{
 "metadata": {
  "name": "",
  "signature": "sha256:b43dc356ad1ac081584f432443f25e40be6bcc73a8915d9570e7d670650e503b"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<small><i>This notebook was put together by [Jake Vanderplas](http://www.vanderplas.com) for PyCon 2014. Source and license info is on [GitHub](https://github.com/jakevdp/sklearn_pycon2014/).</i></small>"
     ]
    },
    {
     "cell_type": "heading",
     "level": 1,
     "metadata": {},
     "source": [
      "Basic principles of machine learning"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Here is where we start diving into the field of machine learning.\n",
      "\n",
      "By the end of this section you will\n",
      "\n",
      "- Know the basic categories of supervised learning, including classification and regression problems.\n",
      "- Know the basic categories of unsupervised learning, including dimensionality reduction and clustering.\n",
      "- Know the basic syntax of the Scikit-learn **estimator** interface.\n",
      "- Know why data should be split into a **training set** and **test set**\n",
      "\n",
      "In addition, we will go over several basic tools within scikit-learn which can be used to accomplish the above tasks."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%matplotlib inline\n",
      "import numpy as np\n",
      "import matplotlib.pyplot as plt"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Problem setting"
     ]
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "A simple definition of machine learning"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Machine Learning (ML) is about building programs with **tunable parameters** (typically an\n",
      "array of floating point values) that are adjusted automatically so as to improve\n",
      "their behavior by **adapting to previously seen data.**\n",
      "\n",
      "In most ML applications, the data is in a 2D array of shape ``[n_samples x n_features]``,\n",
      "where the number of features is the same for each object, and each feature column refers\n",
      "to a related piece of information about each sample.\n",
      "\n",
      "Machine learning can be broken into two broad regimes:\n",
      "*supervised learning* and *unsupervised learning*.\n",
      "We\u2019ll introduce these concepts here, and discuss them in more detail below."
     ]
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "Introducing the scikit-learn estimator object"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Every algorithm is exposed in scikit-learn via an ''Estimator'' object. For instance a linear regression is:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.linear_model import LinearRegression"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Estimator parameters**: All the parameters of an estimator can be set when it is instantiated:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "model = LinearRegression(normalize=True)\n",
      "print model.normalize"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "True\n"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print model"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "LinearRegression(copy_X=True, fit_intercept=True, normalize=True)\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Estimated parameters**: When data is fitted with an estimator, parameters are estimated from the data at hand. All the estimated parameters are attributes of the estimator object ending by an underscore:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "x = np.array([0, 1, 2])\n",
      "y = np.array([0, 1, 2])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.plot(x, y, marker='o');"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEACAYAAABI5zaHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAEYhJREFUeJzt3V9oVGfCx/HftJEtY4rRm4iTAcWka9Q4jrXkpuqEpfV1\nxCGgFwpi0LCEFLVebeHtuzSB1CIsFm1E7IUBEVTwxuwa4243HWWVMWBzIaRg4jY4iSUgRUgu1Jie\n98JudsZM5u+ZmfPn+4FAJufJmYfD6dMvT3KixzAMQwAAR3mr3BMAAJiPxR0AHIjFHQAciMUdAByI\nxR0AHIjFHQAcKO3iHo/H1dTUpHXr1mn9+vU6ffp0ynFHjx5VXV2dAoGAhoaGijJRAED2KtIdXLRo\nkb7++mtt3LhR09PTev/99/XRRx+pvr5+bkxfX59GR0c1MjKie/fuqb29XbFYrOgTBwAsLG25L1++\nXBs3bpQkVVZWqr6+Xk+ePEka09vbq5aWFklSY2Ojnj17psnJySJNFwCQjaz33MfGxjQ0NKTGxsak\nr09MTMjv98+9rqmp0fj4uHkzBADkLKvFfXp6Wnv27NGpU6dUWVk57/ibf8HA4/GYMzsAQF7S7rlL\n0szMjHbv3q39+/erubl53nGfz6d4PD73enx8XD6fb9642tpaPXr0qMDpAoDbrJZhjOb8XWnL3TAM\ntba2au3atTp27FjKMZFIRBcuXJAkxWIxVVVVqbq6et64R48eyTAMPkz6+OKLL8o+B6d8cC25nlb5\n+Oc/Da1caWjFis8lGb995BfFacv9zp07unjxojZs2KBgMChJOn78uB4/fixJamtrUzgcVl9fn2pr\na7V48WL19PTkNREAcKvpaelPf5L++lfp3DnJMD7Wp59+rkePvsz7nGkX9w8//FC//vprxpN0d3fn\nPQEAcLOBAam1VWpqkh48kKqqJGmrJOmbb/6smzfzO2/GPXdYUygUKvcUHINraS6uZ3berPVwOPn4\nzp1btXPnVnk8XXmd32MYRkn+sQ6Px6MSvRUAWFpirZ88+Z9aTy3ftZNyB4ASyVTrZuIPhwFACQwM\nSA0N0vPnr/fWi7mwS5Q7ABRVKWs9EeUOAEVS6lpPRLkDgMnKVeuJKHcAMFE5az0R5Q4AJrBCrSei\n3AGgQFap9USUOwDkyWq1nohyB4A8WLHWE1HuAJADK9d6IsodALJk9VpPRLkDQAZ2qfVElDsApGGn\nWk9EuQNACnas9USUOwC8wa61nohyB4Df2L3WE1HuACBn1Hoiyh2Aqzmp1hNR7gBcy2m1nohyB+A6\nTq31RJQ7AFdxcq0notwBuIIbaj0R5Q7A8dxS64kodwCO5bZaT0S5A3AkN9Z6IsodgKO4udYTUe4A\nHMPttZ6Icgdge9T6fJQ7AFuj1lOj3AHYErWeHuUOwHao9cwodwC2Qa1nj3IHYAvUem4odwCWRq3n\nh3IHYFnUev4odwCWQ60XjnIHYCnUujkodwCWQK2bi3IHUHbUuvkodwBlQ60XD+UOoCyo9eKi3AGU\nFLVeGpQ7gJKh1kuHcgdQdNR66WUs90OHDqm6uloNDQ0pj0ejUS1ZskTBYFDBYFBdXV2mTxKAfVHr\n5ZGx3A8ePKgjR47owIEDC47Ztm2bent7TZ0YAHuj1ssrY7lv2bJFS5cuTTvGMAzTJgTA/qj18iv4\nB6oej0d3795VIBBQOBzW8PCwGfMCYEPT09Inn0gtLdKZM9L581JVVbln5U4F/0B106ZNisfj8nq9\nunHjhpqbm/Xw4cOUYzs6OuY+D4VCCoVChb49AIsYGJBaW6Wmpte1zqKen2g0qmg0WvB5PEYWeypj\nY2PatWuXHjx4kPGEq1at0v3797Vs2bLkN/J42L4BHIi99eLKd+0seFtmcnJy7o0HBwdlGMa8hR2A\nM7G3bl0Zt2X27dunW7du6enTp/L7/ers7NTMzIwkqa2tTVevXtXZs2dVUVEhr9ery5cvF33SAMqL\nWre+rLZlTHkjtmUAR0jcWz95kr31Yst37eQJVQBZodbthb8tAyAj9tbth3IHsCBq3b4odwApUev2\nRrkDSEKtOwPlDmAOte4clDsAat2BKHfA5ah1Z6LcAZei1p2NcgdciFp3PsodcBFq3T0od8AlqHV3\nodwBh6PW3YlyBxyMWncvyh1wIGodlDvgMNQ6JModcAxqHYkod8ABqHW8iXIHbIxax0Iod8CmqHWk\nQ7kDNkOtIxuUO2Aj1DqyRbkDNkCtI1eUO2Bx1DryQbkDFkWtoxCUO2BB1DoKRbkDFkKtwyyUO2AR\n1DrMRLkDZUatoxgod6CMqHUUC+UOlAG1jmKj3IESo9ZRCpQ7UCLUOkqJcgdKgFpHqVHuQBFR6ygX\nyh0oEmod5US5Ayaj1mEFlDtgImodVkG5Ayag1mE1lDtQIGodVkS5A3mi1mFllDuQB2odVke5Azmg\n1mEXlDuQJWoddkK5AxlQ67Ajyh1Ig1qHXVHuQArUOuwuY7kfOnRI1dXVamhoWHDM0aNHVVdXp0Ag\noKGhIVMnCJQatQ4nyFjuBw8e1JEjR3TgwIGUx/v6+jQ6OqqRkRHdu3dP7e3tisVipk8UMNv167d1\n+vTf9eJFhX73u1f64x8/1sDAVmodjpBxcd+yZYvGxsYWPN7b26uWlhZJUmNjo549e6bJyUlVV1eb\nNknAbNev39ann97Uo0dfzn1tYOBzhULSgwdbVVVVvrkBZij4B6oTExPy+/1zr2tqajQ+Pl7oaYGi\nOn3670kLuyS9evWl3n77HyzscARTfqBqGEbSa4/Hk3JcR0fH3OehUEihUMiMtwdy9uJF6lv/+fO3\nSzwTIFk0GlU0Gi34PAUv7j6fT/F4fO71+Pi4fD5fyrGJiztQLtPT0r///SrlsXfemS3xbIBkb4Zv\nZ2dnXucpeFsmEonowoULkqRYLKaqqir222FZ//lNmN///mOtWvV50rHVq/9XR458VKaZAebKWO77\n9u3TrVu39PTpU/n9fnV2dmpmZkaS1NbWpnA4rL6+PtXW1mrx4sXq6ekp+qSBXM3/vfWtun5d+uab\nP+v587f1zjuzOnLkf7Rz59ZyTxUwhcd4c8O8WG/k8czbmwdKYWBAam2VmpqkkyfFD0xhK/munTyh\nCsfiKVO4GX9bBo7EU6ZwO8odjkKtA69R7nAMah34L8odtketA/NR7rA1ah1IjXKHLVHrQHqUO2yH\nWgcyo9xhG9Q6kD3KHbZArQO5odxhadQ6kB/KHZZFrQP5o9xhOdQ6UDjKHZZCrQPmoNxhCdQ6YC7K\nHWVHrQPmo9xRNtQ6UDyUO8qCWgeKi3JHSVHrQGlQ7igZah0oHcodRUetA6VHuaOoqHWgPCh3FAW1\nDpQX5Q7TUetA+VHuMA21DlgH5Q5TUOuAtVDuKAi1DlgT5Y68UeuAdVHuyBm1Dlgf5Y6cUOuAPVDu\nyAq1DtgL5Y6MqHXAfih3LIhaB+yLckdK1Dpgb5Q7klDrgDNQ7phDrQPOQbmDWgcciHJ3OWodcCbK\n3aWodcDZKHcXotYB56PcXYRaB9yDcncJah1wF8rd4ah1wJ0odwej1gH3otwdiFoHQLk7DLUOQKLc\nHYNaB5AoY7n39/drzZo1qqur04kTJ+Ydj0ajWrJkiYLBoILBoLq6uooyUSyMWgfwprTlPjs7q8OH\nD+u7776Tz+fTBx98oEgkovr6+qRx27ZtU29vb1EnivmodQALSVvug4ODqq2t1cqVK7Vo0SLt3btX\n165dmzfOMIyiTRCpUesA0km7uE9MTMjv98+9rqmp0cTERNIYj8eju3fvKhAIKBwOa3h4uDgzhSRp\nakr65BOppUU6c0Y6f16qqir3rABYTdptGY/Hk/EEmzZtUjwel9fr1Y0bN9Tc3KyHDx+mHNvR0TH3\neSgUUigUymmybjcwILW2Sk1Nr2udRR1wnmg0qmg0WvB5PEaaPZVYLKaOjg719/dLkr766iu99dZb\n+uyzzxY84apVq3T//n0tW7Ys+Y08HrZv8jQ19Xpv/W9/Y28dcJt818602zKbN2/WyMiIxsbG9PLl\nS125ckWRSCRpzOTk5NwbDw4OyjCMeQs78jcwIG3YIL14wd46gOyl3ZapqKhQd3e3tm/frtnZWbW2\ntqq+vl7nzp2TJLW1tenq1as6e/asKioq5PV6dfny5ZJM3OmodQCFSLstY+obsS2TtcS99ZMn2VsH\n3CzftZMnVC2EWgdgFv62jEWwtw7ATJR7mVHrAIqBci8jah1AsVDuZUCtAyg2yr3EqHUApUC5lwi1\nDqCUKPcSoNYBlBrlXkTUOoByodyLhFoHUE6Uu8modQBWQLmbiFoHYBWUuwmodQBWQ7kXiFoHYEWU\ne56odQBWRrnngVoHYHWUew6odQB2QblniVoHYCeUewbUOgA7otzToNYB2BXlngK1DsDuKPc3UOsA\nnIBy/w21DsBJKHdR6wCcx9XlTq0DcCrXlju1DsDJXFfu1DoAN3BVuVPrANzCFeVOrQNwG8eXO7UO\nwI0cW+7UOgA3c2S5U+sA3M5R5U6tA8Brjil3ah0A/sv25U6tA8B8ti53ah0AUrNluVPrAJCe7cqd\nWgeAzGxT7tQ6AGTPFuVOrQNAbixd7tQ6AOTHsuVOrQNA/ixX7tQ6ABTOUuVOrQOAOSxR7tQ6AJir\n7OVOrQOA+TIu7v39/VqzZo3q6up04sSJlGOOHj2quro6BQIBDQ0NZfXGU1NSe7vU0iKdOSOdPy9V\nVeU2eQBAamkX99nZWR0+fFj9/f0aHh7WpUuX9OOPPyaN6evr0+joqEZGRvTtt9+qvb0945tS64WL\nRqPlnoJjcC3NxfW0hrSL++DgoGpra7Vy5UotWrRIe/fu1bVr15LG9Pb2qqWlRZLU2NioZ8+eaXJy\nMuX5/vCH/1M4fJtaNwH/AZmHa2kurqc1pF3cJyYm5Pf7517X1NRoYmIi45jx8fGU5xsY6NK//nVT\nf/nLbWodAIoo7eLu8XiyOolhGFl/39TUl+rp+UdW5wUA5Cftr0L6fD7F4/G51/F4XDU1NWnHjI+P\ny+fzpTjbakmvF/2bNyWPpyv/WUOS1NnZWe4pOAbX0lxcT/OsXr06r+9Lu7hv3rxZIyMjGhsb04oV\nK3TlyhVdunQpaUwkElF3d7f27t2rWCymqqoqVVdXzzuXYYzmNUEAQO7SLu4VFRXq7u7W9u3bNTs7\nq9bWVtXX1+vcuXOSpLa2NoXDYfX19am2tlaLFy9WT09PSSYOAFiYx3hzwxwAYHumP6FarIee3CjT\ntYxGo1qyZImCwaCCwaC6uvg5xkIOHTqk6upqNTQ0LDiG+zJ7ma4n92b24vG4mpqatG7dOq1fv16n\nT59OOS7n+9Mw0atXr4zVq1cbP/30k/Hy5UsjEAgYw8PDSWOuX79u7NixwzAMw4jFYkZjY6OZU3CM\nbK7l999/b+zatatMM7SX27dvGz/88IOxfv36lMe5L3OT6Xpyb2bv559/NoaGhgzDMIypqSnjvffe\nM2XdNLXczX7oyc2yuZbS/F9DRWpbtmzR0qVLFzzOfZmbTNdT4t7M1vLly7Vx40ZJUmVlperr6/Xk\nyZOkMfncn6Yu7mY/9ORm2VxLj8eju3fvKhAIKBwOa3h4uNTTdAzuS3Nxb+ZnbGxMQ0NDamxsTPp6\nPvenqX/ytxgPPblVNtdk06ZNisfj8nq9unHjhpqbm/Xw4cMSzM6ZuC/Nw72Zu+npae3Zs0enTp1S\nZWXlvOO53p+mlru5Dz25WzbX8t1335XX65Uk7dixQzMzM/rll19KOk+n4L40F/dmbmZmZrR7927t\n379fzc3N847nc3+aurgnPvT08uVLXblyRZFIJGlMJBLRhQsXJCntQ09ul821nJycnPu/+eDgoAzD\n0LJly8oxXdvjvjQX92b2DMNQa2ur1q5dq2PHjqUck8/9aeq2DA89mSeba3n16lWdPXtWFRUV8nq9\nunz5cplnbV379u3TrVu39PTpU/n9fnV2dmpmZkYS92U+Ml1P7s3s3blzRxcvXtSGDRsUDAYlSceP\nH9fjx48l5X9/8hATADhQ2f+ZPQCA+VjcAcCBWNwBwIFY3AHAgVjcAcCBWNwBwIFY3AHAgVjcAcCB\n/h8qEpStW5V4jAAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107d6a190>"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X = x[:, np.newaxis] # The input data for sklearn is 2D: (samples == 3 x features == 1)\n",
      "X"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 7,
       "text": [
        "array([[0],\n",
        "       [1],\n",
        "       [2]])"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "model.fit(X, y) \n",
      "model.coef_"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 8,
       "text": [
        "array([ 1.])"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Supervised Learning: Classification and regression"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "In **Supervised Learning**, we have a dataset consisting of both features and labels.\n",
      "The task is to construct an estimator which is able to predict the label of an object\n",
      "given the set of features. A relatively simple example is predicting the species of \n",
      "iris given a set of measurements of its flower. This is a relatively simple task. \n",
      "Some more complicated examples are:\n",
      "\n",
      "- given a multicolor image of an object through a telescope, determine\n",
      "  whether that object is a star, a quasar, or a galaxy.\n",
      "- given a photograph of a person, identify the person in the photo.\n",
      "- given a list of movies a person has watched and their personal rating\n",
      "  of the movie, recommend a list of movies they would like\n",
      "  (So-called *recommender systems*: a famous example is the [Netflix Prize](http://en.wikipedia.org/wiki/Netflix_prize)).\n",
      "\n",
      "What these tasks have in common is that there is one or more unknown\n",
      "quantities associated with the object which needs to be determined from other\n",
      "observed quantities.\n",
      "\n",
      "Supervised learning is further broken down into two categories, **classification** and **regression**.\n",
      "In classification, the label is discrete, while in regression, the label is continuous. For example,\n",
      "in astronomy, the task of determining whether an object is a star, a galaxy, or a quasar is a\n",
      "classification problem: the label is from three distinct categories. On the other hand, we might\n",
      "wish to estimate the age of an object based on such observations: this would be a regression problem,\n",
      "because the label (age) is a continuous quantity."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Classification\n",
      "K nearest neighbors (kNN) is one of the simplest learning strategies: given a new, unknown observation, look up in your reference database which ones have the closest features and assign the predominant class.\n",
      "\n",
      "Let's try it out on our iris classification problem:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn import neighbors, datasets\n",
      "iris = datasets.load_iris()\n",
      "X, y = iris.data, iris.target\n",
      "knn = neighbors.KNeighborsClassifier(n_neighbors=1)\n",
      "knn.fit(X, y)\n",
      "# What kind of iris has 3cm x 5cm sepal and 4cm x 2cm petal?\n",
      "print iris.target_names[knn.predict([[3, 5, 4, 2]])]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "['virginica']\n"
       ]
      }
     ],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# A plot of the sepal space and the prediction of the KNN\n",
      "from fig_code import plot_iris_classification\n",
      "plot_iris_classification(neighbors.KNeighborsClassifier, n_neighbors=3)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEKCAYAAAD6q1UVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nADAHYIPqAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107d91ed0>"
       ]
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "#### Quick Exercise\n",
      "Now use as an estimator on the same problem: ``sklearn.svm.SVC``.\n",
      "\n",
      "(Note that you don't have to know what it is to use it. If you finish early, do the same plot as above)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.svm import SVC\n",
      "unknown_iris = [[3, 5, 4, 2]]\n",
      "\n",
      "# Use SVC to train and predict the species of the unknown iris...\n",
      "\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 11
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Regression\n",
      "The simplest possible regression setting is the linear regression one:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Create some simple data\n",
      "np.random.seed(0)\n",
      "X = np.random.random(size=(20, 1))\n",
      "y = 3 * X.squeeze() + 2 + np.random.normal(size=20)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 12
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Fit a linear regression to it\n",
      "from sklearn.linear_model import LinearRegression\n",
      "model = LinearRegression(fit_intercept=True)\n",
      "model.fit(X, y)\n",
      "print \"Model coefficient: %.5f, and intercept: %.5f\" % (model.coef_, model.intercept_)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Model coefficient: 3.93491, and intercept: 1.46229\n"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Plot the data and the model prediction\n",
      "X_test = np.linspace(0, 1, 100)[:, np.newaxis]\n",
      "y_test = model.predict(X_test)\n",
      "import pylab as pl\n",
      "plt.plot(X.squeeze(), y, 'o')\n",
      "plt.plot(X_test.squeeze(), y_test);"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAW0AAAEACAYAAAB4ayemAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAHylJREFUeJzt3XtQU+e+N/AvAoJXFG8ooGhQSbzgtdYrsYqAWm9J3te2\nY63u2k5ntrSd9nT2rrsjtrtqu3vaCp6zp7NP657d17ZnXHireEHdjW7vrVr1mAgaoeIFb1UUkEvC\nev/gGIViWCRZyVrJ9zOTGUzCWj+ekS8rz1rr+YWIoiiCiIhUoZW/CyAiIukY2kREKsLQJiJSEYY2\nEZGKMLSJiFSEoU1EpCLNhvaqVaswaNAgDBkyBM8//zyqq6t9URcRETXBZWgXFxfjb3/7G06cOIEz\nZ87A4XDgu+++81VtRETUSJirFzt27Ijw8HBUVlYiNDQUlZWViI2N9VVtRETUiMsj7ejoaLz11lvo\n3bs3evXqhU6dOmHq1Km+qo2IiBpxGdo2mw2ff/45iouLcfXqVZSXl2P9+vW+qo2IiBpxOT3y008/\nYdy4cejSpQsAYN68eTh06BBeeOEF53sSExNhs9nkrZKIKMBoNBpcuHChxd/n8kg7KSkJR44cwYMH\nDyCKIvbs2QOdTtfgPTabDaIo8iGKWL58ud9rUMqDY8Gx4Fi4frh7sOsytJOTk/Hiiy9i1KhRGDp0\nKADglVdecWtHRETkOZfTIwDwzjvv4J133vFFLURE1AzeEelFer3e3yUoBsfiEY7FIxwLz4WIouhR\nE4SQkBB4uAkioqDjbnbySJuISEUY2kREKsLQJiJSEYY2EZGKMLSJiFSEoU1EpCIMbSIiFWFoExGp\nCEObiEhFGNpERCrC0CYiUhGGNhGRijC0iYhUhKFNRKQizTZBICKiJ8vL24/s7HxUV4chIsKOzMxp\nmDFjkmz7Y2gTEbkpL28/Xn99F2y2D53P2WzLAEC24Ob0CBGRm7Kz8xsENgDYbB8iJ2e3bPtsNrQL\nCgowfPhw5yMqKgrZ2dmyFUREpBbV1U1PVlRVhcq2z2anRwYOHIiTJ08CAOrq6hAbG4u5c+fKVhAR\nkVpERNibfD4y0iHbPls0PbJnzx5oNBrEx8fLVQ8RkWpkZk6DRrOswXMazbtYujRVtn226ETkd999\nh+eff16uWoiIVOXhycacnPdQVRWKyEgHli5Nl/XqEcnd2GtqahAbGwuLxYJu3bo92gC7sRMRtZi7\n2Sn5SHvHjh0YOXJkg8B+KCsry/m1Xq+HXq9vcSFERIHMbDbDbDZ7vB3JR9rz589HRkYGFi5c2HAD\nPNImImoxd7NTUmhXVFSgT58+KCoqQocOHbyyYyKiYCZraMuxYyKiYOZudvKOSCIiFWFoExGpCEOb\niEhFGNpERCrC0CYiUhGGNhGRijC0iYhUhKFNRKQiDG0iIhVhaBMRqQhDm4hIRRjaREQqwtAmIlIR\nhjYRkYq0qEckEVEgy8vbj+zsfFRXhyEiwo7MzGmy9nt0B0ObiAj1gf3667tgs33ofM5mq++0rqTg\n5vQIERGA7Oz8BoENADbbh8jJ2e2niprG0CYiAlBd3fTEQ1VVqI8rcY2hTUQEICLC3uTzkZEOH1fi\nWrOhfffuXRiNRmi1Wuh0Ohw5csQXdRER+VRm5jRoNMsaPKfRvIulS1P9VFHTmm3su3DhQqSkpGDx\n4sWw2+2oqKhAVFTUow2wsS8RBYi8vP3IydmNqqpQREY6sHRpqmwnIWXpxl5WVobhw4fj4sWLXt8x\nEVEwk6Ube1FREbp164ZFixZhxIgRWLJkCSorK90ukoiIPOPyOm273Y4TJ05g7dq1GD16NN544w2s\nXr0a77//foP3ZWVlOb/W6/XQ6/Vy1EpEpFpmsxlms9nj7bicHiktLcXYsWNRVFQEADhw4ABWr16N\nbdu2PdoAp0eIiFpMlumRmJgYxMfHo7CwEACwZ88eDBo0yL0KiYjIY81ePXLq1Cm8/PLLqKmpgUaj\nwbp163j1CBGRh2S5ekTOHRMRBTNZpkeIiEhZGNpERCrC0CYiUhGGNhGRijC0iYhUhKFNRKQiDG0i\nIhVhaBMRqQhDm4hIRRjaREQqwtAmIlIRhjYRkY/9+uBXt7/XZRMEIiLyjpsVN7H53GYIVgFHLrvf\nIJ2r/BERyaS0vBSbrJsgWAX8dPUnpGnSYNKZkNE/Ax0iOnBpViJfyMvbj+zsfFRXhyEiwo7MzGmy\ndewm9bly7wo2WjdCsAo4ff00pvefDqPWiLTENLQNb+t8n7vZyekRohbIy9uP11/fBZvtQ+dzNtsy\nAGBwB7FLZZeQa8mFYBVgvWnFswOfxdtj30aqJhWRYZFe3RePtIma8KSj6bS0PyE//8+/eX9a2nvY\nufMDP1RK/nLxzkUIFgG51lzYfrVh9sDZMOqMmNJvClqHtm72+3mkTeQlro6mq6ub/pWpqgr1SW3k\nX4W3C5FrycUGywZcuX8Fc5Pm4s+T/wx9gh7hoeE+qYGhTdRIdnZ+g8AGAJvtQ+TkvIeIiKaPjCIj\nHb4ojfzActPinPq4UXEDBq0Bn6Z9iom9JyK0le//WEsK7YSEBHTs2BGhoaEIDw/HsWPH5K6LyG9c\nHU3/2789A5ttWYNQ12jexdKl6b4qj2QmiiLO3DgDwSJAsAi4X3MfBq0BazPWYnzv8WgV4t/bWySF\ndkhICMxmM6Kjo+Wuh8jvIiLsTT4fGelwnmzMyXkPVVWhiIx0YOnSdJ6EVDlRFHHi2gnkWnMhWATU\n1tXCqDXiq9lf4anYp/we1I+TPD3Ck40ULDIzp7k8mp4xYxJDOgCIoohjV47VH1FbBYS1CoNBa8C3\nhm8xoucIhISE+LvEJkm6eqRfv36IiopCaGgoXn31VSxZsuTRBnj1CAWgvLz9yMnZ/djRdCqDOgDU\niXU4cvkINpzdgFxrLtq1bgeD1gCTzoShPYb6NKhlvXrk4MGD6NmzJ27evInU1FQkJSVh4sSJztez\nsrKcX+v1euj1+hYXQqQkPJoOHI46Bw5cOgDBImDjuY2IbhMNg9aAHS/swKDug3xWh9lshtls9ng7\nLb5Oe8WKFWjfvj3eeuut+g3wSJuIFMZeZ8e+4n0QLAI2nduEmPYxMOlMMOgMSOqa5O/yAMh4pF1Z\nWQmHw4EOHTqgoqIC+fn5WL58uVtFEhHJpdZRi38W/ROCRcDmgs3oE9UHJp0JBxYfQGJ0or/L85pm\nQ/v69euYO3cuAMBut+OFF17AtGnTZC+MiKg51fZq7Lm4B4JVwNaCrRjQZQCMWiN+nPQjEjol+Ls8\nWfA2diJSlQe1D7DLtgu51lxsK9yGwd0HQycOx7lNQMj9aNUs4sXb2IkoYFXUVGDHhR0QLAJ2XtiJ\nET1HwKgz4uOpH+PE/vNBtYgXj7SJSJHuV99H3vk8CBYBuy/uxpjYMTDqjJiTNAfd23V3vk+ti3jx\nSJuIVK+sqgzfF34PwSLgn0X/xITeE2DQGvDFzC/QpW2XJr8n2BbxYmgTkV/deXAHWwq2QLAI2P/L\nfqQkpMCkM2Hd7HXo3KZzs9/vatmBQMTQJlKoQO6Qc6vyVn2/RIuAQyWHMKXfFDw/5Hl8Y/gGHSM6\ntmhbzS07EGgY2kQKFIgdcq6XX8emc5sgWAT8ePVHTNNMw+LhiyH8HwHtW7d3e7vBtogXT0QSKZBa\nT641dvX+1fp+iRYBp66fQkZiBow6I9IT0xv0SwxGPBFJFEDUfHKtcb/EmQNm4q2xb8nSLzEYMbSJ\nFEhtJ9eK7hQ516K+8OsFzB44G3+a+CdUn2uNv679Af9efRxrI44G1Ly8vzC0iRRIDSfXHvZLzLXm\n4lLZJcxNmosPJn/g7JeYl7cfb78ZWPPySsA5bSKFUuKa3o37Jc5LmgeDzoBJfSYhrFXDY8BAmZeX\nC+e0iQKMEtb0FkUR/3PjfyBYBGywbMC96nswaA3IycjB+PjxLhvbqnleXskY2kTUgCiKOFl60nlE\nXW2vhlFnxJezvsSYuDGS+yWqbV5eLRjaRARRFPHj1R+dHchbhbSCQWvA+nnrMbLnSLfacKlhXl6N\nOKdNFKTqxDocvXzU2dg2MiwSJp0JRp0RyT2SvdIvUYnz8krhbnYytMnvAvl2baVx1DlwsORgfb9E\n60Z0iuxU39h2kAmDug1SbAfyQMQTkaRKgXi7ttLY6+zY/8t+Z1DHtI+BUWfE7gW7oe2m9Xd51EI8\n0ia/CvTLwvz1KaLWUYsfin+o75d4bjP6dOoDg9YAg9aA/l36y75/ah6PtEmVAvmyMF9/injYLzHX\nmoutBVvRv0t/GLVGHH35KPp27uv1/ZF/SApth8OBUaNGIS4uDt9//73cNVEQCeTLwrKz8xsENgDY\nbB8iJ+c9r4X2g9oHyLflQ7AKzn6JRq0RK/QrEB8V75V9kLJICu01a9ZAp9Ph/v37ctdDQSaQLwuT\n61NEZW0ldpzfAcEqYMf5HRgWMwwmnQkfTf0IvTr08mjbpHzNhvbly5exfft2LFu2DJ9++qkvaqIg\nEshrIXvzU0R5TTnyCvMgWAXk2/IxutdomHQmfJ72OXq07+FpqV7Bq4B8o9nQfvPNN/GXv/wF9+7d\n80U9FISUcLu2HDz9FFFWVYZthdsgWAXsvbgX43uPh0FrwF9n/BVd23b1qDZvB6ySrwIKtD8mLkN7\n27Zt6N69O4YPHw6z2fzE92VlZTm/1uv10Ov1XiqPSL3c+RRx58EdbC3YCsEqYF/xPugT9DBoDfhq\n1leS+iVKIUfA+mL+3h1K+mNiNptd5qhkogt//OMfxbi4ODEhIUGMiYkR27ZtKy5YsKDBe5rZBBE1\n42bFTfG/jv+XmP7/0sWOqzqKc76bI64/vV4sqyqTZX/Tpi0TAfE3j7S0P7m9zZSU5U1uMyVlufcK\nd4McP6u3uJudLo+0V65ciZUrVwIA9u3bh08++QT/+Mc/PP9LQRTkrpdfr29saxVw7MoxpGnSsGjY\nImwwbfCoX6IUcpwgVepVQIF4SWmLrtPmLa5E7mvcL3F6/+l4bdRr2DJ/i0/7JcoRsEq9Ckipf0w8\nwTsiiWRUUlbibMNluWnBzAEzYdKZ/Novsal5Xo3mXaxZ49lVO0pcHEqun9UbuGAUkUI07pc4a+As\nGHVGTOk7BRFhEf4uD4AyA1YuSv1ZGdpEfnT+9nlnUF8qu4Q5SXNg1BkxOWEywkPD/V0eKRBDm8jH\nrDetzrWor5dfr1+Q6Qn9EokaY2gTyUx8rF+iYBVQVlWGedp5MGgNmNB7gst+iUSNMbSJZCCKIn4u\n/dkZ1FX2Khi1Rhh0Bjwd97TkfolEjTG0ibxEFEX8dPUnZ1ADgFFrhFFnxKheo3jpK3kFQ5vIA031\nSzRqjTANMnmtXyLR4xjaRC3kqHPgUMkhCBYBudZcREVGORvbsl8iyY2hTSSBvc6Of/3yr/p+iec2\nonu77jDpTDBoDeyXSD7F0CZ6glpHLczFZggWAZvObUJ8VLwzqAuPXAuoZTtJPdgjkugx1fZq7C3a\nC8EiYGvBViRGJ8Koa9gvUUnLdhJJxSNtChhV9irsurALudZcbCvcBl03HYw6I+Zp56F3VO/fvD/Q\nO8GTsvFIm4JSZW0ldl7YCcEiYPv57RgWMwxGnRGrpqxCbMdYl98biMt2UuBjaJPqNNUv0aA14NO0\nTxHTPkbydgJx2U4KfAxtUoWH/RJzrbnYW7QX4+LHwaA14D+n/ye6tevm1jaVugY0kSuc0ybFatwv\nMSUhBQatAbMHzvZqv0QlLttJgY+X/FFAuFV5C1vObYFgFXDw0kE80/cZmHQmzBwwE1GRUf4uj8hr\nGNqkWjcqbmCTdZOzX2Jqv1SYdCbMGDBD9n6JRP7C0CZVuXb/Wn2/RKuAk9dOIqN/Bkw6E9IT033a\nL5HIX2QL7aqqKqSkpKC6uho1NTWYPXs2Vq1a5fGOKfiUlJU4g/rsjbOYOWAmjDojpmmm+a1fIpG/\nyHqkXVlZibZt28Jut2PChAn45JNPMGHCBI92TMHhYb/EXGsuCm8XYvbA2Yrrl0jkD7LeXNO2bf3H\n1ZqaGjgcDkRHR7d4RxQ8Hu+X+EvZL5ibNBcr9CvYL5HICySFdl1dHUaMGAGbzYbXXnsNOp1O7rpI\nZc7dOle/FrVFQGl5KeZp5+GjqR8hJSGF/RKJvEjSb1OrVq3w888/o6ysDGlpaTCbzdDr9c7Xs7Ky\nnF/r9foGr1FgEkURZ2+edQb1nao7MGgNWJO+hv0SiZpgNpthNps93k6Lrx754IMP0KZNG7z99tv1\nG+CcdtAQRRGnrp9yBvUD+wP2SyRyk2xz2rdu3UJYWBg6deqEBw8eYPfu3Vi+fLlbRZL6iKKI49eO\nO4NahAiD1oCv537NfolEftBsaF+7dg0LFy5EXV0d6urqsGDBAkyZMsUXtZGfPOyX+PBkYuvQ1jDq\njNhg2oBhMcMCPqjz8vazMQIpFm+uIQD1Qf14v8SOER1h0Bpg0pkwuPvggA/qh5pqjKDRLMOaNWkM\nbvIq3hFJLeaoc+Bfl/6FDWc3YOO5jejWtlt9Gy6dAbpuwXmFEBsjkK+wCQJJ0rhfYlzHOJh0Jux7\naR8GdBng7/L8jo0RSOlUF9qcb2y5GkcN9lzcg1xLLrYUbIEmWgOj1ogjLx9Bv879/F2eorAxAimd\nqkKbjVilq7JXId+WD8EiNOiXuFy/vMl+iVSPjRFI6VQ1p835Rtee1C9xbtLcZvsl0iNsjEC+EBRz\n2pxv/K3ymnJsP78dgkXALtsujO41GkadscX9EumRGTMmMaRJsVQV2pxvrHev+h62FW6DYBEa9Ev8\nj+n/4Xa/RCKl4Hkr11QV2sE833i36m59v0SLAHOx2dkv8ctZX3qtXyKRv/G8VfNUNacNBNd84+3K\n29hSsAWCRcDBkoOYnDCZ/RIpoAXTeaugmNMGAn++8UbFDWw+txmCRcDRK0eR2i8VC5MX4r+N/40O\nER38XR6RrHjeqnmKDe1gmte6dv8aNp3bBMEi4MS1E0hPTMerI1/Fpv+7Ce1at/N3eUQ+w/NWzVNk\naAfDvNble5fr+yVaBJy5cQYzB8zE62NexzTNNLQJb+Pv8oj8IpjPW0mlyDntQJ3XKr5bjFxLLgSr\ngMLbhZg1cBaMWiOm9pvKfolE/ytYzlsF1Jx2IM1rXfj1gjOoi+8WY87AOchKycLkvpPROrS1v8sj\nUpxAP2/lKUWGttrntc7dOucM6qv3r2Je0jysnrKa/RKJyGOKTBC1zWs11S9xXtI8fJb2GSb2nsh+\niUTkNYqc0waUP6/VuF9iZW0ljDojDFoDxsaPZb9EInKJTRB8oHG/xDqxrr67yyATRvcaHTTdXYjI\ncwxtmdSJdTh25ZgzqFuHtnYG9fCY4QxqInKLbKFdUlKCF198ETdu3EBISAheeeUVZGZmerxjJWvc\nL7FD6w7OoB7SfQiDmog8Jltol5aWorS0FMOGDUN5eTlGjhyJzZs3Q6vVerRjpXnYL1GwCNho3Yiu\nbbsGfb9EIpKPbNdpx8TEICamfl3m9u3bQ6vV4urVq87QVjN7nb1Bv8TYDrEw6Uwwv2Rmv0QiUqQW\nXfJXXFyMkydPYsyYMXLVI7saRw32XtyLXGt9v8S+nfrCpDPh8O8Os18iESme5NAuLy+H0WjEmjVr\n0L59+wavZWVlOb/W6/XQ6/Xeqs8rquxV2G3bDcEq4PuC76HtpoVRa8R7k95Dn059/F0eEQUBs9kM\ns9ns8XYkXT1SW1uLmTNnIiMjA2+88UbDDSh0TvtB7YP6folWAXmFeUiOSYZRa8Rc7VzEdYzzd3lE\nFORkOxEpiiIWLlyILl264LPPPvPajuVQUVNR3y/RKmDXhV0Y2WskjFoj5iTNQc8OPf1dHhGRk2yh\nfeDAAUyaNAlDhw51Xuq2atUqpKene7Rjb7lXfQ95hXkQrAL2XNyDsXFjYdAaMCdpDvslEpFiBdXN\nNXer7uL7gu8hWAX8UPQDJvWZBIPWgNlJsxHdJtqntRARuSPgQ/t25e36xrZWAQcuHcDkhMkwaA14\nduCz6BTZSfb9028FU3chIm8LqPW0H7pZcRObz23GBssGHL1yFFP7TcWCoQvwneE79kv0s2DoLkSk\nRIo70i4tL8Um6yYIVgHHrx5HWmIaTDoTMhIz2C9RQQK1uxCRr6j6SLtxv8QZ/Wdg6VNLkaZJY79E\nhQqk7kJEauK30H68X2LBrQLMGjgL74x/B6n9UtkvUQXU3l2ISK18GtqN+yXOHjgby1OW45m+z7Bf\nosqorbsQUaCQfU674FZB/VrUj/VLNOgMSOmTgvDQcE92TX6m9O5CREqmmEv+RFGE5aYFudZcbLBs\nwO3K2zBoDTDoDOyXSET0v/wa2nV1dTh9/bTziLq8phxGrREGnQHj4sexXyIRUSN+De3E7ETY6+z1\nTQO0BoyOHc2gJiJywa+hffzqcfZLJCJqAcXMaRMRUfPczU7OYRARqQhDm4hIRRjaREQqIssdkVyy\nk4hIHl4PbS7ZSUQkH69Pj2Rn5zcIbACw2T5ETs5ub++KiCjoNBvaixcvRo8ePTBkyBBJG+SSnURE\n8mk2tBctWoSdO3dK3iCX7CQikk+zoT1x4kR07txZ8gYzM6dBo1nW4Ln6JTtTW14dERE14PUTkQ9P\nNubkvPfYkp3pPAlJROQFslzyN2PGJIY0EZEMvBLaWVlZzq/1ej30er03NktEFDDMZjPMZrPH25G0\nYFRxcTGeffZZnDlz5rcb4IJRREQtJtuCUc899xzGjRuHwsJCxMfHY926dW4VSEREnuPSrEREfsCl\nWYmIgoAsV48QPQkXEyPyDEObfIaLiRF5jtMj5DNcTIzIcwxt8hkuJkbkOYY2+QwXEyPyHEObfIaL\niRF5jtdpk0/l5e1HTs7uxxYTS+VJSApK7mYnQ5uIyA94cw0RURBgaBMRqQhDm4hIRRjaREQqwtAm\nIlIRhjYRkYowtImIVIShTUSkIgxtIiIVYWgTEalIs6G9c+dOJCUloX///vjoo498URMRET2By9B2\nOBz4/e9/j507d8JiseDbb7+F1Wr1VW2qYzab/V2CYnAsHuFYPMKx8JzL0D527BgSExORkJCA8PBw\nzJ8/H1u2bPFVbarD/5CPcCwe4Vg8wrHwnMvQvnLlCuLj453/jouLw5UrV2QvioiImuYytENCQnxV\nBxERSSG6cPjwYTEtLc3575UrV4qrV69u8B6NRiMC4IMPPvjgowUPjUbjKn6fyGUTBLvdjoEDB2Lv\n3r3o1asXnnrqKXz77bfQarVP+hYiIpJR0+2xH74YFoa1a9ciLS0NDocDv/vd7xjYRER+5HG7MSIi\n8h3Jd0RKuckmMzMT/fv3R3JyMk6ePOm1IpWmubFYv349kpOTMXToUIwfPx6nT5/2Q5W+IfXmqx9/\n/BFhYWHYuHGjD6vzLSljYTabMXz4cAwePBh6vd63BfpQc2Nx69YtpKenY9iwYRg8eDD+/ve/+75I\nH1i8eDF69OiBIUOGPPE9Lc5NKRPfdrtd1Gg0YlFRkVhTUyMmJyeLFoulwXvy8vLEjIwMURRF8ciR\nI+KYMWPcmmRXOiljcejQIfHu3buiKIrijh07gnosHr5v8uTJ4owZM0RBEPxQqfykjMWdO3dEnU4n\nlpSUiKIoijdv3vRHqbKTMhbLly8X//CHP4iiWD8O0dHRYm1trT/KldX+/fvFEydOiIMHD27ydXdy\nU9KRtpSbbLZu3YqFCxcCAMaMGYO7d+/i+vXrUjavKlLGYuzYsYiKigJQPxaXL1/2R6myk3rzVU5O\nDoxGI7p16+aHKn1Dylh88803MBgMiIuLAwB07drVH6XKTspY9OzZE/fu3QMA3Lt3D126dEFYmMtT\nbKo0ceJEdO7c+Ymvu5ObkkJbyk02Tb0nEMOqpTccffnll5g+fbovSvM5qf8vtmzZgtdeew1A4F77\nL2Uszp8/j19//RWTJ0/GqFGj8PXXX/u6TJ+QMhZLlizB2bNn0atXLyQnJ2PNmjW+LlMR3MlNSX/a\npP6iiY3OaQbiL2hLfqYffvgBX331FQ4ePChjRf4jZSzeeOMNrF69GiEhIRBF8Tf/RwKFlLGora3F\niRMnsHfvXlRWVmLs2LF4+umn0b9/fx9U6DtSxmLlypUYNmwYzGYzbDYbUlNTcerUKXTo0MEHFSpL\nS3NTUmjHxsaipKTE+e+SkhLnR7wnvefy5cuIjY2VsnlVkTIWAHD69GksWbIEO3fudPnxSM2kjMXx\n48cxf/58APUnn3bs2IHw8HDMmjXLp7XKTcpYxMfHo2vXrmjTpg3atGmDSZMm4dSpUwEX2lLG4tCh\nQ1i2bBkAQKPRoG/fvigoKMCoUaN8Wqu/uZWbUibTa2trxX79+olFRUVidXV1syciDx8+HLAn36SM\nxS+//CJqNBrx8OHDfqrSN6SMxeNeeuklMTc314cV+o6UsbBareKUKVNEu90uVlRUiIMHDxbPnj3r\np4rlI2Us3nzzTTErK0sURVEsLS0VY2Njxdu3b/ujXNkVFRVJOhEpNTclHWk/6SabL774AgDw6quv\nYvr06di+fTsSExPRrl07rFu3zv0/PwomZSzef/993LlzxzmPGx4ejmPHjvmzbFlIGYtgIWUskpKS\nkJ6ejqFDh6JVq1ZYsmQJdDqdnyv3Pilj8e6772LRokVITk5GXV0dPv74Y0RHR/u5cu977rnnsG/f\nPty6dQvx8fFYsWIFamtrAbifm7y5hohIRdhujIhIRRjaREQqwtAmIlIRhjYRkYowtImIVIShTUSk\nIgxtIiIVYWgTEanI/weulC1MVehgOgAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1080fd890>"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Unsupervised Learning"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Unsupervised Learning** addresses a different sort of problem. Here the data has no labels,\n",
      "and we are interested in finding similarities between the objects in question. In a sense,\n",
      "you can think of unsupervised learning as a means of discovering labels from the data itself.\n",
      "Unsupervised learning comprises tasks such as *dimensionality reduction*, *clustering*, and\n",
      "*density estimation*. For example, in the iris data discussed above, we can used unsupervised\n",
      "methods to determine combinations of the measurements which best display the structure of the\n",
      "data. As we\u2019ll see below, such a projection of the data can be used to visualize the\n",
      "four-dimensional dataset in two dimensions. Some more involved unsupervised learning problems are:\n",
      "\n",
      "- given detailed observations of distant galaxies, determine which features or combinations of\n",
      "  features summarize best the information.\n",
      "- given a mixture of two sound sources (for example, a person talking over some music),\n",
      "  separate the two (this is called the [blind source separation](http://en.wikipedia.org/wiki/Blind_signal_separation) problem).\n",
      "- given a video, isolate a moving object and categorize in relation to other moving objects which have been seen.\n",
      "\n",
      "Sometimes the two may even be combined: e.g. Unsupervised learning can be used to find useful\n",
      "features in heterogeneous data, and then these features can be used within a supervised\n",
      "framework."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Dimensionality Reduction\n",
      "\n",
      "Principle Component Analysis (PCA) is a dimension reduction technique that can find the combinations of variables that explain the most variance. Here we'll show an example of using PCA for visualization.\n",
      "\n",
      "Consider the iris dataset. It cannot be visualized in a single 2D plot, as it has 4 features. We are going to extract 2 combinations of sepal and petal dimensions to visualize it:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X, y = iris.data, iris.target\n",
      "from sklearn.decomposition import PCA\n",
      "pca = PCA(n_components=2)\n",
      "pca.fit(X)\n",
      "X_reduced = pca.transform(X)\n",
      "print \"Reduced dataset shape:\", X_reduced.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Reduced dataset shape: (150, 2)\n"
       ]
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.scatter(X_reduced[:, 0], X_reduced[:, 1], c=y);"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEACAYAAAC9Gb03AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1080ff650>"
       ]
      }
     ],
     "prompt_number": 16
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print \"Meaning of the 2 components:\"\n",
      "for component in pca.components_:\n",
      "    print \" + \".join(\"%.2f x %s\" % (value, name)\n",
      "                     for value, name in zip(component, iris.feature_names))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Meaning of the 2 components:\n",
        "0.36 x sepal length (cm) + -0.08 x sepal width (cm) + 0.86 x petal length (cm) + 0.36 x petal width (cm)\n",
        "-0.66 x sepal length (cm) + -0.73 x sepal width (cm) + 0.18 x petal length (cm) + 0.07 x petal width (cm)\n"
       ]
      }
     ],
     "prompt_number": 17
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Clustering\n",
      "\n",
      "Clustering groups together observations that are homogeneous with respect to a given criterion, finding ''clusters'' in the data.\n",
      "\n",
      "Note that these clusters will uncover relevent hidden structure of the data only if the criterion used highlights it."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.cluster import KMeans\n",
      "k_means = KMeans(n_clusters=3, random_state=0)\n",
      "k_means.fit(X_reduced)\n",
      "y_pred = k_means.predict(X_reduced)\n",
      "\n",
      "plt.scatter(X_reduced[:, 0], X_reduced[:, 1], c=y_pred);"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEACAYAAAC9Gb03AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYuSo0BNCiJGjQk8IIUbu/wG6C7XsYYA9NQAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x105cbedd0>"
       ]
      }
     ],
     "prompt_number": 18
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "A recap on Scikit-learn's estimator interface"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Scikit-learn strives to have a uniform interface across all methods,\n",
      "and we\u2019ll see examples of these below. Given a scikit-learn *estimator*\n",
      "object named `model`, the following methods are available:\n",
      "\n",
      "- Available in **all Estimators**\n",
      "  + `model.fit()` : fit training data. For supervised learning applications,\n",
      "    this accepts two arguments: the data `X` and the labels `y` (e.g. `model.fit(X, y)`).\n",
      "    For unsupervised learning applications, this accepts only a single argument,\n",
      "    the data `X` (e.g. `model.fit(X)`).\n",
      "- Available in **supervised estimators**\n",
      "  + `model.predict()` : given a trained model, predict the label of a new set of data.\n",
      "    This method accepts one argument, the new data `X_new` (e.g. `model.predict(X_new)`),\n",
      "    and returns the learned label for each object in the array.\n",
      "  + `model.predict_proba()` : For classification problems, some estimators also provide\n",
      "    this method, which returns the probability that a new observation has each categorical label.\n",
      "    In this case, the label with the highest probability is returned by `model.predict()`.\n",
      "  + `model.score()` : for classification or regression problems, most (all?) estimators implement\n",
      "    a score method.  Scores are between 0 and 1, with a larger score indicating a better fit.\n",
      "- Available in **unsupervised estimators**\n",
      "  + `model.transform()` : given an unsupervised model, transform new data into the new basis.\n",
      "    This also accepts one argument `X_new`, and returns the new representation of the data based\n",
      "    on the unsupervised model.\n",
      "  + `model.fit_transform()` : some estimators implement this method,\n",
      "    which more efficiently performs a fit and a transform on the same input data."
     ]
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": [
      "Measuring Performance"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "An important piece of machine learning is **model validation**: that is, determining how well your model will generalize from the training data to future unlabeled data. Let's look at an example using the *nearest neighbor classifier*. This is a very simple classifier: it simply stores all training data, and for any unknown quantity, simply returns the label of the closest training point.\n",
      "\n",
      "With the iris data, it very easily returns the correct prediction for each of the input points:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.neighbors import KNeighborsClassifier\n",
      "X, y = iris.data, iris.target\n",
      "clf = KNeighborsClassifier(n_neighbors=1)\n",
      "clf.fit(X, y)\n",
      "y_pred = clf.predict(X)\n",
      "print(np.all(y == y_pred))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "True\n"
       ]
      }
     ],
     "prompt_number": 19
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "A more useful way to look at the results is to view the **confusion matrix**, or the matrix showing the frequency of inputs and outputs:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.metrics import confusion_matrix\n",
      "print(confusion_matrix(y, y_pred))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "[[50  0  0]\n",
        " [ 0 50  0]\n",
        " [ 0  0 50]]\n"
       ]
      }
     ],
     "prompt_number": 20
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "For each class, all 50 training samples are correctly identified. But this **does not mean that our model is perfect!** In particular, such a model generalizes extremely poorly to new data. We can simulate this by splitting our data into a *training set* and a *testing set*. Scikit-learn contains some convenient routines to do this:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.cross_validation import train_test_split\n",
      "Xtrain, Xtest, ytrain, ytest = train_test_split(X, y)\n",
      "clf.fit(Xtrain, ytrain)\n",
      "ypred = clf.predict(Xtest)\n",
      "print(confusion_matrix(ytest, ypred))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "[[12  0  0]\n",
        " [ 0 11  3]\n",
        " [ 0  0 12]]\n"
       ]
      }
     ],
     "prompt_number": 21
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This paints a better picture of the true performance of our classifier: apparently there is some confusion between the second and third species, which we might anticipate given what we've seen of the data above.\n",
      "\n",
      "This is why it's **extremely important** to use a train/test split when evaluating your models.  We'll go into more depth on model evaluation later in this tutorial."
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Exercise: Interactive Demo on linearly separable data"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "To get a feel for how a classification task works, please run the **svm_gui.py** file available in the repository (use the command below). It will open an interactive window which allows you to click and add points and see how this affects the model.\n",
      "\n",
      "What is the fewest number of points you can create that are not linearly separable (that is, for which a linear kernel fails to build a correct discriminative model)?"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# %run fig_code/svm_gui.py\n",
      "#    or, if this doesn't work, try\n",
      "# !python fig_code/svm_gui.py"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 22
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Flow chart: how do I choose what to do with my data set?"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This is a flow chart created by scikit-learn super-contributor [Andreas Mueller](https://github.com/amueller) which gives a nice summary of which algorithms to choose in various situations. Keep it around as a handy reference!"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<a href=\"http://scikit-learn.org/dev/tutorial/machine_learning_map/index.html\"><img src=\"http://scikit-learn.org/dev/_static/ml_map.png\" width=100%></a>\n",
      "\n",
      "Original source on the [scikit-learn website](http://scikit-learn.org/stable/tutorial/machine_learning_map/)"
     ]
    }
   ],
   "metadata": {}
  }
 ]
}