = cross_validate(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 266, in cross_validate\n",
            "    results = parallel(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/utils/parallel.py\", line 63, in __call__\n",
            "    return super().__call__(iterable_with_config)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1863, in __call__\n",
            "    return output if self.return_generator else list(output)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1792, in _get_sequential_output\n",
            "    res = func(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/utils/parallel.py\", line 123, in __call__\n",
            "    return self.function(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 686, in _fit_and_score\n",
            "    estimator.fit(X_train, y_train, **fit_params)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/svm/_base.py\", line 252, in fit\n",
            "    fit(X, y, sample_weight, solver_type, kernel, random_seed=seed)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/svm/_base.py\", line 331, in _dense_fit\n",
            "    ) = libsvm.fit(\n",
            "KeyboardInterrupt\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'KeyboardInterrupt' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3473, in run_ast_nodes\n",
            "    if (await self.run_code(code, result,  async_=asy)):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3575, in run_code\n",
            "    self.showtraceback(running_compiled_code=True)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2101, in showtraceback\n",
            "    stb = self.InteractiveTB.structured_traceback(etype,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1367, in structured_traceback\n",
            "    return FormattedTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1267, in structured_traceback\n",
            "    return VerboseTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1124, in structured_traceback\n",
            "    formatted_exception = self.format_exception_as_a_whole(etype, evalue, etb, number_of_lines_of_context,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1082, in format_exception_as_a_whole\n",
            "    last_unique, recursion_repeat = find_recursion(orig_etype, evalue, records)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 382, in find_recursion\n",
            "    return len(records), 0\n",
            "TypeError: object of type 'NoneType' has no len()\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'TypeError' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1101, in get_records\n",
            "    return _fixed_getinnerframes(etb, number_of_lines_of_context, tb_offset)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 248, in wrapped\n",
            "    return f(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 281, in _fixed_getinnerframes\n",
            "    records = fix_frame_records_filenames(inspect.getinnerframes(etb, context))\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1662, in getinnerframes\n",
            "    frameinfo = (tb.tb_frame,) + getframeinfo(tb, context)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1620, in getframeinfo\n",
            "    filename = getsourcefile(frame) or getfile(frame)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 829, in getsourcefile\n",
            "    module = getmodule(object, filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 861, in getmodule\n",
            "    file = getabsfile(object, _filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 845, in getabsfile\n",
            "    return os.path.normcase(os.path.abspath(_filename))\n",
            "  File \"/usr/lib/python3.10/posixpath.py\", line 384, in abspath\n",
            "    cwd = os.getcwd()\n",
            "OSError: [Errno 107] Transport endpoint is not connected\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\n",
            "    exec(code_obj, self.user_global_ns, self.user_ns)\n",
            "  File \"<ipython-input-9-0deb7089d17a>\", line 23, in <cell line: 17>\n",
            "    score = np.mean(cross_val_score(svm_clf, X[current_features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 515, in cross_val_score\n",
            "    cv_results = cross_validate(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 266, in cross_validate\n",
            "    results = parallel(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/utils/parallel.py\", line 63, in __call__\n",
            "    return super().__call__(iterable_with_config)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1863, in __call__\n",
            "    return output if self.return_generator else list(output)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/joblib/parallel.py\", line 1792, in _get_sequential_output\n",
            "    res = func(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/utils/parallel.py\", line 123, in __call__\n",
            "    return self.function(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/model_selection/_validation.py\", line 686, in _fit_and_score\n",
            "    estimator.fit(X_train, y_train, **fit_params)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/svm/_base.py\", line 252, in fit\n",
            "    fit(X, y, sample_weight, solver_type, kernel, random_seed=seed)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/sklearn/svm/_base.py\", line 331, in _dense_fit\n",
            "    ) = libsvm.fit(\n",
            "KeyboardInterrupt\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'KeyboardInterrupt' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3473, in run_ast_nodes\n",
            "    if (await self.run_code(code, result,  async_=asy)):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3575, in run_code\n",
            "    self.showtraceback(running_compiled_code=True)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2101, in showtraceback\n",
            "    stb = self.InteractiveTB.structured_traceback(etype,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1367, in structured_traceback\n",
            "    return FormattedTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1267, in structured_traceback\n",
            "    return VerboseTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1124, in structured_traceback\n",
            "    formatted_exception = self.format_exception_as_a_whole(etype, evalue, etb, number_of_lines_of_context,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1082, in format_exception_as_a_whole\n",
            "    last_unique, recursion_repeat = find_recursion(orig_etype, evalue, records)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 382, in find_recursion\n",
            "    return len(records), 0\n",
            "TypeError: object of type 'NoneType' has no len()\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'TypeError' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3030, in _run_cell\n",
            "    return runner(coro)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/async_helpers.py\", line 78, in _pseudo_sync_runner\n",
            "    coro.send(None)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3257, in run_cell_async\n",
            "    has_raised = await self.run_ast_nodes(code_ast.body, cell_name,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3492, in run_ast_nodes\n",
            "    self.showtraceback()\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2101, in showtraceback\n",
            "    stb = self.InteractiveTB.structured_traceback(etype,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1367, in structured_traceback\n",
            "    return FormattedTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1267, in structured_traceback\n",
            "    return VerboseTB.structured_traceback(\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1142, in structured_traceback\n",
            "    formatted_exceptions += self.format_exception_as_a_whole(etype, evalue, etb, lines_of_context,\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1082, in format_exception_as_a_whole\n",
            "    last_unique, recursion_repeat = find_recursion(orig_etype, evalue, records)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 382, in find_recursion\n",
            "    return len(records), 0\n",
            "TypeError: object of type 'NoneType' has no len()\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2099, in showtraceback\n",
            "    stb = value._render_traceback_()\n",
            "AttributeError: 'TypeError' object has no attribute '_render_traceback_'\n",
            "\n",
            "During handling of the above exception, another exception occurred:\n",
            "\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 1101, in get_records\n",
            "    return _fixed_getinnerframes(etb, number_of_lines_of_context, tb_offset)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 248, in wrapped\n",
            "    return f(*args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.10/dist-packages/IPython/core/ultratb.py\", line 281, in _fixed_getinnerframes\n",
            "    records = fix_frame_records_filenames(inspect.getinnerframes(etb, context))\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1662, in getinnerframes\n",
            "    frameinfo = (tb.tb_frame,) + getframeinfo(tb, context)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 1620, in getframeinfo\n",
            "    filename = getsourcefile(frame) or getfile(frame)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 829, in getsourcefile\n",
            "    module = getmodule(object, filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 861, in getmodule\n",
            "    file = getabsfile(object, _filename)\n",
            "  File \"/usr/lib/python3.10/inspect.py\", line 845, in getabsfile\n",
            "    return os.path.normcase(os.path.abspath(_filename))\n",
            "  File \"/usr/lib/python3.10/posixpath.py\", line 384, in abspath\n",
            "    cwd = os.getcwd()\n",
            "OSError: [Errno 107] Transport endpoint is not connected\n"
          ]
        }
      ],
      "source": [
        "from sklearn.svm import SVC\n",
        "from sklearn.model_selection import cross_val_score, LeaveOneGroupOut\n",
        "import numpy as np\n",
        "\n",
        "# Define the groups based on your sample grouping (every 10 rows)\n",
        "groups = np.repeat(np.arange(len(X)//10), 10)\n",
        "\n",
        "# Initialize feature set F to include all features\n",
        "features = []\n",
        "\n",
        "# Initialize SVM classifier\n",
        "svm_clf = SVC(kernel='linear', C=1)\n",
        "\n",
        "print(\"\\n### Forward Feature Selection for SVM ###\")\n",
        "\n",
        "# Iterate over features\n",
        "while len(features) < len(X.columns):  # Continue until all features are selected\n",
        "    scores = []\n",
        "    for feature in X.columns:\n",
        "        if feature not in features:\n",
        "            # Evaluate performance with the current feature set F\n",
        "            current_features = features + [feature]\n",
        "            score = np.mean(cross_val_score(svm_clf, X[current_features], y.values.ravel(), cv=LeaveOneGroupOut().split(X, y.values.ravel(), groups), error_score='raise'))\n",
        "            scores.append((feature, score))\n",
        "\n",
        "    # Find the feature that maximally improves performance\n",
        "    best_feature, best_score = max(scores, key=lambda x: x[1])\n",
        "\n",
        "    # Update feature set to include the feature that maximally improves performance\n",
        "    features.append(best_feature)\n",
        "    print(f\"Added feature: {best_feature}, Updated Performance: {best_score}\")\n",
        "\n",
        "# Resulting features after forward feature selection for SVM\n",
        "print(\"Final selected features for SVM:\", features)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import re\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Provided text\n",
        "text = \"\"\"\n",
        "Added feature: ratio_beta_37, Updated Performance: 0.6611111111111112\n",
        "Added feature: ratio_beta_34, Updated Performance: 0.711111111111111\n",
        "Added feature: ratio_beta_44, Updated Performance: 0.7611111111111111\n",
        "Added feature: alpha_eo_9, Updated Performance: 0.7888888888888889\n",
        "Added feature: alpha_ec_11, Updated Performance: 0.8\n",
        "Added feature: beta_ec_6, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_0, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_10, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_14, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_12, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_13, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_32, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_43, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_1, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_2, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_27, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_3, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_8, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_15, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_17, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_21, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_29, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_33, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_34, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_24, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_35, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_36, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_41, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_44, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_42, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_47, Updated Performance: 0.8111111111111112\n",
        "Added feature: theta_ec_6, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_37, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_43, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_23, Updated Performance: 0.8111111111111112\n",
        "Added feature: theta_ec_13, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_eo_2, Updated Performance: 0.8111111111111112\n",
        "Added feature: theta_ec_7, Updated Performance: 0.8111111111111112\n",
        "Added feature: theta_ec_10, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_26, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_10, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_16, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_22, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_13, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_4, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_12, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_20, Updated Performance: 0.8111111111111112\n",
        "Added feature: beta_ec_9, Updated Performance: 0.8111111111111112\n",
        "Added feature: alpha_ec_24, Updated Performance: 0.8111111111111112\n",
        "Added feature: ratio_beta_21, Updated Performance: 0.8166666666666668\n",
        "Added feature: ratio_alpha_6, Updated Performance: 0.8333333333333335\n",
        "Added feature: alpha_eo_31, Updated Performance: 0.8500000000000001\n",
        "Added feature: beta_eo_26, Updated Performance: 0.8722222222222222\n",
        "Added feature: ratio_alpha_35, Updated Performance: 0.8777777777777778\n",
        "Added feature: theta_ec_32, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_0, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_18, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_8, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_37, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_9, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_15, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_19, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_36, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_47, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_40, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_20, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_14, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_27, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_32, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_39, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_28, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_38, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_45, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_46, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_eo_1, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_7, Updated Performance: 0.8944444444444445\n",
        "Added feature: theta_ec_36, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_eo_20, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_eo_15, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_38, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_25, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_40, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_5, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_11, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_1, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_19, Updated Performance: 0.8944444444444445\n",
        "Added feature: theta_ec_8, Updated Performance: 0.8944444444444445\n",
        "Added feature: theta_ec_17, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_39, Updated Performance: 0.8944444444444445\n",
        "Added feature: beta_ec_30, Updated Performance: 0.8944444444444445\n",
        "Added feature: alpha_ec_35, Updated Performance: 0.8944444444444445\n",
        "Added feature: theta_ec_4, Updated Performance: 0.8944444444444445\n",
        "Added feature: theta_ec_1, Updated Performance: 0.8944444444444445\n",
        "Added feature: theta_ec_3, Updated Performance: 0.8944444444444445\n",
        "Added feature: theta_eo_8, Updated Performance: 0.9000000000000001\n",
        "Added feature: ratio_beta_47, Updated Performance: 0.9055555555555554\n",
        "Added feature: alpha_ec_33, Updated Performance: 0.9166666666666666\n",
        "Added feature: alpha_ec_3, Updated Performance: 0.9166666666666666\n",
        "Added feature: theta_ec_5, Updated Performance: 0.9166666666666666\n",
        "Added feature: alpha_ec_2, Updated Performance: 0.9166666666666666\n",
        "Added feature: alpha_ec_44, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_21, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_26, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_46, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_ec_31, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_11, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_12, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_14, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_16, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_41, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_18, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_28, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_4, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_34, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_25, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_2, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_9, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_15, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_8, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_19, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_22, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_28, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_35, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_27, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_0, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_41, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_9, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_20, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_4, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_37, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_29, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_44, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_17, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_30, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_42, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_13, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_5, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_4, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_47, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_8, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_45, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_10, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_7, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_38, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_31, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_44, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_40, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_6, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_2, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_17, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_25, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_33, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_27, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_13, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_21, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_45, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_28, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_eo_16, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_0, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_25, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_12, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_43, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_30, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_42, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_23, Updated Performance: 0.9222222222222223\n",
        "Added feature: ratio_beta_12, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_46, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_24, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_18, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_25, Updated Performance: 0.9166666666666666\n",
        "Added feature: beta_eo_16, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_39, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_7, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_41, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_26, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_16, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_ec_45, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_18, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_36, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_6, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_27, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_46, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_12, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_42, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_ec_22, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_10, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_37, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_47, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_33, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_0, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_14, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_22, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_21, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_29, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_37, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_43, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_38, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_39, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_38, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_43, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_47, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_41, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_eo_6, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_33, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_3, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_35, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_36, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_5, Updated Performance: 0.9166666666666666\n",
        "Added feature: ratio_beta_20, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_44, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_28, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_11, Updated Performance: 0.9222222222222223\n",
        "Added feature: alpha_eo_23, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_15, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_45, Updated Performance: 0.9222222222222223\n",
        "Added feature: theta_ec_34, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_23, Updated Performance: 0.9222222222222223\n",
        "Added feature: beta_eo_31, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_ec_42, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_ec_5, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_eo_20, Updated Performance: 0.9277777777777779\n",
        "Added feature: beta_eo_11, Updated Performance: 0.9277777777777779\n",
        "Added feature: beta_eo_34, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_eo_19, Updated Performance: 0.9277777777777779\n",
        "Added feature: beta_ec_18, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_ec_23, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_ec_6, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_4, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_ec_29, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_11, Updated Performance: 0.9277777777777779\n",
        "Added feature: beta_eo_1, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_10, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_eo_24, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_ec_30, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_3, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_ec_7, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_eo_30, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_17, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_eo_32, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_18, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_15, Updated Performance: 0.9277777777777779\n",
        "Added feature: ratio_beta_1, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_46, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_47, Updated Performance: 0.9277777777777779\n",
        "Added feature: beta_eo_32, Updated Performance: 0.9277777777777779\n",
        "Added feature: ratio_alpha_2, Updated Performance: 0.9277777777777779\n",
        "Added feature: theta_eo_5, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_ec_31, Updated Performance: 0.9222222222222223\n",
        "Added feature: ratio_theta_42, Updated Performance: 0.9277777777777779\n",
        "Added feature: alpha_eo_3, Updated Performance: 0.9277777777777779\n",
        "Added feature: ratio_alpha_23, Updated Performance: 0.9333333333333333\n",
        "Added feature: ratio_alpha_18, Updated Performance: 0.9388888888888888\n",
        "Added feature: alpha_eo_26, Updated Performance: 0.9388888888888888\n",
        "Added feature: beta_eo_19, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_22, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_42, Updated Performance: 0.9388888888888888\n",
        "Added feature: ratio_theta_15, Updated Performance: 0.9388888888888888\n",
        "Added feature: alpha_eo_35, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_23, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_24, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_25, Updated Performance: 0.9388888888888888\n",
        "Added feature: alpha_ec_17, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_7, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_44, Updated Performance: 0.9388888888888888\n",
        "Added feature: ratio_beta_6, Updated Performance: 0.9333333333333333\n",
        "Added feature: theta_eo_31, Updated Performance: 0.9333333333333333\n",
        "Added feature: ratio_alpha_20, Updated Performance: 0.9388888888888888\n",
        "Added feature: beta_eo_46, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_2, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_20, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_0, Updated Performance: 0.9388888888888888\n",
        "Added feature: alpha_eo_29, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_9, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_38, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_21, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_30, Updated Performance: 0.9388888888888888\n",
        "Added feature: ratio_beta_7, Updated Performance: 0.9388888888888888\n",
        "Added feature: beta_eo_39, Updated Performance: 0.9388888888888888\n",
        "Added feature: ratio_beta_3, Updated Performance: 0.9388888888888888\n",
        "Added feature: ratio_beta_11, Updated Performance: 0.9388888888888888\n",
        "Added feature: ratio_alpha_11, Updated Performance: 0.9388888888888888\n",
        "Added feature: beta_eo_24, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_26, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_33, Updated Performance: 0.9388888888888888\n",
        "Added feature: alpha_eo_21, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_37, Updated Performance: 0.9388888888888888\n",
        "Added feature: theta_eo_39, Updated Performance: 0.9388888888888888\n",
        "Added feature: ratio_beta_45, Updated Performance: 0.9333333333333333\n",
        "Added feature: ratio_alpha_31, Updated Performance: 0.9388888888888888\n",
        "Added feature: ratio_beta_18, Updated Performance: 0.9444444444444444\n",
        "Added feature: ratio_alpha_8, Updated Performance: 0.9499999999999998\n",
        "Added feature: alpha_eo_14, Updated Performance: 0.9499999999999998\n",
        "Added feature: alpha_eo_34, Updated Performance: 0.9499999999999998\n",
        "Added feature: theta_eo_1, Updated Performance: 0.9499999999999998\n",
        "Added feature: theta_eo_12, Updated Performance: 0.9499999999999998\n",
        "Added feature: theta_eo_43, Updated Performance: 0.9499999999999998\n",
        "Added feature: ratio_alpha_4, Updated Performance: 0.9499999999999998\n",
        "Added feature: beta_eo_40, Updated Performance: 0.9499999999999998\n",
        "Added feature: theta_eo_13, Updated Performance: 0.9499999999999998\n",
        "Added feature: theta_eo_14, Updated Performance: 0.9499999999999998\n",
        "Added feature: theta_eo_35, Updated Performance: 0.9499999999999998\n",
        "Added feature: theta_eo_40, Updated Performance: 0.9499999999999998\n",
        "Added feature: ratio_alpha_0, Updated Performance: 0.9500000000000001\n",
        "Added feature: theta_eo_32, Updated Performance: 0.9500000000000001\n",
        "Added feature: theta_eo_19, Updated Performance: 0.9500000000000001\n",
        "Added feature: ratio_beta_15, Updated Performance: 0.9555555555555555\n",
        "Added feature: ratio_beta_17, Updated Performance: 0.9611111111111111\n",
        "Added feature: alpha_eo_22, Updated Performance: 0.9611111111111111\n",
        "Added feature: theta_eo_27, Updated Performance: 0.9611111111111111\n",
        "Added feature: alpha_ec_16, Updated Performance: 0.9611111111111111\n",
        "Added feature: theta_eo_34, Updated Performance: 0.9611111111111111\n",
        "Added feature: theta_eo_36, Updated Performance: 0.9611111111111111\n",
        "Added feature: theta_eo_41, Updated Performance: 0.9611111111111111\n",
        "Added feature: theta_eo_45, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_alpha_14, Updated Performance: 0.9611111111111111\n",
        "Added feature: alpha_eo_40, Updated Performance: 0.9611111111111111\n",
        "Added feature: theta_eo_28, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_alpha_16, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_beta_13, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_beta_26, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_beta_29, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_alpha_29, Updated Performance: 0.9611111111111111\n",
        "Added feature: theta_eo_29, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_beta_16, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_theta_34, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_alpha_13, Updated Performance: 0.9611111111111111\n",
        "Added feature: ratio_beta_36, Updated Performance: 0.9666666666666666\n",
        "\"\"\"\n",
        "\n",
        "# Extracting names and feature names\n",
        "names = re.findall(r'Added feature: (\\w+_\\w+),', text)\n",
        "feature_names = re.findall(r'Added feature: (\\w+_\\w+),.*?Updated Performance: (\\d+\\.\\d+)', text)\n",
        "\n",
        "# Extracting accuracy values\n",
        "accuracy_values = [float(value) for _, value in feature_names]\n",
        "\n",
        "# Plotting the performance values\n",
        "plt.plot(accuracy_values, marker='o')\n",
        "plt.xlabel('Added Features')\n",
        "plt.ylabel('Mean Accuracy')\n",
        "plt.title('Performance Over Features Added')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "xT4FJchiZ8us",
        "outputId": "3f6506e8-9e3e-4273-9f96-312740a1b57f"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [
        "Mz-mppveFfZH",
        "SWasGtbjw8o2",
        "fQ8xz8KCu4Lv",
        "DN4rixIgs24n",
        "ILkI6mdu06FV",
        "cUd6llzT0_lw",
        "7h5nsgLc5tWM"
      ],
      "provenance": [],
      "authorship_tag": "ABX9TyN4f+N/86GA/q0yS0tbl4wV",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}