Skip to content

Instantly share code, notes, and snippets.

@ksetdekov
Created June 1, 2021 20:28
Show Gist options
  • Save ksetdekov/f4d8856bb2df9cbe3df6de015d707251 to your computer and use it in GitHub Desktop.
Save ksetdekov/f4d8856bb2df9cbe3df6de015d707251 to your computer and use it in GitHub Desktop.
plot_decision_regions example
Display the source blob
Display the rendered blob
Raw
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "plot_decision_regions_ex.ipynb",
"provenance": []
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
}
},
"cells": [
{
"cell_type": "code",
"metadata": {
"id": "JEVt9p5VbsgR"
},
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import multiprocessing\n",
"\n",
"from sklearn.model_selection import train_test_split, GridSearchCV\n",
"from sklearn.linear_model import LogisticRegression\n",
"from sklearn.metrics import roc_auc_score\n",
"from sklearn import svm\n",
"from sklearn.ensemble import RandomForestClassifier\n",
"\n",
"import matplotlib.pyplot as plt\n",
"from mlxtend.plotting import plot_decision_regions\n",
"import matplotlib.gridspec as gridspec\n",
"import itertools"
],
"execution_count": 1,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "XQtwsatDbhvt"
},
"source": [
"data = pd.read_csv('https://raw.githubusercontent.com/ksetdekov/ml_dpo_2021/master/hw/hw3/heart.csv')\n",
"X_train, X_test, y_train, y_test = train_test_split(data.drop('target', axis=1), data['target'], test_size=0.25, random_state=13)\n",
"\n",
"\n",
"logreg = LogisticRegression(penalty='none', random_state=13, max_iter=1000)\n",
"\n",
"\n",
"svm_base = svm.SVC(C=1, kernel='linear', random_state=13)\n",
"\n",
"\n",
"rfset = RandomForestClassifier(n_estimators=200)"
],
"execution_count": 2,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "O_isqc6XdQf8",
"outputId": "8125acb1-93fb-4204-cc5a-2a7edd9a79f5"
},
"source": [
"X_train.mean()"
],
"execution_count": 3,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"age 54.656388\n",
"sex 0.678414\n",
"cp 0.929515\n",
"trestbps 131.436123\n",
"chol 246.691630\n",
"fbs 0.149780\n",
"restecg 0.533040\n",
"thalach 148.660793\n",
"exang 0.321586\n",
"oldpeak 1.041410\n",
"slope 1.383260\n",
"ca 0.722467\n",
"thal 2.312775\n",
"dtype: float64"
]
},
"metadata": {
"tags": []
},
"execution_count": 3
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 519
},
"id": "AbzqENUFb-iC",
"outputId": "b0d54016-eb93-4f0b-d0f8-0639f0aeb341"
},
"source": [
"width=1000\n",
"\n",
"filler_feature_values_0={0:54.656388, 1: 0.5, 2: 0.929515, 3:131.436123, 4:246.691630, 5:0.149780, 6:0.533040, 7:148.660793, 8:0.321586, 9:1.041410, 10:1.383260, 11:0.722467, 12:2.312775}\n",
"filler_feature_ranges_0={0: width, 1: width, 2: width, 3: width,4: width, 5:width, 6:width, 7:width, 8:width, 9:width, 10:width, 11:width, 12:width}\n",
"\n",
"feature_index=[9,0]\n",
"filler_feature_values_0.pop(feature_index[0])\n",
"filler_feature_values_0.pop(feature_index[1])\n",
"\n",
"filler_feature_ranges_0.pop(feature_index[0])\n",
"filler_feature_ranges_0.pop(feature_index[1])\n",
"\n",
"\n",
"gs = gridspec.GridSpec(1, 3)\n",
"\n",
"fig = plt.figure(figsize=(15,4))\n",
"labels = ['Logistic Regression', 'SVM', 'random forest']\n",
"for clf, lab, grd in zip([logreg, svm_base, rfset],\n",
" labels,\n",
" itertools.product([0, 1, 2], repeat=1)):\n",
"\n",
" clf.fit(X_train, y_train)\n",
" ax = plt.subplot(gs[grd[0]])\n",
" fig = plot_decision_regions(X= X_train.to_numpy(), y=y_train.to_numpy(), clf=clf,\n",
" feature_index=feature_index, \n",
" filler_feature_values=filler_feature_values_0, \n",
" filler_feature_ranges=filler_feature_ranges_0, legend=2)\n",
" plt.title(lab)\n",
"\n",
"plt.show()"
],
"execution_count": 4,
"outputs": [
{
"output_type": "stream",
"text": [
"/usr/local/lib/python3.7/dist-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
"STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
"\n",
"Increase the number of iterations (max_iter) or scale the data as shown in:\n",
" https://scikit-learn.org/stable/modules/preprocessing.html\n",
"Please also refer to the documentation for alternative solver options:\n",
" https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
" extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
"/usr/local/lib/python3.7/dist-packages/mlxtend/plotting/decision_regions.py:244: MatplotlibDeprecationWarning: Passing unsupported keyword arguments to axis() will raise a TypeError in 3.3.\n",
" ax.axis(xmin=xx.min(), xmax=xx.max(), y_min=yy.min(), y_max=yy.max())\n",
"/usr/local/lib/python3.7/dist-packages/mlxtend/plotting/decision_regions.py:244: MatplotlibDeprecationWarning: Passing unsupported keyword arguments to axis() will raise a TypeError in 3.3.\n",
" ax.axis(xmin=xx.min(), xmax=xx.max(), y_min=yy.min(), y_max=yy.max())\n",
"/usr/local/lib/python3.7/dist-packages/mlxtend/plotting/decision_regions.py:244: MatplotlibDeprecationWarning: Passing unsupported keyword arguments to axis() will raise a TypeError in 3.3.\n",
" ax.axis(xmin=xx.min(), xmax=xx.max(), y_min=yy.min(), y_max=yy.max())\n"
],
"name": "stderr"
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x288 with 3 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "zMkeRr5zkLOn"
},
"source": [
""
],
"execution_count": 4,
"outputs": []
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment