05-model-evaluation-and-improvement.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [],
   "source": [
    "from preamble import *\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model evaluation and improvement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set score: 0.88\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import make_blobs\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# create a synthetic dataset\n",
    "X, y = make_blobs(random_state=0)\n",
    "# split data and labels into a training and a test set\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "# Instantiate a model and fit it to the training set\n",
    "logreg = LogisticRegression().fit(X_train, y_train)\n",
    "# evaluate the model on the test set\n",
    "print(\"Test set score: {:.2f}\".format(logreg.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3wAAACqCAYAAAD2m02NAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e5daddd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_cross_validation()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Cross-validation in scikit-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validation scores: [ 0.961  0.922  0.958]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "iris = load_iris()\n",
    "logreg = LogisticRegression()\n",
    "\n",
    "scores = cross_val_score(logreg, iris.data, iris.target)\n",
    "print(\"Cross-validation scores: {}\".format(scores))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross validation scores: [ 1.     0.967  0.933  0.9    1.   ]\n"
     ]
    }
   ],
   "source": [
    "scores = cross_val_score(logreg, iris.data, iris.target, cv=5)\n",
    "print(\"Cross validation scores: {}\".format(scores))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score: 0.96\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score: {:.2f}\".format(scores.mean()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Benefits of cross-validation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Stratified K-Fold cross-validation and other strategies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iris labels:\n",
      "[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
      " 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1\n",
      " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2\n",
      " 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2\n",
      " 2 2]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "iris = load_iris()\n",
    "print(\"Iris labels:\\n{}\".format(iris.target))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA/wAAAFjCAYAAACADgBEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ndVs6AEsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3620247eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_stratified_cross_validation()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### More control over cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "kfold = KFold(n_splits=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validation scores:\n",
      "[ 1.     0.933  0.433  0.967  0.433]\n"
     ]
    }
   ],
   "source": [
    "print(\"Cross-validation scores:\\n{}\".format(\n",
    "      cross_val_score(logreg, iris.data, iris.target, cv=kfold)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validation scores:\n",
      "[ 0.  0.  0.]\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=3)\n",
    "print(\"Cross-validation scores:\\n{}\".format(\n",
    "      cross_val_score(logreg, iris.data, iris.target, cv=kfold)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validation scores:\n",
      "[ 0.9   0.96  0.96]\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=3, shuffle=True, random_state=0)\n",
    "print(\"Cross-validation scores:\\n{}\".format(\n",
    "    cross_val_score(logreg, iris.data, iris.target, cv=kfold)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Leave-One-Out cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of cv iterations:  150\n",
      "Mean accuracy: 0.95\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import LeaveOneOut\n",
    "loo = LeaveOneOut()\n",
    "scores = cross_val_score(logreg, iris.data, iris.target, cv=loo)\n",
    "print(\"Number of cv iterations: \", len(scores))\n",
    "print(\"Mean accuracy: {:.2f}\".format(scores.mean()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Shuffle-Split cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [],
   "source": [
    "mglearn.plots.plot_shuffle_split()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![shuffle_split](images/06_shuffle_split.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validation scores:\n",
      "[ 0.96   0.947  0.933  0.92   0.92   0.88   0.84   0.933  0.907  0.84 ]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import ShuffleSplit\n",
    "shuffle_split = ShuffleSplit(test_size=.5, train_size=.5, n_splits=10)\n",
    "scores = cross_val_score(logreg, iris.data, iris.target, cv=shuffle_split)\n",
    "print(\"Cross-validation scores:\\n{}\".format(scores))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Cross-validation with groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzwAAACLCAYAAACk0Zy+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e30c6da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_group_kfold()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validation scores:\n",
      "[ 0.75   0.8    0.667]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GroupKFold\n",
    "# create synthetic dataset\n",
    "X, y = make_blobs(n_samples=12, random_state=0)\n",
    "# assume the first three samples belong to the same group,\n",
    "# then the next four etc.\n",
    "groups = [0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 3, 3]\n",
    "scores = cross_val_score(logreg, X, y, groups, cv=GroupKFold(n_splits=3))\n",
    "print(\"Cross-validation scores:\\n{}\".format(scores))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Grid Search"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Simple Grid-Search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Size of training set: 112   size of test set: 38\n",
      "Best score: 0.97\n",
      "Best parameters: {'C': 100, 'gamma': 0.001}\n"
     ]
    }
   ],
   "source": [
    "# naive grid search implementation\n",
    "from sklearn.svm import SVC\n",
    "X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target,\n",
    "                                                    random_state=0)\n",
    "print(\"Size of training set: {}   size of test set: {}\".format(\n",
    "      X_train.shape[0], X_test.shape[0]))\n",
    "\n",
    "best_score = 0\n",
    "\n",
    "for gamma in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "    for C in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "        # for each combination of parameters\n",
    "        # train an SVC\n",
    "        svm = SVC(gamma=gamma, C=C)\n",
    "        svm.fit(X_train, y_train)\n",
    "        # evaluate the SVC on the test set \n",
    "        score = svm.score(X_test, y_test)\n",
    "        # if we got a better score, store the score and parameters\n",
    "        if score > best_score:\n",
    "            best_score = score\n",
    "            best_parameters = {'C': C, 'gamma': gamma}\n",
    "            \n",
    "print(\"Best score: {:.2f}\".format(best_score))\n",
    "print(\"Best parameters: {}\".format(best_parameters))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The danger of overfitting the parameters and the validation set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1MAAABsCAYAAACPfOh5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJztnXmcHVWZv58vSYAoO8gqm7ihgqMIooIgoijK4gIiKMEN\nf6O4/NRxQRiCG47KgMs448YigiwqsikKIsOmiCCyyE5YRBIgQEBIIAnv/PGem1Sqq7vv7b7dXYHv\n8/ncTydVT5177r1V55z3LFWKCIwxxhhjjDHG9MYyE50BY4wxxhhjjFkacTBljDHGGGOMMSPAwZQx\nxhhjjDHGjAAHU8YYY4wxxhgzAhxMGWOMMcYYY8wIcDBljDHGGGOMMSPAwZQxLUfSdElPSHr1KNPZ\nrqTz7/3KmzHmyY2k2yTdWtu2XylL9u0hnWPKMRv0P5dLvM+A/BpjzFjiYMqYHpC0YWkQHDWObxvl\n1ba0Ws0E/VbGPNloKi9GUo70peyRdL6kJ8b6fZYG3EHWGxNZJ/SrU7SftDFPSyuTJzoDxphh+Tbw\nU+COUaZzKbApcN+oc2SMeSrzC+APwN0T8N7DBUs7jFdGjOmBNgb5bczTUomDKWN6Q+P9hhFxP3B/\nH9KZB9w4+hwtNYz7b2XMU4GIeBh4eKLz0UREzJjoPIwjLuN6YyK/rzb+Vm3M01KJp/kZ0yWSDgFu\nJXtyOmsGnqiuHahOu5C0paSzJM2WtLCzVkDS9pK+L+laSXMkPSrp6nLMcg3v2zgUX7adJ2n1kt4/\nJM2TdI2k/RrSaZwSUqbNLJS0jKQDJd1Y0rlD0lclTRnk+9hH0hUl/7Mk/VjSOl1Mw6mns5mkn0qa\nUd73HkmXSzpC0qSaO0nShyT9oXx3j5Q8fFiSKt6wv5UxSzOSXl7O558P4Vwnaa6kVcr/p0g6oJRL\nt5XrbbakcyS9oYf3njbYtSRpR0kXSvpnSftUSc8bIq39JP1M0i2lLJkj6SJJ+9S8DUu58ur87xLX\n9HkVr3HNlKRlJX1W0lWl3Jgj6QJJezS4i6aDlX+fKOne8l1eJulN3X5XJb1tJZ0h6c7ynd9dyrAB\n0/MkTZX0OUl/Kd/hw5IukbRXzTsaOI8s4zp1xBOlLPe0rRq91AmSdpL0q/Kbz5N0s6SvSVq5Id1h\n6y9JM4DOb31+9bfqMu/TJF1c0p6rrJvPlrRng7uepO+U62mepPsknSbpZTVvVHkyS+KRKWO65/fA\nysDHgSuBX1b2XVlzXwkcCFwI/AhYA3i87PsM8DzgEuBMYHngVcB0YDtJO0ZEdeh9qKH4VYCLgceA\nU4DlgD2AoyQtjIjjuvhcnbR/CmwD/Bp4CNgZ+DTwDOB91QMkfRr4KjlidnTxX1fyMmeI/C6BpM3I\n6YdPAKcDM4CVgGcD/wp8Hni0uJPJ7+v1wPXA8cA84DXkVMitgGkl6V5+K2OWOiLiUkk3ADtLWjUi\nHqjul7QlWc6cEhEPls2rAUeS1+lvgXuBdYBdgF9Jen9EdLueZMA1LuntwIlkeXQiMJMsU/4AXDVI\nOt8FrgH+l5w2uDpZ9hwn6bkRcUjxHiTLyPcAG5R/dzpQbhsmX1PIz/tq4DrgO8DTgLcDJ0l6cUQc\n1JC3jYA/AbcAPya/v3cAvyzl9P8O8pmq7/0GstyaQ5Zxd5V0NiXLuC9U3JXJsuvFwBVk3bEMsBNw\ngqQXRESnAXxq+az7AeeXV4fq92GSruqEEnQdAswmf7d7gM2BTwFvlPSKiPhncbutv44Adge2A45h\n8e8zbD0p6SvAZ8lA8CTyPFoH2JI8f0+uuC8lz/NVgN8APyfbHrsDF0naPSLOLvqI82QaiAi//PKr\nyxewIVlwHjXI/u3K/oXA+wdxNhpk+6HluD1q2w8p219d2955n+8BqmzfFJgPXDNI3v69tv33Zftl\nwMqV7VOBm0paa1a2b0wGhjOBdWtpndDJV5ff5zfKZ3hzw76Va/+fXtI+svZ5BfywpLNLt7+VX34t\n7S+ykbUQ+FDDvv8q+3aubFu2fs2W7SsCV5PrKZer7ZsB3FrbNq2kvW9l29PJBuhjwEtq/uGV8mqD\n2r6NG/IzGTi3pLVObd/vhypfBsnv58r7nwEsU9m+RvEXAltXtm9Yye9BtbReX/ad2eVv9POSzosa\n9q1W+/8xxf1kbfuyZCfXAmDzyvbGMt2vQX+L4erv15T9FwIr1vbtW/YdXtnWS/3VWI93kef7yPXS\nyzXsW63y70nAzWTwtk3NWxv4OxnITxltnvwa+PI0P2PGhisj4odNOyLitkGO+SYZGOzUw/s8Sla8\ni3qTIuI6sud5U0lP6zKdAD4dEXMq6cwlR3+WAapTBPYhC+5vR8Q/aul0Gne9Mm9Ahip5kSTgALLn\n+hO1zxvAJyt5M+apwnHktTuturGMxLyD7FXv9EQTEY83XLNEroE6CliV7PEeCbuV44+PiL/U9h1K\n9qgPIBrWOEXEAjIYnAy8doT5qfJesiH8iYhYNAU5Iu4DvkiWu+9vOO524Mu1vP2WbNxu1eV7d8qq\npjJu0VpYSauR5defI+Lwmvc4OaNhGWDvLt/X9M5Hyd9r/3JNLCIifkyOYDXVMUPWX31gPg0jRtXz\nB3gT8CyyXr6o5s0EvkYGVf24nkwNT/MzZmz402A7SoDzcXKI/blkr3BnukoA6/XwPjdFmXJQ487y\nd1XKNLkuuHyYdDr8S/l7cV2OiDsk3Un2AHbDScDHgNMk/Yzsjb44IuprHp5LTo25EThYGrBuVsBc\nclTOmKcEEXGXpN8BO0p6fkRcX3btSl4vh1eDBwBJLyCn725LThdavpokvZU/VV5ajr+gIZ8PSbqS\nnGa3BJLWJzthdiCn703tU3466a8AbAL8PSJualA6661e0rDvymrHTYU7ga27zMLxwFuAP0k6iRxZ\nuzgi7qp5W5KdVFGmmtVZtvx1GTd2bE0GLns21DGQv8EzKtNqu62/RsPxZEfi3ySdTE6H/UNEPFTz\nXlH+bjTI+fMcsp7clEoHi+kPDqaMGRtmNm0s635+T1acV5PrCu4lC3DIqWwDbkIxBA8Osn1B+Ttp\nkP0DaCicB0unswh31iBJzaLLYCoiLpO0DTm3/G3Au8iBqBuAQyPixKKuXv4+h8WLZpt4ejfva8yT\niGPI9YrTyOlslH8Huc5nEZK2Bn5HXs+/A04j1zs+QXaS7EZv5U+V4cqFAWWipI0p04vJqVW/IUew\nFpLrlaaNIj/1fA12G/fO9lUa9g1VvnY1syciTpX0ZnL0/D3A/mQZdznwuYg4t6idMm5LBh8dDFzG\njSWrk9fGUHVMACsAD/RQf42Gj5Nr9t5Djk5+Flgg6VfkrJRbKnmHXEc1XN5Nn3EwZczYMNgizt3I\nivKoiFhiWomktclgqu10gq61yMXcddbqJbGIuBTYtUxN2gJ4A/AR4HhJ90TEeSyeInRqRAxVWRjz\nVONU8pp8l6QDyXVAbyBHVa6uuQeRI1HbR8SF1R2SPkuWTyOlc40Odv2v3bDtk+So935Ru1lOuXvd\nfqPITz1fTe8POTpX9fpORPwa+LWkqcDLgTcDHwLOkPSSMqLYef8jIuJTY5UXMyRzyPW4a3R7QJf1\n14gpI6PfAr4laQ3yhi57AXsCL5D0woiYz+IbP+0aEWeN5j1N73jNlDG90VkP1PWIT41nkwXeqQ37\nth9hmuPNX8jpAtvUdyhv/77+SBKNiPkR8ceImE5OnRCLG3fXk73EW6t2u/QhGO1vZUzriXx+3MnA\nusCO5JqOycCxDfomwP31QKqw/SizcgV5zW5X3yFpJRZPD67nB/IhwE35aeqUWljS7OoZOWUa9C3A\nepI2aVA6D/ltmubcVyJibkScX4Klr5DTxt5Ydv+JHCHctockXcb1xnDf1x+BVSX1PJVymPqrm/fu\n5j3ui4hfRsRe5PTUTYAXVfIuGqbSDoHPnz7hYMqY3niArOA3GOHxt5EF3vbVjZKeRd5qfGm4LekJ\n5DSXj0h6Zm3fV+mhYJb0CknLN+zq9CI/ChARC8nbn68LfLvpGElr1yrB0f5WxiwtHEOWK9OAd5PT\nhk9o8G4DVpP0oupGSe8j71I3Gk4jr7m9JW1R23coi6fb1fMDA8vDnag9jqHC7PK3l+v6KLK983VJ\ni9o9paf/YLKcOLqH9LpG+YyppjKxXsbdS66PeZmkg6r5rKT1LEkbVTaN5Lt4KjNcnXAEeR39QNI6\n9Z2Snibp5ZX/d1V/FWaXtLv+rZTPRntlw/YpLJ7W13mP08hOgw9LemP9mHLc1rX89pwn04yn+RnT\nAxHxiKRLgW0l/YS8IcJC4LSIuKaLJM4gb1/6CUmbk6M8G5J34jmTHL5vG0v0AEfErcqHTX4Z+GtZ\nVD2HXLexKvBXYLMu0/40sIOkC8lbFP8TeCHZWzsb+H7F/SL5vI8PArsoH9R5F7AmuZbqVeSzva4r\n+Rztb2XMUkFEXCLpZvIZc1OA08ud6uocSd4t9OKymH0OeafOV5HPqRvwANshqJcLj0jan1wHemEp\nF+4mR7BfSN6Yoj7q8l1yLcjPygL+f5A97TuRo21N5eHvSj5PLetG5gK3R8RPhsjrN8gyZTeyzPoV\n+ZypPcjn6P1HRFzS7QfvkW+Ro2IXk8Hj4+R0sB3IMq+6ruYAcvbCocC7JV1ErkFbl7xxwMuAd7I4\nCL2BLAP3krSAvPtgAD+OiDsxSzBEnXB6RFwdEedJ+gxwGHBTOU9mkOuMNiRHXS8kn4MGvdVfnUeQ\nfLU8n+qBkqcl7hZZYyr5fKibyZHT28lpuq8Dnk/WZTeUdBZIeit5c4mzJF1C3n3wUXK2yJbkY03W\nYfHdB0eSJ9PEWN533S+/nowv8vajp5E3jlhA5XkrZGG7EDh4iOPXI29pfCfwCHkjik+SIzoLgd/V\n/MGeMzXArew7uuRtg8q2xryRBeqCQdIZ8DyZyr59yAL+UbLCP5bskbuanErUzXe5I/lgymvIgvxh\nMhg6Alh/kGP2Ac4hn78xr3yPF5CLc9fr9rfyy68n04tcBL+wnOe7D+HtTD4wfA750O1fkwFP47VO\nNhJvqW0bqlx4bbke/0k2KH9B3o1zQJlU/K3Ju6DNLnm6gHyI8GDl1TLAl8hOqceKc95Q+S3blyUX\n719Vyt055J3R9mxwNyzp/miQ73DQMrPBfTs54nQDubbtwZKHLwCrN/iTyfVUF5UycS4ZPJ1DrsVZ\nteZvUfY9UCnj/NygwX+PYesE4JVkkPv3UsfMIqexfh14acXrqf4ib2t/RTn/Fg53DpVz4VPAWeUc\n6NS1lwAfACY3HLMGOYX0qnINPlTOvZPJQHyZ0eTJr+aXypdpjDGjRtKKZGH/l4h41UTnxxhjjDFm\nLPGaKWNMz0hao9zmvbptEvCf5K2MmxaUG2OMMcY8qfDIlDGmZyR9kJymci45zW418i5CzyWnDLwq\nIh6buBwaY4wxxow9vgGFMWYkXEouxN2WxXcVmkHeJOJrDqSMMcYY81TAI1PGGGOMMcYYMwK8ZsoY\nY4wxxhhjRoCDKWOMMcYYY4wZAQ6mjDHGGGOMMWYE9HQDiqlTp86cN2/eWmOVGWOMMcb0h+WXX37W\n3Llz165umzx58uyFCxeuNlF5MmPLpEmT7l+wYEHnpkBMmTJl5oIFC9xuM2YYJk+ePGv+/PlrD28O\npKcbUEiKqn/++eezxx57cMopp7D99tsPe7x9+/bt27dvf3x8SUSEatti+vTpw6Y/Y8YMTjnlFPbY\nYw823nhj+0uJ/+ijjy7xm3f7exvzVGf69OkDystuGfE0vzZUFPbt27dv3779/tLWQMH+8L4xZvwZ\nUTDVtorCvn379u3btz962hwo2B+9b4zpPz0HU22rKOzbt2/fvn37o6dtDX/7/fWNMWNDz8FUmyoK\n+/bt27dv336z3wtta/jb769vjBk7eg6m2lRR2Ldv3759+/ab/W5pW8Pffn99Y8zY0nMw1aaKwr59\n+/bt27ff7HdD2xr+9vvrG2PGnp6DqeFoc8Vi3759+/bt20/a1vC331/fGDM+9DWYaltFYd++ffv2\n7dsfSNsa/vb76xtjxo++BVNtqyjs27dv3759+wNpW8Pffn99Y8z40pdgqm0VhX379u3bt2+/mTY1\n/O331zfGjD+jDqbaVlHYt2/fvn379genLQ1/+/33jTHjz6iCqbZVFPbt27dv3779oWlLw99+/31j\nzPgz4mCqbRWFffv27du3b3/0tDVQsD+8b4wZf0YUTLWtorBv3759+/btj542Bwr2R+8bY/pPz8FU\n2yoK+/bt27dv3/7oaVvD335/fWPM2NBzMNWmisK+ffv27du33+z3Qtsa/vb76xtjxo6eg6k2VRT2\n7du3b9++/Wa/W9rW8LffX98YM7b0HEy1qaKwb9++ffv27Tf73dC2hr/9/vrGmLGn52BqONpcsdi3\nb9++ffv2k7Y1/O331zfGjA99DabaVlHYt2/fvn379gfStoa//f76xpjxo2/BVNsqCvv27du3b9/+\nQNrW8LffX98YM770JZhqW0Vh3759+/bt22+mTQ1/+/31jTHjz6iDqbZVFPbt27dv3779wWlLw99+\n/31jzPgzqmCqbRWFffv27du3b39o2tLwt99/3xgz/ow4mGpbRWHfvn379u3bHz1tDRTsD+8bY8af\nEQVTbaso7Nu3b9++ffujp82Bgv3R+8aY/tNzMNW2isK+ffv27du3P3ra1vC331/fGDM29BxMtami\nsG/fvn379u03+73Qtoa//f76xpixo+dgqk0VhX379u3bt2+/2e+WtjX87ffXN8aMLT0HU22qKOzb\nt2/fvn37zX43tK3hb7+/vjFm7Ok5mBqONlcs9u3bt2/fvv2kbQ1/+/31jTHjQ1+DqbZVFPbt27dv\n3779gbSt4W+/v74xZvzoWzDVtorCvn379u3btz+QtjX87ffXN8aML30JptpWUdi3b9++ffv2m2lT\nw99+f31jzPgz6mCqbRWFffv27du3b39w2tLwt99/3xgz/owqmGpbRWHfvn379u3bH5q2NPzt9983\nxow/Iw6m2lZR2Ldv3759+/ZHT1sDBfvD+8aY8WdEwVTbKgr79u3bt2/f/uhpc6Bgf/S+Mab/9BxM\nta2isG/fvn379u2PnrY1/O331zfGjA09B1Ntqijs27dv3759+81+L7St4W+/v74xZuzoOZhqU0Vh\n3759+/bt22/2u6VtDX/7/fWNMWNLz8FUmyoK+/bt27dv336z3w1ta/jb769vjBl7eg6mhqPNFYt9\n+/bt27dvP2lbw99+f31jzPjQ12CqbRWFffv27du3b38gbWv42++vb4wZP/oWTLWtorBv3759+/bt\nD6RtDX/7/fWNMeNLX4KptlUU9u3bt2/fvv1m2tTwt99f3xgz/ow6mGpbRWHfvn379u3bH5y2NPzt\n9983xow/ioiu5alTp86cN2/eWmOYH2OMMcb0geWXX37W3Llz165umzJlyqwFCxasOVF5MmPL5MmT\n75k/f/6idtqUKVNmLliwwO02Y4Zh8uTJs+bPn7/28OZAegqmjDHGGGOMMcYkfbsBhTHGGGOMMcY8\nlXAwZYwxxhhjjDEjwMGUMcYYY4wxxowAB1PGGGOMMcYYMwIcTBljjDHGGGPMCHAwZYwxxhhjjDEj\nwMGUMcYYY4wxxowAB1PGPImQNF/Sq7t0t5M0fxjns5JmSnpI0haSfiXpUyPI1+ckndbrccYYM5FI\n2lDSE5LWHcP3uEbSHmOVvhlfyvnyyjF+D58zLcLBlDHjiKTzS0H79tr2rcr2W8c5S4M+tVvSesCX\nge0iYqWIuDwido6Ib1ScAZWGpN9LOnCJN4k4LCJ263fmjTGLypV5pdPjAUlXSHrrROerVyTNkLT3\nROejgUHLyV4YLDCLiBdFxCn9eA/THbVr5iFJD5e/L5zovFXxObN04GDKmPElgL8BH6ht/0DZ3iY2\nBhZGxA0TnRFjzJAE8IWIWAlYHfgpcJKkZ/eakKRJktTvDI4XkqZMdB6GQPQpMDOjZtE1U14rlr/X\nTnTGavicWQpwMGXM+PML4CWSNgKQtALwNuDoqiRpqqRvSrpD0j2SfiFp/cr+FSQdK2l26dHdt/5G\nknaX9OfSW31tt72+kvYEfgtMKj12N5Xti0adJF1JFvK/LT1635f0bWBb4OBy3HXFPUTSOZX0Z5Sp\nf+cW7ypJr6jsnyzpCEmzJP1D0r9JuqnpMxpjFhMRTwDfBSYBmwFI+rKkWzrXsqSPdfxKz/d7JV0L\n/BN4hqR3SLpS0hxJd0n6H0lTK8fNkPR5SeeVdP8qaTNJe5X3eEDSDyQtUzlmfUmnSLq7pPk9SU8v\n+04HNgB+WMqTs8v2SZIOlHSDpPslXShpi0qaR0v6Sfk7Gziy/p1IWraUT7MkPVjSeltl/7Yl3dkl\n758Y6jserlxVTqG+oKR3j6Sjyq4ry98by2f8fOW73Lt2/B9LXv8maf/avvmS9pR0c8nDSZ3v0YwO\nSTuX82RSZdvTyzm+bfn/oNdTQ3rTOvVnZdvRkr5f+f9Rynr+IeX0vXdWdJ8zSwEOpowZf+YBxwPv\nL/9/J3A+MLPmHQlsVV4bArOBM6RFvcbfBDYBng9sDuxG5ZqW9DrgB8BHI2JVYBrwHUnbDJfBiDgZ\neCM5MrViRDynwfkXstfsdaVHb/+I+AhwIfDFctym1UNqSbwHOABYCTgXOLay70Bgp/LZNwaeSTa0\njDFDoByZOQB4HPhr2Xwt8MqIWJEcBT+slA9V3glsD6wI3Ac8CLwzIlYmO0i2AQ6qHbMv8P+AVYCr\ngFNLGpuRZdKuwDtKvpYDzgOuIcuzFwDrAd8CiIhdgTuA95Xy5A3lPb4A7AK8nhx1Owo4W9LKlXy8\nHTgLWAP4ZMPXMg3YAnheRKwC7FC+EyS9oBz7HxGxOvAm4MOS3tWQzrDlqqTNgbOLszawPnBMOfzF\n5e9zymf8ckP6GwO/Bv4LWI0sJw+rBn9koPw68nt+LvAS4KNN+TU9czYwnzwPOuwJ3B0RF5b/d3M9\nVRluZOlC8npZmTzfj5H0/LLP58xSgIMpYyaGHwL7ld6v/cmKdxElYNoX+HxEzIyIucDHgU2Brcr+\nvYGDIuLeiHgY+AwZ3HT4KPDNiLgEICL+DPykpNtPRjol6H8i4vqICPL72ETSimXfu8nGze0R8Rj5\n2TzVwZjBOUjS/cCdZPDx1oi4FSAiToiIWeXf55PBw2trx08vZcmCiHgiIn4TEdeVY24F/rvhmO9H\nxI0RsRA4gez4ODAi5kXEnWQn0cuKu0tJ69CIeDwi5gCHAPtUOohgYHnyEeDfSlkQEXE0cDdLNnYv\nioiflf3zGr6bx4EVgBdJmhQRd0XE9WXfvwInR8SZJX83ko3SaQ3pwPDl6geB0yPiuIiYHxGPRcQF\ntTSGKjP3Ai4vxz8REZcC32Nx5xtkWfiZiJgbEfcCv2Tx92y656Ay2nl/Ga25v4zs/gR4b8Xbjwzi\nga6vp66JiKMj4sFy/p5MdkxsX9N8zrSYyROdAWOeikTEtZJuBw4GnkH2hlWnijwDWA64rXLMI5Lu\nIXs6Z5T9t1eOmVF7m42B7StTVkR2oNQr9omiOhL3SPm7IvAw2WN9R2dnRMyTdO845s2YpY0vRcRX\nmnZI+ijZsHomWQ4sTwY/HYIly5LOCMzB5Mj3smR7YVYt6bsr/36UHMm+v7at00GyEbBhCfgWvQ2w\nkBzBqabVycMaZBB0hqSoHDO5fJYOt9WPrXEcsCZwBPAcSecCny5B4sbAa7T4hh0qrzsaUxq+XN0I\nuGKY/AxFp3yvcgs5yteh/j0/wuLv2XTPYNfM0cBfy/m3MvAKcuQW6Op66prSkXAoOfq1Vtn8NLIN\n0C0+ZyYYj0wZM3H8gJw286MyOlPlXuAxsmIGFq2tWpOs5O8je1s3qhyzcS2N28ne5tXKa9WIWDki\ndunjZ2gaLXqiD+neRU4FAnL9GL1VLsYYQHm3za8CH+iUA8CZDOzpfqJyzBRyyt4JwDPL1Lj6yHev\n3A7cUCmPOmXS0yOiE0gtUXZExH3kGq4da8esGBFfa8p7E6W3/usRsSU5XXgui0cabgeOqqW/SkRs\nPsTnGKpcvQ0YMC26ks/hvsM7WbJch5zOfecwx5k+EXnTpcvJGRLTgHMj4h/Q0/XU4WGgvjapeme+\nvYH3AW8p59Kq5MhUJz2fM0sBDqaMmTh+Sq4D+FZ9Rwmufgx8UdI6kp4GHA5cB1xWpiKcABwqaU1J\nKwGHsWRwcyTw/yVtI2kZ5SLsl6qyeLsP3M3AhsNMoOe7iLFkhXEc8G+SNpK0PPnZlto7jBkzgawE\nLADuU/Imcj1klfq1tWx5PRgRj5d1RQeMMh9nAssqbzyzAuTjFyTtXnFmMrA8+SZwuMqdCZU33nm9\npLW7fWNJryll32Syk+oRckQM8mYde0l6s/LGN5Mkbaoln9dX/X6GK1e/B+wqaZ+yb3lJ25V995b3\nHSzYgqwXtpD0rpKXrcip4D/s9vOavnAMOdVvXypT/OjueqpyJbCm8sYWkvQWoHpurUiu0Zpdzr/3\nsnidFPicWSpwMGXM+LIo2Clz6c8rawea+DjwZ+AysrdzLWDXyijWx8ih/evJheans7iBQEScQy6O\n/To5knUX8J8M7CUbUf4LnycDvtmS/rtsOwJ4WZmDfnWX6dS3HQacA/wJuJXM+91kQ8gYsyRDrSf8\nDdkxcxnZMHsreUfRQY+PiEfItURfl/QQ8G3ypjndvufADOa6zx3IG09cL+lB8hqvNhy/BLy7lCdn\nlW3TyfUdp5VjbiDXJfXSflmL7KC5nyxLNiAbm0TeCvvNZHl7NzmV8WjyZhaLsl/5HEOWqxFxFbAz\n8KGS1u3Au8q+eeTUyRPLOp3PNaR/Wzn+IyX9Y8m1sz/v4fOa7jhYA58ztXPZdyLwLHLKXfWB8z1d\nT2Uq6cfImSizyQ7Un1XcY4FLgZvJkaTnU5mK73Nm6UADZxcZY0y7UN7C9QHg1RHxx4nOjzHGGGMM\neGTKGNNCJK0qaacy7WFlsmd8BtkbaIwxxhjTChxMGWPayDLklJ/Z5F2J1iWnOC4c8ihjjDHGmHHE\n0/yMMcba+R6qAAAAUElEQVQYY4wxZgR4ZMoYY4wxxhhjRoCDKWOMMcYYY4wZAQ6mjDHGGGOMMWYE\nOJgyxhhjjDHGmBHgYMoYY4wxxhhjRoCDKWOMMcYYY4wZAf8HQSMSr914EEkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e3058278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_threefold_split()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Size of training set: 84   size of validation set: 28   size of test set: 38\n",
      "\n",
      "Best score on validation set: 0.96\n",
      "Best parameters:  {'C': 10, 'gamma': 0.001}\n",
      "Test set score with best parameters: 0.92\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "# split data into train+validation set and test set\n",
    "X_trainval, X_test, y_trainval, y_test = train_test_split(\n",
    "    iris.data, iris.target, random_state=0)\n",
    "# split train+validation set into training and validation set\n",
    "X_train, X_valid, y_train, y_valid = train_test_split(\n",
    "    X_trainval, y_trainval, random_state=1)\n",
    "print(\"Size of training set: {}   size of validation set: {}   size of test set:\"\n",
    "      \" {}\\n\".format(X_train.shape[0], X_valid.shape[0], X_test.shape[0]))\n",
    "\n",
    "best_score = 0\n",
    "\n",
    "for gamma in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "    for C in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "        # for each combination of parameters\n",
    "        # train an SVC\n",
    "        svm = SVC(gamma=gamma, C=C)\n",
    "        svm.fit(X_train, y_train)\n",
    "        # evaluate the SVC on the test set \n",
    "        score = svm.score(X_valid, y_valid)\n",
    "        # if we got a better score, store the score and parameters\n",
    "        if score > best_score:\n",
    "            best_score = score\n",
    "            best_parameters = {'C': C, 'gamma': gamma}\n",
    "\n",
    "# rebuild a model on the combined training and validation set,\n",
    "# and evaluate it on the test set\n",
    "svm = SVC(**best_parameters)\n",
    "svm.fit(X_trainval, y_trainval)\n",
    "test_score = svm.score(X_test, y_test)\n",
    "print(\"Best score on validation set: {:.2f}\".format(best_score))\n",
    "print(\"Best parameters: \", best_parameters)\n",
    "print(\"Test set score with best parameters: {:.2f}\".format(test_score))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Grid-search with cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=100, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape=None, degree=3, gamma=0.01, kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# reference: manual_grid_search_cv\n",
    "for gamma in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "    for C in [0.001, 0.01, 0.1, 1, 10, 100]:\n",
    "        # for each combination of parameters\n",
    "        # train an SVC\n",
    "        svm = SVC(gamma=gamma, C=C)\n",
    "        # perform cross-validation\n",
    "        scores = cross_val_score(svm, X_trainval, y_trainval, cv=5)\n",
    "        # compute mean cross-validation accuracy\n",
    "        score = np.mean(scores)\n",
    "        # if we got a better score, store the score and parameters\n",
    "        if score > best_score:\n",
    "            best_score = score\n",
    "            best_parameters = {'C': C, 'gamma': gamma}\n",
    "# rebuild a model on the combined training and validation set\n",
    "svm = SVC(**best_parameters)\n",
    "svm.fit(X_trainval, y_trainval)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzIAAAFCCAYAAAAqiOjFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMzMzs9H5fzsu3AsZ2IyjAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e2f62dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_cross_val_selection()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "hide_input": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApwAAAGnCAYAAAAaHodWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nXTFwEhEREZGu/g8taYkDq1xHlAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e31b1400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_grid_search_overview()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parameter grid:\n",
      "{'C': [0.001, 0.01, 0.1, 1, 10, 100], 'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}\n"
     ]
    }
   ],
   "source": [
    "param_grid = {'C': [0.001, 0.01, 0.1, 1, 10, 100],\n",
    "              'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}\n",
    "print(\"Parameter grid:\\n{}\".format(param_grid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.svm import SVC\n",
    "grid_search = GridSearchCV(SVC(), param_grid, cv=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target,\n",
    "                                                    random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise',\n",
       "       estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape=None, degree=3, gamma='auto', kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False),\n",
       "       fit_params={}, iid=True, n_jobs=1,\n",
       "       param_grid={'C': [0.001, 0.01, 0.1, 1, 10, 100], 'gamma': [0.001, 0.01, 0.1, 1, 10, 100]},\n",
       "       pre_dispatch='2*n_jobs', refit=True, return_train_score=True,\n",
       "       scoring=None, verbose=0)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_search.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set score: 0.97\n"
     ]
    }
   ],
   "source": [
    "print(\"Test set score: {:.2f}\".format(grid_search.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters: {'C': 100, 'gamma': 0.01}\n",
      "Best cross-validation score: 0.97\n"
     ]
    }
   ],
   "source": [
    "print(\"Best parameters: {}\".format(grid_search.best_params_))\n",
    "print(\"Best cross-validation score: {:.2f}\".format(grid_search.best_score_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best estimator:\n",
      "SVC(C=100, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.01, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n"
     ]
    }
   ],
   "source": [
    "print(\"Best estimator:\\n{}\".format(grid_search.best_estimator_))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Analyzing the result of cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "      <th>...</th>\n",
       "      <th>std_fit_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>std_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.17e-03</td>\n",
       "      <td>4.16e-04</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>1.03e-04</td>\n",
       "      <td>2.95e-05</td>\n",
       "      <td>0.01</td>\n",
       "      <td>2.85e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.16e-03</td>\n",
       "      <td>4.40e-04</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>3.97e-05</td>\n",
       "      <td>4.30e-05</td>\n",
       "      <td>0.01</td>\n",
       "      <td>2.85e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.09e-03</td>\n",
       "      <td>4.02e-04</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>4.59e-05</td>\n",
       "      <td>2.40e-05</td>\n",
       "      <td>0.01</td>\n",
       "      <td>2.85e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.18e-03</td>\n",
       "      <td>4.37e-04</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>6.21e-05</td>\n",
       "      <td>2.89e-05</td>\n",
       "      <td>0.01</td>\n",
       "      <td>2.85e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.28e-03</td>\n",
       "      <td>4.37e-04</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>1.32e-04</td>\n",
       "      <td>5.38e-05</td>\n",
       "      <td>0.01</td>\n",
       "      <td>2.85e-03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   mean_fit_time  mean_score_time  mean_test_score  mean_train_score  \\\n",
       "0       1.17e-03         4.16e-04             0.37              0.37   \n",
       "1       1.16e-03         4.40e-04             0.37              0.37   \n",
       "2       1.09e-03         4.02e-04             0.37              0.37   \n",
       "3       1.18e-03         4.37e-04             0.37              0.37   \n",
       "4       1.28e-03         4.37e-04             0.37              0.37   \n",
       "\n",
       "        ...        std_fit_time std_score_time std_test_score  std_train_score  \n",
       "0       ...            1.03e-04       2.95e-05           0.01         2.85e-03  \n",
       "1       ...            3.97e-05       4.30e-05           0.01         2.85e-03  \n",
       "2       ...            4.59e-05       2.40e-05           0.01         2.85e-03  \n",
       "3       ...            6.21e-05       2.89e-05           0.01         2.85e-03  \n",
       "4       ...            1.32e-04       5.38e-05           0.01         2.85e-03  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "# convert to Dataframe\n",
    "results = pd.DataFrame(grid_search.cv_results_)\n",
    "# Show the first 5 rows\n",
    "display(results.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PolyCollection at 0x7f35e31ce668>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAR0AAAEKCAYAAAAvuYFYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e31b1b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "scores = np.array(results.mean_test_score).reshape(6, 6)\n",
    "\n",
    "# plot the mean cross-validation score\n",
    "mglearn.tools.heatmap(scores, xlabel='gamma', xticklabels=param_grid['gamma'],\n",
    "                      ylabel='C', yticklabels=param_grid['C'], cmap=\"viridis\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7f35e3237d30>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsIAAAElCAYAAADuu4R+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e3237da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 3, figsize=(13, 5))\n",
    "\n",
    "param_grid_linear = {'C': np.linspace(1, 2, 6),\n",
    "                     'gamma':  np.linspace(1, 2, 6)}\n",
    "\n",
    "param_grid_one_log = {'C': np.linspace(1, 2, 6),\n",
    "                     'gamma':  np.logspace(-3, 2, 6)}\n",
    "\n",
    "param_grid_range = {'C': np.logspace(-3, 2, 6),\n",
    "                     'gamma':  np.logspace(-7, -2, 6)}\n",
    "\n",
    "for param_grid, ax in zip([param_grid_linear, param_grid_one_log,\n",
    "                           param_grid_range], axes):\n",
    "    grid_search = GridSearchCV(SVC(), param_grid, cv=5)\n",
    "    grid_search.fit(X_train, y_train)\n",
    "    scores = grid_search.cv_results_['mean_test_score'].reshape(6, 6)\n",
    "\n",
    "    # plot the mean cross-validation scores\n",
    "    scores_image = mglearn.tools.heatmap(\n",
    "        scores, xlabel='gamma', ylabel='C', xticklabels=param_grid['gamma'],\n",
    "        yticklabels=param_grid['C'], cmap=\"viridis\", ax=ax)\n",
    "    \n",
    "plt.colorbar(scores_image, ax=axes.tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "List of grids:\n",
      "[{'kernel': ['rbf'], 'C': [0.001, 0.01, 0.1, 1, 10, 100], 'gamma': [0.001, 0.01, 0.1, 1, 10, 100]}, {'kernel': ['linear'], 'C': [0.001, 0.01, 0.1, 1, 10, 100]}]\n"
     ]
    }
   ],
   "source": [
    "param_grid = [{'kernel': ['rbf'],\n",
    "               'C': [0.001, 0.01, 0.1, 1, 10, 100],\n",
    "               'gamma': [0.001, 0.01, 0.1, 1, 10, 100]},\n",
    "              {'kernel': ['linear'],\n",
    "               'C': [0.001, 0.01, 0.1, 1, 10, 100]}]\n",
    "print(\"List of grids:\\n{}\".format(param_grid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters: {'kernel': 'rbf', 'C': 100, 'gamma': 0.01}\n",
      "Best cross-validation score: 0.97\n"
     ]
    }
   ],
   "source": [
    "grid_search = GridSearchCV(SVC(), param_grid, cv=5)\n",
    "grid_search.fit(X_train, y_train)\n",
    "print(\"Best parameters: {}\".format(grid_search.best_params_))\n",
    "print(\"Best cross-validation score: {:.2f}\".format(grid_search.best_score_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>...</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mean_fit_time</th>\n",
       "      <td>0.00078</td>\n",
       "      <td>0.00083</td>\n",
       "      <td>0.00079</td>\n",
       "      <td>0.00082</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00034</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>0.00032</td>\n",
       "      <td>0.00034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_score_time</th>\n",
       "      <td>0.00027</td>\n",
       "      <td>0.00028</td>\n",
       "      <td>0.00026</td>\n",
       "      <td>0.00033</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00019</td>\n",
       "      <td>0.00025</td>\n",
       "      <td>0.00019</td>\n",
       "      <td>0.00018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_test_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_C</th>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.001</td>\n",
       "      <td>...</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_gamma</th>\n",
       "      <td>0.001</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_kernel</th>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>rbf</td>\n",
       "      <td>...</td>\n",
       "      <td>linear</td>\n",
       "      <td>linear</td>\n",
       "      <td>linear</td>\n",
       "      <td>linear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>params</th>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 0.001}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 0.01}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 0.1}</td>\n",
       "      <td>{'kernel': 'rbf', 'C': 0.001, 'gamma': 1}</td>\n",
       "      <td>...</td>\n",
       "      <td>{'kernel': 'linear', 'C': 0.1}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 1}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 10}</td>\n",
       "      <td>{'kernel': 'linear', 'C': 100}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rank_test_score</th>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>27</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split0_test_score</th>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>...</td>\n",
       "      <td>0.96</td>\n",
       "      <td>1</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split0_train_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split1_test_score</th>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.35</td>\n",
       "      <td>...</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.96</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split1_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split2_test_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split2_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split3_test_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>...</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split3_train_score</th>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.37</td>\n",
       "      <td>...</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split4_test_score</th>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>...</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>split4_train_score</th>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.36</td>\n",
       "      <td>...</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_fit_time</th>\n",
       "      <td>7.5e-05</td>\n",
       "      <td>8.6e-05</td>\n",
       "      <td>0.00014</td>\n",
       "      <td>9.6e-05</td>\n",
       "      <td>...</td>\n",
       "      <td>2.8e-05</td>\n",
       "      <td>9.5e-05</td>\n",
       "      <td>1.6e-05</td>\n",
       "      <td>3.3e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_score_time</th>\n",
       "      <td>2.2e-05</td>\n",
       "      <td>1.6e-05</td>\n",
       "      <td>1.7e-05</td>\n",
       "      <td>8e-05</td>\n",
       "      <td>...</td>\n",
       "      <td>1.9e-05</td>\n",
       "      <td>5.6e-05</td>\n",
       "      <td>1.8e-05</td>\n",
       "      <td>9.5e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_test_score</th>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>0.011</td>\n",
       "      <td>...</td>\n",
       "      <td>0.033</td>\n",
       "      <td>0.022</td>\n",
       "      <td>0.034</td>\n",
       "      <td>0.034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std_train_score</th>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>...</td>\n",
       "      <td>0.012</td>\n",
       "      <td>0.0055</td>\n",
       "      <td>0.007</td>\n",
       "      <td>0.0055</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>23 rows × 42 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                               0   \\\n",
       "mean_fit_time                                             0.00078   \n",
       "mean_score_time                                           0.00027   \n",
       "mean_test_score                                              0.37   \n",
       "mean_train_score                                             0.37   \n",
       "param_C                                                     0.001   \n",
       "param_gamma                                                 0.001   \n",
       "param_kernel                                                  rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 0.001}   \n",
       "rank_test_score                                                27   \n",
       "split0_test_score                                            0.38   \n",
       "split0_train_score                                           0.36   \n",
       "split1_test_score                                            0.35   \n",
       "split1_train_score                                           0.37   \n",
       "split2_test_score                                            0.36   \n",
       "split2_train_score                                           0.37   \n",
       "split3_test_score                                            0.36   \n",
       "split3_train_score                                           0.37   \n",
       "split4_test_score                                            0.38   \n",
       "split4_train_score                                           0.36   \n",
       "std_fit_time                                              7.5e-05   \n",
       "std_score_time                                            2.2e-05   \n",
       "std_test_score                                              0.011   \n",
       "std_train_score                                            0.0029   \n",
       "\n",
       "                                                              1   \\\n",
       "mean_fit_time                                            0.00083   \n",
       "mean_score_time                                          0.00028   \n",
       "mean_test_score                                             0.37   \n",
       "mean_train_score                                            0.37   \n",
       "param_C                                                    0.001   \n",
       "param_gamma                                                 0.01   \n",
       "param_kernel                                                 rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 0.01}   \n",
       "rank_test_score                                               27   \n",
       "split0_test_score                                           0.38   \n",
       "split0_train_score                                          0.36   \n",
       "split1_test_score                                           0.35   \n",
       "split1_train_score                                          0.37   \n",
       "split2_test_score                                           0.36   \n",
       "split2_train_score                                          0.37   \n",
       "split3_test_score                                           0.36   \n",
       "split3_train_score                                          0.37   \n",
       "split4_test_score                                           0.38   \n",
       "split4_train_score                                          0.36   \n",
       "std_fit_time                                             8.6e-05   \n",
       "std_score_time                                           1.6e-05   \n",
       "std_test_score                                             0.011   \n",
       "std_train_score                                           0.0029   \n",
       "\n",
       "                                                             2   \\\n",
       "mean_fit_time                                           0.00079   \n",
       "mean_score_time                                         0.00026   \n",
       "mean_test_score                                            0.37   \n",
       "mean_train_score                                           0.37   \n",
       "param_C                                                   0.001   \n",
       "param_gamma                                                 0.1   \n",
       "param_kernel                                                rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 0.1}   \n",
       "rank_test_score                                              27   \n",
       "split0_test_score                                          0.38   \n",
       "split0_train_score                                         0.36   \n",
       "split1_test_score                                          0.35   \n",
       "split1_train_score                                         0.37   \n",
       "split2_test_score                                          0.36   \n",
       "split2_train_score                                         0.37   \n",
       "split3_test_score                                          0.36   \n",
       "split3_train_score                                         0.37   \n",
       "split4_test_score                                          0.38   \n",
       "split4_train_score                                         0.36   \n",
       "std_fit_time                                            0.00014   \n",
       "std_score_time                                          1.7e-05   \n",
       "std_test_score                                            0.011   \n",
       "std_train_score                                          0.0029   \n",
       "\n",
       "                                                           3   \\\n",
       "mean_fit_time                                         0.00082   \n",
       "mean_score_time                                       0.00033   \n",
       "mean_test_score                                          0.37   \n",
       "mean_train_score                                         0.37   \n",
       "param_C                                                 0.001   \n",
       "param_gamma                                                 1   \n",
       "param_kernel                                              rbf   \n",
       "params              {'kernel': 'rbf', 'C': 0.001, 'gamma': 1}   \n",
       "rank_test_score                                            27   \n",
       "split0_test_score                                        0.38   \n",
       "split0_train_score                                       0.36   \n",
       "split1_test_score                                        0.35   \n",
       "split1_train_score                                       0.37   \n",
       "split2_test_score                                        0.36   \n",
       "split2_train_score                                       0.37   \n",
       "split3_test_score                                        0.36   \n",
       "split3_train_score                                       0.37   \n",
       "split4_test_score                                        0.38   \n",
       "split4_train_score                                       0.36   \n",
       "std_fit_time                                          9.6e-05   \n",
       "std_score_time                                          8e-05   \n",
       "std_test_score                                          0.011   \n",
       "std_train_score                                        0.0029   \n",
       "\n",
       "                                 ...                \\\n",
       "mean_fit_time                    ...                 \n",
       "mean_score_time                  ...                 \n",
       "mean_test_score                  ...                 \n",
       "mean_train_score                 ...                 \n",
       "param_C                          ...                 \n",
       "param_gamma                      ...                 \n",
       "param_kernel                     ...                 \n",
       "params                           ...                 \n",
       "rank_test_score                  ...                 \n",
       "split0_test_score                ...                 \n",
       "split0_train_score               ...                 \n",
       "split1_test_score                ...                 \n",
       "split1_train_score               ...                 \n",
       "split2_test_score                ...                 \n",
       "split2_train_score               ...                 \n",
       "split3_test_score                ...                 \n",
       "split3_train_score               ...                 \n",
       "split4_test_score                ...                 \n",
       "split4_train_score               ...                 \n",
       "std_fit_time                     ...                 \n",
       "std_score_time                   ...                 \n",
       "std_test_score                   ...                 \n",
       "std_train_score                  ...                 \n",
       "\n",
       "                                                38  \\\n",
       "mean_fit_time                              0.00034   \n",
       "mean_score_time                            0.00019   \n",
       "mean_test_score                               0.95   \n",
       "mean_train_score                              0.97   \n",
       "param_C                                        0.1   \n",
       "param_gamma                                    NaN   \n",
       "param_kernel                                linear   \n",
       "params              {'kernel': 'linear', 'C': 0.1}   \n",
       "rank_test_score                                 11   \n",
       "split0_test_score                             0.96   \n",
       "split0_train_score                            0.97   \n",
       "split1_test_score                             0.91   \n",
       "split1_train_score                            0.98   \n",
       "split2_test_score                                1   \n",
       "split2_train_score                            0.94   \n",
       "split3_test_score                             0.91   \n",
       "split3_train_score                            0.98   \n",
       "split4_test_score                             0.95   \n",
       "split4_train_score                            0.97   \n",
       "std_fit_time                               2.8e-05   \n",
       "std_score_time                             1.9e-05   \n",
       "std_test_score                               0.033   \n",
       "std_train_score                              0.012   \n",
       "\n",
       "                                              39  \\\n",
       "mean_fit_time                             0.0004   \n",
       "mean_score_time                          0.00025   \n",
       "mean_test_score                             0.97   \n",
       "mean_train_score                            0.98   \n",
       "param_C                                        1   \n",
       "param_gamma                                  NaN   \n",
       "param_kernel                              linear   \n",
       "params              {'kernel': 'linear', 'C': 1}   \n",
       "rank_test_score                                1   \n",
       "split0_test_score                              1   \n",
       "split0_train_score                          0.99   \n",
       "split1_test_score                           0.96   \n",
       "split1_train_score                          0.98   \n",
       "split2_test_score                              1   \n",
       "split2_train_score                          0.98   \n",
       "split3_test_score                           0.95   \n",
       "split3_train_score                          0.99   \n",
       "split4_test_score                           0.95   \n",
       "split4_train_score                          0.99   \n",
       "std_fit_time                             9.5e-05   \n",
       "std_score_time                           5.6e-05   \n",
       "std_test_score                             0.022   \n",
       "std_train_score                           0.0055   \n",
       "\n",
       "                                               40  \\\n",
       "mean_fit_time                             0.00032   \n",
       "mean_score_time                           0.00019   \n",
       "mean_test_score                              0.96   \n",
       "mean_train_score                             0.99   \n",
       "param_C                                        10   \n",
       "param_gamma                                   NaN   \n",
       "param_kernel                               linear   \n",
       "params              {'kernel': 'linear', 'C': 10}   \n",
       "rank_test_score                                 3   \n",
       "split0_test_score                            0.96   \n",
       "split0_train_score                           0.99   \n",
       "split1_test_score                               1   \n",
       "split1_train_score                           0.99   \n",
       "split2_test_score                               1   \n",
       "split2_train_score                           0.98   \n",
       "split3_test_score                            0.91   \n",
       "split3_train_score                           0.99   \n",
       "split4_test_score                            0.95   \n",
       "split4_train_score                              1   \n",
       "std_fit_time                              1.6e-05   \n",
       "std_score_time                            1.8e-05   \n",
       "std_test_score                              0.034   \n",
       "std_train_score                             0.007   \n",
       "\n",
       "                                                41  \n",
       "mean_fit_time                              0.00034  \n",
       "mean_score_time                            0.00018  \n",
       "mean_test_score                               0.96  \n",
       "mean_train_score                              0.99  \n",
       "param_C                                        100  \n",
       "param_gamma                                    NaN  \n",
       "param_kernel                                linear  \n",
       "params              {'kernel': 'linear', 'C': 100}  \n",
       "rank_test_score                                  3  \n",
       "split0_test_score                             0.96  \n",
       "split0_train_score                            0.99  \n",
       "split1_test_score                                1  \n",
       "split1_train_score                            0.99  \n",
       "split2_test_score                                1  \n",
       "split2_train_score                            0.99  \n",
       "split3_test_score                             0.91  \n",
       "split3_train_score                               1  \n",
       "split4_test_score                             0.95  \n",
       "split4_train_score                               1  \n",
       "std_fit_time                               3.3e-05  \n",
       "std_score_time                             9.5e-06  \n",
       "std_test_score                               0.034  \n",
       "std_train_score                             0.0055  \n",
       "\n",
       "[23 rows x 42 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "results = pd.DataFrame(grid_search.cv_results_)\n",
    "# we display the transposed table so that it better fits on the page:\n",
    "display(results.T)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Using different cross-validation strategies with grid-search"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Nested cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validation scores:  [ 0.967  1.     0.9    0.967  1.   ]\n",
      "Mean cross-validation score:  0.966666666667\n"
     ]
    }
   ],
   "source": [
    "scores = cross_val_score(GridSearchCV(SVC(), param_grid, cv=5),\n",
    "                         iris.data, iris.target, cv=5)\n",
    "print(\"Cross-validation scores: \", scores)\n",
    "print(\"Mean cross-validation score: \", scores.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def nested_cv(X, y, inner_cv, outer_cv, Classifier, parameter_grid):\n",
    "    outer_scores = []\n",
    "    # for each split of the data in the outer cross-validation\n",
    "    # (split method returns indices)\n",
    "    for training_samples, test_samples in outer_cv.split(X, y):\n",
    "        # find best parameter using inner cross-validation:\n",
    "        best_parms = {}\n",
    "        best_score = -np.inf\n",
    "        # iterate over parameters\n",
    "        for parameters in parameter_grid:\n",
    "            # accumulate score over inner splits\n",
    "            cv_scores = []\n",
    "            # iterate over inner cross-validation\n",
    "            for inner_train, inner_test in inner_cv.split(\n",
    "                    X[training_samples], y[training_samples]):\n",
    "                # build classifier given parameters and training data\n",
    "                clf = Classifier(**parameters)\n",
    "                clf.fit(X[inner_train], y[inner_train])\n",
    "                # evaluate on inner test set\n",
    "                score = clf.score(X[inner_test], y[inner_test])\n",
    "                cv_scores.append(score)\n",
    "            # compute mean score over inner folds\n",
    "            mean_score = np.mean(cv_scores)\n",
    "            if mean_score > best_score:\n",
    "                # if better than so far, remember parameters\n",
    "                best_score = mean_score\n",
    "                best_params = parameters\n",
    "        # build classifier on best parameters using outer training set\n",
    "        clf = Classifier(**best_params)\n",
    "        clf.fit(X[training_samples], y[training_samples])\n",
    "        # evaluate \n",
    "        outer_scores.append(clf.score(X[test_samples], y[test_samples]))\n",
    "    return np.array(outer_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validation scores: [ 0.967  1.     0.967  0.967  1.   ]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import ParameterGrid, StratifiedKFold\n",
    "scores = nested_cv(iris.data, iris.target, StratifiedKFold(5),\n",
    "                   StratifiedKFold(5), SVC, ParameterGrid(param_grid))\n",
    "print(\"Cross-validation scores: {}\".format(scores))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "#### Parallelizing cross-validation and grid-search"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluation Metrics and scoring\n",
    "#### Keep the end-goal in mind"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Metrics for binary classification\n",
    "##### Kinds of Errors\n",
    "##### Imbalanced datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_digits\n",
    "\n",
    "digits = load_digits()\n",
    "y = digits.target == 9\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    digits.data, y, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unique predicted labels: [False]\n",
      "Test score: 0.90\n"
     ]
    }
   ],
   "source": [
    "from sklearn.dummy import DummyClassifier\n",
    "dummy_majority = DummyClassifier(strategy='most_frequent').fit(X_train, y_train)\n",
    "pred_most_frequent = dummy_majority.predict(X_test)\n",
    "print(\"Unique predicted labels: {}\".format(np.unique(pred_most_frequent)))\n",
    "print(\"Test score: {:.2f}\".format(dummy_majority.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.92\n"
     ]
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "tree = DecisionTreeClassifier(max_depth=2).fit(X_train, y_train)\n",
    "pred_tree = tree.predict(X_test)\n",
    "print(\"Test score: {:.2f}\".format(tree.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dummy score: 0.82\n",
      "logreg score: 0.98\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "dummy = DummyClassifier().fit(X_train, y_train)\n",
    "pred_dummy = dummy.predict(X_test)\n",
    "print(\"dummy score: {:.2f}\".format(dummy.score(X_test, y_test)))\n",
    "\n",
    "logreg = LogisticRegression(C=0.1).fit(X_train, y_train)\n",
    "pred_logreg = logreg.predict(X_test)\n",
    "print(\"logreg score: {:.2f}\".format(logreg.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Confusion matrices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix:\n",
      "[[401   2]\n",
      " [  8  39]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "confusion = confusion_matrix(y_test, pred_logreg)\n",
    "print(\"Confusion matrix:\\n{}\".format(confusion))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAl0AAAHjCAYAAAAZqqqzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\naoGdLkmSpBbY6ZIkSWrBXwzcQ4qqM/JQAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e383c7b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_confusion_matrix_illustration()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAc0AAAEACAYAAADRMy13AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e2d3c128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_binary_confusion_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Most frequent class:\n",
      "[[403   0]\n",
      " [ 47   0]]\n",
      "\n",
      "Dummy model:\n",
      "[[362  41]\n",
      " [ 45   2]]\n",
      "\n",
      "Decision tree:\n",
      "[[390  13]\n",
      " [ 24  23]]\n",
      "\n",
      "Logistic Regression\n",
      "[[401   2]\n",
      " [  8  39]]\n"
     ]
    }
   ],
   "source": [
    "print(\"Most frequent class:\")\n",
    "print(confusion_matrix(y_test, pred_most_frequent))\n",
    "print(\"\\nDummy model:\")\n",
    "print(confusion_matrix(y_test, pred_dummy))\n",
    "print(\"\\nDecision tree:\")\n",
    "print(confusion_matrix(y_test, pred_tree))\n",
    "print(\"\\nLogistic Regression\")\n",
    "print(confusion_matrix(y_test, pred_logreg))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Relation to accuracy\n",
    "\\begin{equation}\n",
    "\\text{Accuracy} = \\frac{\\text{TP} + \\text{TN}}{\\text{TP} + \\text{TN} + \\text{FP} + \\text{FN}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Precision, recall and f-score\n",
    "\\begin{equation}\n",
    "\\text{Precision} = \\frac{\\text{TP}}{\\text{TP} + \\text{FP}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "hide_input": false
   },
   "source": [
    "\\begin{equation}\n",
    "\\text{Recall} = \\frac{\\text{TP}}{\\text{TP} + \\text{FN}}\n",
    "\\end{equation}\n",
    "\\begin{equation}\n",
    "\\text{F} = 2 \\cdot \\frac{\\text{precision} \\cdot \\text{recall}}{\\text{precision} + \\text{recall}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f1 score most frequent: 0.00\n",
      "f1 score dummy: 0.04\n",
      "f1 score tree: 0.55\n",
      "f1 score logistic regression: 0.89\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/andy/checkout/scikit-learn/sklearn/metrics/classification.py:1113: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "print(\"f1 score most frequent: {:.2f}\".format(\n",
    "    f1_score(y_test, pred_most_frequent)))\n",
    "print(\"f1 score dummy: {:.2f}\".format(f1_score(y_test, pred_dummy)))\n",
    "print(\"f1 score tree: {:.2f}\".format(f1_score(y_test, pred_tree)))\n",
    "print(\"f1 score logistic regression: {:.2f}\".format(\n",
    "    f1_score(y_test, pred_logreg)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "   not nine       0.90      1.00      0.94       403\n",
      "       nine       0.00      0.00      0.00        47\n",
      "\n",
      "avg / total       0.80      0.90      0.85       450\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/andy/checkout/scikit-learn/sklearn/metrics/classification.py:1113: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_test, pred_most_frequent,\n",
    "                            target_names=[\"not nine\", \"nine\"]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "   not nine       0.89      0.90      0.89       403\n",
      "       nine       0.05      0.04      0.04        47\n",
      "\n",
      "avg / total       0.80      0.81      0.81       450\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, pred_dummy,\n",
    "                            target_names=[\"not nine\", \"nine\"]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "   not nine       0.98      1.00      0.99       403\n",
      "       nine       0.95      0.83      0.89        47\n",
      "\n",
      "avg / total       0.98      0.98      0.98       450\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, pred_logreg,\n",
    "                            target_names=[\"not nine\", \"nine\"]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Taking uncertainty into account"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from mglearn.datasets import make_blobs \n",
    "X, y = make_blobs(n_samples=(400, 50), centers=2, cluster_std=[7.0, 2],        \n",
    "                  random_state=22)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "svc = SVC(gamma=.05).fit(X_train, y_train)                                     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1MAAAH6CAYAAADiNBaTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIiIyCCqmREREREREBuH/A4+hTjmHV2qkAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e32d6dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mglearn.plots.plot_decision_threshold()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.97      0.89      0.93       104\n",
      "          1       0.35      0.67      0.46         9\n",
      "\n",
      "avg / total       0.92      0.88      0.89       113\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, svc.predict(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "y_pred_lower_threshold = svc.decision_function(X_test) > -.8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       1.00      0.82      0.90       104\n",
      "          1       0.32      1.00      0.49         9\n",
      "\n",
      "avg / total       0.95      0.83      0.87       113\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred_lower_threshold))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Precision-Recall curves and ROC curves"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "precision, recall, thresholds = precision_recall_curve(\n",
    "    y_test, svc.decision_function(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f35e2ec5550>"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEPCAYAAABY9lNGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e32fca90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create a similar dataset as before, but with more samples\n",
    "# to get a smoother curve\n",
    "X, y = make_blobs(n_samples=(4000, 500), centers=2, cluster_std=[7.0, 2],\n",
    "                  random_state=22)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "\n",
    "svc = SVC(gamma=.05).fit(X_train, y_train)\n",
    "\n",
    "precision, recall, thresholds = precision_recall_curve(\n",
    "    y_test, svc.decision_function(X_test))\n",
    "# find threshold closest to zero:\n",
    "close_zero = np.argmin(np.abs(thresholds))\n",
    "plt.plot(precision[close_zero], recall[close_zero], 'o', markersize=10,\n",
    "         label=\"threshold zero\", fillstyle=\"none\", c='k', mew=2)\n",
    "\n",
    "plt.plot(precision, recall, label=\"precision recall curve\")\n",
    "plt.xlabel(\"Precision\")\n",
    "plt.ylabel(\"Recall\")\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f35e2da7e10>"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEPCAYAAABY9lNGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkCsIMzMr5ArCzMwKuYIwM7NC/wdNsx6jLlZ0+gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e2e3b978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "rf = RandomForestClassifier(n_estimators=100, random_state=0, max_features=2)\n",
    "rf.fit(X_train, y_train)\n",
    "\n",
    "# RandomForestClassifier has predict_proba, but not decision_function\n",
    "precision_rf, recall_rf, thresholds_rf = precision_recall_curve(\n",
    "    y_test, rf.predict_proba(X_test)[:, 1])\n",
    "\n",
    "plt.plot(precision, recall, label=\"svc\")\n",
    "\n",
    "plt.plot(precision[close_zero], recall[close_zero], 'o', markersize=10,\n",
    "         label=\"threshold zero svc\", fillstyle=\"none\", c='k', mew=2)\n",
    "\n",
    "plt.plot(precision_rf, recall_rf, label=\"rf\")\n",
    "\n",
    "close_default_rf = np.argmin(np.abs(thresholds_rf - 0.5))\n",
    "plt.plot(precision_rf[close_default_rf], recall_rf[close_default_rf], '^', c='k',\n",
    "         markersize=10, label=\"threshold 0.5 rf\", fillstyle=\"none\", mew=2)\n",
    "plt.xlabel(\"Precision\")\n",
    "plt.ylabel(\"Recall\")\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f1_score of random forest: 0.610\n",
      "f1_score of svc: 0.656\n"
     ]
    }
   ],
   "source": [
    "print(\"f1_score of random forest: {:.3f}\".format(\n",
    "        f1_score(y_test, rf.predict(X_test))))\n",
    "print(\"f1_score of svc: {:.3f}\".format(f1_score(y_test, svc.predict(X_test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average precision of random forest: 0.666\n",
      "Average precision of svc: 0.663\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import average_precision_score\n",
    "ap_rf = average_precision_score(y_test, rf.predict_proba(X_test)[:, 1])\n",
    "ap_svc = average_precision_score(y_test, svc.decision_function(X_test))\n",
    "print(\"Average precision of random forest: {:.3f}\".format(ap_rf))\n",
    "print(\"Average precision of svc: {:.3f}\".format(ap_svc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Receiver Operating Characteristics (ROC) and AUC\n",
    "\\begin{equation}\n",
    "\\text{FPR} = \\frac{\\text{FP}}{\\text{FP} + \\text{TN}}\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f35e2e8acf8>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEPCAYAAABY9lNGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJQgREYlICUJERCL6/xeg7PzX1LZrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e2e8af98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "fpr, tpr, thresholds = roc_curve(y_test, svc.decision_function(X_test))\n",
    "\n",
    "plt.plot(fpr, tpr, label=\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR (recall)\")\n",
    "# find threshold closest to zero:\n",
    "close_zero = np.argmin(np.abs(thresholds))\n",
    "plt.plot(fpr[close_zero], tpr[close_zero], 'o', markersize=10,\n",
    "         label=\"threshold zero\", fillstyle=\"none\", c='k', mew=2)\n",
    "plt.legend(loc=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f35e0b5bbe0>"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEPCAYAAABY9lNGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlHLo/wOwO3Dx/An/QgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e2cac5c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "fpr_rf, tpr_rf, thresholds_rf = roc_curve(y_test, rf.predict_proba(X_test)[:, 1])\n",
    "\n",
    "plt.plot(fpr, tpr, label=\"ROC Curve SVC\")\n",
    "plt.plot(fpr_rf, tpr_rf, label=\"ROC Curve RF\")\n",
    "\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR (recall)\")\n",
    "plt.plot(fpr[close_zero], tpr[close_zero], 'o', markersize=10,\n",
    "         label=\"threshold zero SVC\", fillstyle=\"none\", c='k', mew=2)\n",
    "close_default_rf = np.argmin(np.abs(thresholds_rf - 0.5))\n",
    "plt.plot(fpr_rf[close_default_rf], tpr[close_default_rf], '^', markersize=10,\n",
    "         label=\"threshold 0.5 RF\", fillstyle=\"none\", c='k', mew=2)\n",
    "\n",
    "plt.legend(loc=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AUC for Random Forest: 0.937\n",
      "AUC for SVC: 0.916\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "rf_auc = roc_auc_score(y_test, rf.predict_proba(X_test)[:, 1])\n",
    "svc_auc = roc_auc_score(y_test, svc.decision_function(X_test))\n",
    "print(\"AUC for Random Forest: {:.3f}\".format(rf_auc))\n",
    "print(\"AUC for SVC: {:.3f}\".format(svc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gamma = 1.00  accuracy = 0.90  AUC = 0.50\n",
      "gamma = 0.05  accuracy = 0.90  AUC = 1.00\n",
      "gamma = 0.01  accuracy = 0.90  AUC = 1.00\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f35e2bf9c50>"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAELCAYAAADDZxFQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e2bf9da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y = digits.target == 9\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    digits.data, y, random_state=0)\n",
    "\n",
    "plt.figure()\n",
    "\n",
    "for gamma in [1, 0.05, 0.01]:\n",
    "    svc = SVC(gamma=gamma).fit(X_train, y_train)\n",
    "    accuracy = svc.score(X_test, y_test)\n",
    "    auc = roc_auc_score(y_test, svc.decision_function(X_test))\n",
    "    fpr, tpr, _ = roc_curve(y_test , svc.decision_function(X_test))\n",
    "    print(\"gamma = {:.2f}  accuracy = {:.2f}  AUC = {:.2f}\".format(\n",
    "            gamma, accuracy, auc))\n",
    "    plt.plot(fpr, tpr, label=\"gamma={:.3f}\".format(gamma))\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.xlim(-0.01, 1)\n",
    "plt.ylim(0, 1.02)\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Multi-class classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.953\n",
      "Confusion matrix:\n",
      "[[37  0  0  0  0  0  0  0  0  0]\n",
      " [ 0 39  0  0  0  0  2  0  2  0]\n",
      " [ 0  0 41  3  0  0  0  0  0  0]\n",
      " [ 0  0  1 43  0  0  0  0  0  1]\n",
      " [ 0  0  0  0 38  0  0  0  0  0]\n",
      " [ 0  1  0  0  0 47  0  0  0  0]\n",
      " [ 0  0  0  0  0  0 52  0  0  0]\n",
      " [ 0  1  0  1  1  0  0 45  0  0]\n",
      " [ 0  3  1  0  0  0  0  0 43  1]\n",
      " [ 0  0  0  1  0  1  0  0  1 44]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    digits.data, digits.target, random_state=0)\n",
    "lr = LogisticRegression().fit(X_train, y_train)\n",
    "pred = lr.predict(X_test)\n",
    "print(\"Accuracy: {:.3f}\".format(accuracy_score(y_test, pred)))\n",
    "print(\"Confusion matrix:\\n{}\".format(confusion_matrix(y_test, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": false,
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQYAAAEZCAYAAAB1tYtyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35e2c81dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "scores_image = mglearn.tools.heatmap(\n",
    "    confusion_matrix(y_test, pred), xlabel='Predicted label',\n",
    "    ylabel='True label', xticklabels=digits.target_names,\n",
    "    yticklabels=digits.target_names, cmap=plt.cm.gray_r, fmt=\"%d\")    \n",
    "plt.title(\"Confusion matrix\")\n",
    "plt.gca().invert_yaxis()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       1.00      1.00      1.00        37\n",
      "          1       0.89      0.91      0.90        43\n",
      "          2       0.95      0.93      0.94        44\n",
      "          3       0.90      0.96      0.92        45\n",
      "          4       0.97      1.00      0.99        38\n",
      "          5       0.98      0.98      0.98        48\n",
      "          6       0.96      1.00      0.98        52\n",
      "          7       1.00      0.94      0.97        48\n",
      "          8       0.93      0.90      0.91        48\n",
      "          9       0.96      0.94      0.95        47\n",
      "\n",
      "avg / total       0.95      0.95      0.95       450\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Micro average f1 score: 0.953\n",
      "Macro average f1 score: 0.954\n"
     ]
    }
   ],
   "source": [
    "print(\"Micro average f1 score: {:.3f}\".format(f1_score(y_test, pred, average=\"micro\")))\n",
    "print(\"Macro average f1 score: {:.3f}\".format(f1_score(y_test, pred, average=\"macro\")))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Regression metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using evaluation metrics in model selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Default scoring: [ 0.9  0.9  0.9]\n",
      "Explicit accuracy scoring: [ 0.9  0.9  0.9]\n",
      "AUC scoring: [ 0.994  0.99   0.996]\n"
     ]
    }
   ],
   "source": [
    "# default scoring for classification is accuracy\n",
    "print(\"Default scoring: {}\".format(\n",
    "      cross_val_score(SVC(), digits.data, digits.target == 9)))\n",
    "# providing scoring=\"accuracy\" doesn't change the results\n",
    "explicit_accuracy =  cross_val_score(SVC(), digits.data, digits.target == 9,\n",
    "                                     scoring=\"accuracy\")\n",
    "print(\"Explicit accuracy scoring: {}\".format(explicit_accuracy))\n",
    "roc_auc =  cross_val_score(SVC(), digits.data, digits.target == 9,\n",
    "                           scoring=\"roc_auc\")\n",
    "print(\"AUC scoring: {}\".format(roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Grid-Search with accuracy\n",
      "Best parameters: {'gamma': 0.0001}\n",
      "Best cross-validation score (accuracy)): 0.970\n",
      "Test set AUC: 0.992\n",
      "Test set accuracy: 0.973\n",
      "\n",
      "Grid-Search with AUC\n",
      "Best parameters: {'gamma': 0.01}\n",
      "Best cross-validation score (AUC): 0.997\n",
      "Test set AUC: 1.000\n",
      "Test set accuracy: 1.000\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    digits.data, digits.target == 9, random_state=0)\n",
    "\n",
    "# we provide a somewhat bad grid to illustrate the point:\n",
    "param_grid = {'gamma': [0.0001, 0.01, 0.1, 1, 10]}\n",
    "# using the default scoring of accuracy:\n",
    "grid = GridSearchCV(SVC(), param_grid=param_grid)\n",
    "grid.fit(X_train, y_train)\n",
    "print(\"Grid-Search with accuracy\")\n",
    "print(\"Best parameters:\", grid.best_params_)\n",
    "print(\"Best cross-validation score (accuracy)): {:.3f}\".format(grid.best_score_))\n",
    "print(\"Test set AUC: {:.3f}\".format(\n",
    "        roc_auc_score(y_test, grid.decision_function(X_test))))\n",
    "print(\"Test set accuracy: {:.3f}\".format(grid.score(X_test, y_test)))\n",
    "\n",
    "# using AUC scoring instead:\n",
    "grid = GridSearchCV(SVC(), param_grid=param_grid, scoring=\"roc_auc\")\n",
    "grid.fit(X_train, y_train)\n",
    "print(\"\\nGrid-Search with AUC\")\n",
    "print(\"Best parameters:\", grid.best_params_)\n",
    "print(\"Best cross-validation score (AUC): {:.3f}\".format(grid.best_score_))\n",
    "print(\"Test set AUC: {:.3f}\".format(\n",
    "        roc_auc_score(y_test, grid.decision_function(X_test))))\n",
    "print(\"Test set accuracy: {:.3f}\".format(grid.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available scorers:\n",
      "['accuracy', 'adjusted_rand_score', 'average_precision', 'f1', 'f1_macro', 'f1_micro', 'f1_samples', 'f1_weighted', 'log_loss', 'mean_absolute_error', 'mean_squared_error', 'median_absolute_error', 'neg_log_loss', 'neg_mean_absolute_error', 'neg_mean_squared_error', 'neg_median_absolute_error', 'precision', 'precision_macro', 'precision_micro', 'precision_samples', 'precision_weighted', 'r2', 'recall', 'recall_macro', 'recall_micro', 'recall_samples', 'recall_weighted', 'roc_auc']\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics.scorer import SCORERS\n",
    "print(\"Available scorers:\\n{}\".format(sorted(SCORERS.keys())))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Summary and outlook"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}