notebooks/32_multiclass_auc.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[![Open in Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/justmarkham/scikit-learn-tips/master?filepath=notebooks%2F32_multiclass_auc.ipynb)\n",
    "\n",
    "[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/justmarkham/scikit-learn-tips/blob/master/notebooks/32_multiclass_auc.ipynb)\n",
    "\n",
    "# 🤖⚡ scikit-learn tip #32 ([video](https://www.youtube.com/watch?v=-s-KdkYmCaA&list=PL5-da3qGB5ID7YYAqireYEew2mWVvgmj6&index=32))\n",
    "\n",
    "AUC is an excellent evaluation metric for binary classification, especially if you have class imbalance.\n",
    "\n",
    "New in scikit-learn 0.22: AUC can be used with multiclass problems! Supports \"one-vs-one\" and \"one-vs-rest\" strategies.\n",
    "\n",
    "See example 👇"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_wine\n",
    "X, y = load_wine(return_X_y=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# only keep two features in order to make this problem harder\n",
    "X = X[:, 0:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "clf = LogisticRegression()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Multiclass AUC with train/test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import roc_auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "clf.fit(X_train, y_train)\n",
    "y_score = clf.predict_proba(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9399801587301587"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use 'ovo' (One-vs-One) or 'ovr' (One-vs-Rest)\n",
    "roc_auc_score(y_test, y_score, multi_class='ovo')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Multiclass AUC with cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9086960878627546"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use 'roc_auc_ovo' (One-vs-One) or 'roc_auc_ovr' (One-vs-Rest)\n",
    "cross_val_score(clf, X, y, cv=5, scoring='roc_auc_ovo').mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Want more tips? [View all tips on GitHub](https://github.com/justmarkham/scikit-learn-tips) or [Sign up to receive 2 tips by email every week](https://scikit-learn.tips) 💌\n",
    "\n",
    "© 2020 [Data School](https://www.dataschool.io). All rights reserved."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}