{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b9dff11d",
   "metadata": {},
   "source": [
    "Summary - Pattern Recognition\n",
    "===============================\n",
    "\n",
    "This notebook is a summary of many of the techniques we have looked at in pattern recognition,\n",
    "as well the interfaces and classes used to implement them in the `scikit.learn` package."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6c7187cf",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from pylab import *\n",
    "from scipy.spatial import distance\n",
    "import tables\n",
    "import random as pyrandom"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47602d2f",
   "metadata": {},
   "source": [
    "MNIST Training and Test Data\n",
    "======================\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32a3f95f",
   "metadata": {},
   "source": [
    "We have tried out a lot of these algorithms on the MNIST data set.\n",
    "The MNIST data set has been widely used for experiments in pattern\n",
    "recognition.\n",
    "Being based on images, it's actually easy to see what the various\n",
    "preprocessing methods do with the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2bb7cd6e",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "mnist = tables.openFile(\"mnist.h5\",\"r\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa3fa63f",
   "metadata": {},
   "source": [
    "Most algorithms work on 1D vectors, while input images are 2D.\n",
    "Furthermore, training data sets are, by convention, matrices whose rows represent training data vectors.\n",
    "The `make2d` function is useful for converting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c27c907b",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def make2d(data):\n",
    "    return data.reshape(len(data),prod(data.shape[1:]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "675cfd37",
   "metadata": {},
   "source": [
    "This is the training data..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5b46fd73",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data = make2d(array(mnist.root.deskewed,'f'))\n",
    "labels = array(mnist.root.labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc4aa5ee",
   "metadata": {},
   "source": [
    "And this is a separate test set..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a42ecd42",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "test_data = make2d(array(mnist.root.test_deskewed,'f'))\n",
    "test_labels = array(mnist.root.test_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48dc37c3",
   "metadata": {},
   "source": [
    "k-Nearest Neighbors\n",
    "==================="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "77746dbd",
   "metadata": {},
   "source": [
    "One of the simplest and most intuitive algorithms for classification\n",
    "is the nearest neighbor classifier.\n",
    "You should always try it out to get some idea of how difficult\n",
    "a classification problem is.\n",
    "For large amounts of training data, the k-NN classifier can be slow,\n",
    "but the implementation in SciKit has various optimizations that can\n",
    "make it a lot faster than the implementation you created yourself."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2c5b9e2f",
   "metadata": {},
   "source": [
    "Scikit classifiers generally start out as an object that is initialized with some parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a8423efe",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "knn = KNeighborsClassifier(n_neighbors=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f3a05322",
   "metadata": {},
   "source": [
    "Training is performed with the `fit` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "47edf44c",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, n_neighbors=1,\n",
       "           warn_on_equidistant=True, weights='uniform')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.fit(data[:1000],labels[:1000])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca363d64",
   "metadata": {},
   "source": [
    "Classification is carried out with the `predict` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f358ed4e",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[7 2 1 0 4 1 9 4 9 9]\n",
      "[7 2 1 0 4 1 4 9 5 9]"
     ]
    }
   ],
   "source": [
    "pred = knn.predict(test_data)\n",
    "print pred[:10]\n",
    "print test_labels[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b32f9870",
   "metadata": {},
   "source": [
    "In the past, we have evaluated predictions directly by comparing labels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "85b5f74d",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.076100000000000001"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(pred!=test_labels)*1.0/len(test_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "609a0c64",
   "metadata": {},
   "source": [
    "However, the package `sklearn.metrics` contains\n",
    "various evaluation functions.\n",
    "The `zero_one_score` function computes the fraction of _correct_ results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fb010cc7",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.92390000000000005"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5dfa0b20",
   "metadata": {},
   "source": [
    "Often, the error rate is more useful."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "7270e0d3",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.076099999999999945"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1-metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8b17a90",
   "metadata": {},
   "source": [
    "Linear Discriminant Analysis\n",
    "=============================="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88280eba",
   "metadata": {},
   "source": [
    "Linear discriminant analysis (LDA) is one way of obtaining a _linear classifier_.\n",
    "It is actually based on parametric models, however, not perceptron learning or other\n",
    "_discriminative_ methods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "60671fd4",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import lda\n",
    "lc = lda.LDA()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2134d759",
   "metadata": {},
   "source": [
    "As before, the `fit` method does the training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8ad9bb6c",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LDA(n_components=None, priors=None)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lc.fit(data,labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0467218c",
   "metadata": {},
   "source": [
    "The `predict` method does the prediction, and we use the `zero_one_score` function to evaluate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "d04b1751",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.081999999999999962"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = lc.predict(test_data)\n",
    "1-metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ecc888e",
   "metadata": {},
   "source": [
    "Preprocessing\n",
    "============"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f69fdc08",
   "metadata": {},
   "source": [
    "We had talked about various forms of preprocessing.\n",
    "Often, feature vector normalization is important.\n",
    "For example, we might transform $v$ into $v/||v||$.\n",
    "\n",
    "Some preprocessing methods need to be trained with a `fit` method,\n",
    "just like a classifier,\n",
    "but instead of a `pred` method, there is a `transform` method.\n",
    "\n",
    "The `fit_transform` method fits a model and transforms the input data at the same time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "20e4cad3",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn import preprocessing\n",
    "normalizer = preprocessing.Normalizer(norm='l2')\n",
    "normalized_data = normalizer.fit_transform(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "176cc95a",
   "metadata": {},
   "source": [
    "Now we can fit a classifier in terms of the transformed data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "59975f3d",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LDA(n_components=None, priors=None)"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lc = lda.LDA()\n",
    "lc.fit(normalized_data,labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16169d97",
   "metadata": {},
   "source": [
    "We also need to remember to transform the test data prior to classification."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "4e359d44",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.077500000000000013"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = lc.predict(normalizer.transform(test_data))\n",
    "1-metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c01be287",
   "metadata": {},
   "source": [
    "Preprocessing is important for achieving low error rates,\n",
    "but many attempts at preprocessing make classification worse, not better.\n",
    "Finding good preprocessing methods is tricky and requires both experience\n",
    "and experimentation."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6675780a",
   "metadata": {},
   "source": [
    "Feature Extraction\n",
    "=================="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52be5255",
   "metadata": {},
   "source": [
    "Feature extraction is a particular form of preprocessing.\n",
    "There really isn't any clear distinction between feature extraction,\n",
    "preprocessing, normalization, etc.\n",
    "Normally, feature extraction aims at reducing the amount of data\n",
    "and information that needs to be processed by a classifier,\n",
    "and often it attempts to enhance features of the input that are\n",
    "relevant to classification and discard parts of the input that\n",
    "are noise or irrelevant."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "id": "187aa313",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from scipy.ndimage.filters import gaussian_filter\n",
    "def extract(data):\n",
    "    return array([gaussian_filter(x.reshape(28,28),2.0).ravel() for x in data])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db256396",
   "metadata": {},
   "source": [
    "Feature extraction is often hard-coded by hand."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "bda722f4",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "extracted = extract(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "d7354e76",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LDA(n_components=None, priors=None)"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lc = lda.LDA()\n",
    "lc.fit(extracted,labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "ed4d5d7e",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.080600000000000005"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = lc.predict(extract(test_data))\n",
    "1-metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2da64650",
   "metadata": {},
   "source": [
    "Confusion Matrix\n",
    "==================="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e10bfc43",
   "metadata": {},
   "source": [
    "We'd looked already at one function in the `metrics` package, but there\n",
    "are other useful functions for evaluation and getting insight in how\n",
    "classifiers work.\n",
    "\n",
    "The _confusion matrix_ tells us which kinds of classes are most often confused.\n",
    "For example, '9' and '7' are confused fairly frequently, as are '8' and '3',\n",
    "due to significant similarity.\n",
    "\n",
    "Based on confusions, we can often conclude what kind of preprocessing might\n",
    "be helpful in reducing the error rate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "cb492d3c",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 951,    0,    0,    2,    1,   12,    3,    0,    9,    2],\n",
       "       [   0, 1105,    2,    5,    1,    2,    4,    0,   16,    0],\n",
       "       [   9,    0,  912,   20,   16,    6,    8,    3,   55,    3],\n",
       "       [   2,    0,    5,  953,    1,   17,    1,    4,   16,   11],\n",
       "       [   0,    4,    8,    1,  922,    5,    3,    2,    5,   32],\n",
       "       [   5,    0,    0,   19,   16,  816,   13,    0,   18,    5],\n",
       "       [  11,    3,   10,    1,   15,   39,  870,    0,    9,    0],\n",
       "       [   2,    5,   15,    3,   16,    1,    0,  899,   10,   77],\n",
       "       [  13,    1,    7,   19,   10,   28,    3,    2,  879,   12],\n",
       "       [   5,    1,    3,   22,   40,    5,    0,   38,    8,  887]])"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.confusion_matrix(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eac5ce4c",
   "metadata": {},
   "source": [
    "Quadratic Discriminant Analysis\n",
    "================================"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32156529",
   "metadata": {},
   "source": [
    "Quadratic Discriminant Analysis (QDA) is what we were looking\n",
    "at under parametric classifiers in previous lectures."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "fce32717",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "QDA(priors=None)"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import qda\n",
    "qc = qda.QDA()\n",
    "qc.fit(data,labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "b777d56c",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.21189999999999998"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = qc.predict(test_data)\n",
    "1-metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90c5c6ad",
   "metadata": {},
   "source": [
    "Note that performance isn't very good, since this attempts to infer\n",
    "a complete covariance matrix from not very much data.\n",
    "\n",
    "Again, this is different from learning a polynomial classifier discriminatively."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "527986ed",
   "metadata": {},
   "source": [
    "Support Vector Machines\n",
    "=============="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd29db17",
   "metadata": {},
   "source": [
    "Support vector machines are a popular machine learning technique.\n",
    "You can think of them as combining ideas from perceptron learning with\n",
    "non-linear transformations of the input space based on kernels.\n",
    "As such, the method is quite versatile and can solve lots of problems.\n",
    "\n",
    "However, both training and classification with SVMs can be kind of slow, and\n",
    "other methods are often significantly faster.\n",
    "\n",
    "Algorithmically, they are a bit more complicated than that, but\n",
    "the numerical details are not all that important for using these methods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "1cbf7121",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1.0, cache_size=200, coef0=0.0, degree=3, gamma=0.00127551020408,\n",
       "  kernel='rbf', probability=False, scale_C=False, shrinking=True,\n",
       "  tol=0.001)"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import svm\n",
    "sv = svm.SVC()\n",
    "sv.fit(data,labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "d6664929",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.033599999999999963"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = sv.predict(test_data)\n",
    "1-metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ada06d25",
   "metadata": {},
   "source": [
    "Note that performance is fairly good compared to the other methods.\n",
    "There are a lot of parameters you can tune to improve them further."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9087c7dc",
   "metadata": {},
   "source": [
    "Logistic Regression\n",
    "==================="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f1706711",
   "metadata": {},
   "source": [
    "Another linear classifier that we had mentioned briefly is _logistic regression_.\n",
    "\n",
    "It computes a linear decision boundary, but uses different methods from LDA or SVM."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "id": "22168f0d",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, dual=False, fit_intercept=True, intercept_scaling=1,\n",
       "          penalty='l2', scale_C=False, tol=0.0001)"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "lr = LogisticRegression()\n",
    "lr.fit(data,labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "db5e04fd",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.04930000000000001"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = lr.predict(test_data)\n",
    "1-metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e6f6896",
   "metadata": {},
   "source": [
    "ROC Curve\n",
    "========="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d6227b3",
   "metadata": {},
   "source": [
    "As shown above, classifiers just give a single error rate.\n",
    "But often, there are different error rates we can achieve.\n",
    "For a binary classifier, there is the _true positive rate_\n",
    "(the percentage of true positives classified as such),\n",
    "and the _false positive rate_ (the percentage of negative\n",
    "samples classified as positive), and there\n",
    "is a tradeoff between the two."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a083eb2f",
   "metadata": {},
   "source": [
    "The basic idea is to use the value of a _discriminant function_ to make different tradeoffs.\n",
    "By choosing different cutoff values, we can increase the true positive rate or decreate the\n",
    "false positive rate.\n",
    "For this, we need to ask the classifier to return a _degree of confidence_ or the value of\n",
    "a discriminant function.\n",
    "There are multiple candidates for this, including `predict_log_proba`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "1d3b4795",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scores = lr.predict_log_proba(test_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d52102ad",
   "metadata": {},
   "source": [
    "The function `metrics.roc_curve` will then generate an ROC curve. \n",
    "It only does this for binary classification tasks, so here we look at\n",
    "0 vs. everything else."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "9687b189",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "fpr,tpr,thresholds = metrics.roc_curve(test_labels==0,scores[:,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "f83021e2",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x99ca150>]"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAD9CAYAAAC/fMwDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(100*fpr[:1000],100*tpr[:1000])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2abe084",
   "metadata": {},
   "source": [
    "Cross Validation\n",
    "================"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2e54c2aa",
   "metadata": {},
   "source": [
    "We have seen several times that evaluating a classifier on the training\n",
    "set yields lower error rates than evaluating it on new data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "id": "b3d161a7",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.999"
     ]
    }
   ],
   "source": [
    "from sklearn import cross_validation\n",
    "from sklearn import linear_model\n",
    "\n",
    "cl = linear_model.LogisticRegression()\n",
    "cl.fit(data[:1000],labels[:1000])\n",
    "pred = cl.predict(data[:1000])\n",
    "print metrics.zero_one_score(labels[:1000],pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "533d0873",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9108"
     ]
    }
   ],
   "source": [
    "pred = cl.predict(test_data)\n",
    "print metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d2521b7",
   "metadata": {},
   "source": [
    "If we don't have a separate test set, we can use _cross validation_ for estimating the\n",
    "true error rate of a classifier.\n",
    "\n",
    "Cross-validation splits the training data itself into a separate test and training set,\n",
    "but it does so multiple times and then reports the error rate for each split."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "5a2dbf2c",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scores = cross_validation.cross_val_score(cl,data[:1000],labels[:1000],cv=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "05f178dc",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.94  0.94  0.94  0.93  0.93  0.95  0.93  0.97  0.9   0.9 ]\n",
      "0.933"
     ]
    }
   ],
   "source": [
    "print scores\n",
    "print mean(scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f84a3992",
   "metadata": {},
   "source": [
    "Dimensionality Reduction via PCA\n",
    "=================="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "919e4aee",
   "metadata": {},
   "source": [
    "We have talked about manual preprocessing and feature extraction steps.\n",
    "We also saw one important automatic feature extraction step.\n",
    "\n",
    "Remember that PCA is a linear transformation that reduces the dimensionality of the input,\n",
    "while preserving distances as much as possible.\n",
    "It is an example of _unsupervised learning_."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "f87d8e97",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PCA(copy=True, n_components=50, whiten=False)"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import decomposition\n",
    "pca = decomposition.PCA(n_components=50)\n",
    "pca.fit(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "539330ba",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, n_neighbors=3,\n",
       "           warn_on_equidistant=True, weights='uniform')"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "knn = KNeighborsClassifier(n_neighbors=3)\n",
    "knn.fit(pca.transform(data),labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "id": "be3b97c1",
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pred = knn.predict(pca.transform(test_data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "17842107",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.01529999999999998"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1-metrics.zero_one_score(test_labels,pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fecf7f53",
   "metadata": {},
   "source": [
    "Clustering with k-Means\n",
    "======================="
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41d267a7",
   "metadata": {},
   "source": [
    "A second form of unsupervised learning we looked at is $k$-means clustering.\n",
    "\n",
    "$k$-means clustering attempts to identify representative patters in the input data.\n",
    "\n",
    "$k$-means clustering has a number of applications:\n",
    "\n",
    "- we can use it to visualize the structure of the data\n",
    "- we can use it to create \"pseudoclasses\" that can then be used with, say, a QDA (this gives a classifier somewhat similar to a kernel-based classifier)\n",
    "- we can use it as kernel centers for kernel-based methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "eddf3158",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KMeans(copy_x=True, init='k-means++', k=40, max_iter=300, n_init=10,\n",
       "    precompute_distances=True,\n",
       "    random_state=<mtrand.RandomState object at 0x7fadbbe4a498>, tol=0.0001,\n",
       "    verbose=0)"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import cluster\n",
    "km = cluster.KMeans(40)\n",
    "km.fit(data[::10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "3fcf7e63",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAD6CAYAAACBIq1uAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i,v in enumerate(km.cluster_centers_):\n",
    "    subplot(5,8,i+1)\n",
    "    imshow(v.reshape(28,28))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "4f8b56dc",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([11, 23, 24,  1,  2, 19,  1, 31, 32, 13])"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pseudoclasses = km.predict(data)\n",
    "pseudoclasses[:10]"
   ]
  }
 ],
 "metadata": {},
 "nbformat": 4,
 "nbformat_minor": 5
}