notebooks/07_knn_sklearn.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# K-nearest neighbors and scikit-learn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Agenda\n",
    "\n",
    "1. Review of the iris dataset\n",
    "2. Human learning on the iris dataset\n",
    "3. K-nearest neighbors (KNN) classification\n",
    "4. Review of supervised learning\n",
    "5. Benefits and drawbacks of scikit-learn\n",
    "6. Requirements for working with data in scikit-learn\n",
    "7. scikit-learn's 4-step modeling pattern\n",
    "8. Tuning a KNN model\n",
    "9. Comparing KNN with other models\n",
    "\n",
    "## Lesson goals\n",
    "\n",
    "1. Learn how the modeling process works\n",
    "2. Learn how scikit-learn works\n",
    "3. Learn how KNN works"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Review of the iris dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# read the iris data into a DataFrame\n",
    "import pandas as pd\n",
    "url = 'http://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data'\n",
    "col_names = ['sepal_length', 'sepal_width', 'petal_length', 'petal_width', 'species']\n",
    "iris = pd.read_csv(url, header=None, names=col_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>sepal_length</th>\n",
       "      <th>sepal_width</th>\n",
       "      <th>petal_length</th>\n",
       "      <th>petal_width</th>\n",
       "      <th>species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   sepal_length  sepal_width  petal_length  petal_width      species\n",
       "0           5.1          3.5           1.4          0.2  Iris-setosa\n",
       "1           4.9          3.0           1.4          0.2  Iris-setosa\n",
       "2           4.7          3.2           1.3          0.2  Iris-setosa\n",
       "3           4.6          3.1           1.5          0.2  Iris-setosa\n",
       "4           5.0          3.6           1.4          0.2  Iris-setosa"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Terminology\n",
    "\n",
    "- **150 observations** (n=150): each observation is one iris flower\n",
    "- **4 features** (p=4): sepal length, sepal width, petal length, and petal width\n",
    "- **Response**: iris species\n",
    "- **Classification problem** since response is categorical"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Human learning on the iris dataset\n",
    "\n",
    "How did we (as humans) predict the species of an iris?\n",
    "\n",
    "1. We observe that the different species had (somewhat) dissimilar measurements.\n",
    "2. We focus on features that seemed to correlate with the response.\n",
    "3. We create a set of rules (using those features) to predict the species of an unknown iris.\n",
    "\n",
    "We assumed that if an **unknown iris** has measurements similar to **previous irises**, then its species is most likely the same as those previous irises."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# allow plots to appear in the notebook\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# increase default figure and font sizes for easier viewing\n",
    "plt.rcParams['figure.figsize'] = (6, 4)\n",
    "plt.rcParams['font.size'] = 14\n",
    "\n",
    "# create a custom colormap\n",
    "from matplotlib.colors import ListedColormap\n",
    "cmap_bold = ListedColormap(['#FF0000', '#00FF00', '#0000FF'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# all features and response need to be numbers for scikit-learn\n",
    "# map each iris species to a number\n",
    "iris['species_num'] = iris.species.map({'Iris-setosa':0, 'Iris-versicolor':1, 'Iris-virginica':2})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x114125ad0>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZwAAAD7CAYAAABexyJvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1141e1910>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create a scatter plot of PETAL LENGTH versus PETAL WIDTH and color by SPECIES\n",
    "iris.plot(kind='scatter', x='petal_length', y='petal_width', c='species_num', colormap=cmap_bold) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x110f021d0>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAD7CAYAAABE+8LhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110ecdad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create a scatter plot of SEPAL LENGTH versus SEPAL WIDTH and color by SPECIES\n",
    "iris.plot(kind='scatter', x='sepal_length', y='sepal_width', c='species_num', colormap=cmap_bold)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K-nearest neighbors (KNN) classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Pick a value for K.\n",
    "2. Search for the K observations in the data that are \"nearest\" to the measurements of the unknown iris.\n",
    "    - Euclidian distance is often used as the distance metric, but other metrics are allowed.\n",
    "3. Use the most popular response value from the K \"nearest neighbors\" as the predicted response value for the unknown iris."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KNN classification map for iris (K=1)\n",
    "\n",
    "![1NN classification map](images/iris_01nn_map.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KNN classification map for iris (K=5)\n",
    "\n",
    "![5NN classification map](images/iris_05nn_map.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KNN classification map for iris (K=15)\n",
    "\n",
    "![15NN classification map](images/iris_15nn_map.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KNN classification map for iris (K=50)\n",
    "\n",
    "![50NN classification map](images/iris_50nn_map.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Question:** What's the \"best\" value for K in this case?\n",
    "\n",
    "**Answer:** The value which produces the most accurate predictions on **unseen data**. We want to create a model that generalizes!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Review of supervised learning\n",
    "\n",
    "![Supervised learning diagram](images/supervised_learning.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Benefits and drawbacks of scikit-learn\n",
    "\n",
    "**Benefits:**\n",
    "\n",
    "- Consistent interface to machine learning models\n",
    "- Provides many tuning parameters but with sensible defaults\n",
    "- Exceptional documentation\n",
    "- Rich set of functionality for companion tasks\n",
    "- Active community for development and support\n",
    "\n",
    "**Potential drawbacks:**\n",
    "\n",
    "- Harder (than R) to get started with machine learning\n",
    "- Less emphasis (than R) on model interpretability\n",
    "\n",
    "Ben Lorica: [Six reasons why I recommend scikit-learn](http://radar.oreilly.com/2013/12/six-reasons-why-i-recommend-scikit-learn.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Requirements for working with data in scikit-learn\n",
    "\n",
    "1. Features and response should be **separate objects**\n",
    "2. Features and response should be entirely **numeric**\n",
    "3. Features and response should be **NumPy arrays** (or easily converted to NumPy arrays)\n",
    "4. Features and response should have **specific shapes** (outlined below)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## scikit-learn's 6-step modeling pattern\n",
    "**Step 1:** Create a feature matrix and response vector\n",
    "\n",
    "**Step 2:** Decide on the estimator you want to to use and import that class\n",
    "\n",
    "**Step 3:** \"Instantiate\" the \"estimator\"\n",
    "\n",
    "**Step 4:** Fit the model with data (aka \"model training\")\n",
    "\n",
    "**Step 5:** Use the model to predict the response for a new observation\n",
    "\n",
    "**Step 6:** Evaluate the error or accuracy of the model\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Let's utilize scikit-learn's 6-step modeling pattern\n",
    "(We will explore evaluating accuracy next lesson)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Step 1:** Create a feature matrix and response vector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>sepal_length</th>\n",
       "      <th>sepal_width</th>\n",
       "      <th>petal_length</th>\n",
       "      <th>petal_width</th>\n",
       "      <th>species</th>\n",
       "      <th>species_num</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>6.7</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>Iris-virginica</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>6.3</td>\n",
       "      <td>2.5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>Iris-virginica</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147</th>\n",
       "      <td>6.5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Iris-virginica</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>6.2</td>\n",
       "      <td>3.4</td>\n",
       "      <td>5.4</td>\n",
       "      <td>2.3</td>\n",
       "      <td>Iris-virginica</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>5.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.1</td>\n",
       "      <td>1.8</td>\n",
       "      <td>Iris-virginica</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal_length  sepal_width  petal_length  petal_width         species  \\\n",
       "145           6.7          3.0           5.2          2.3  Iris-virginica   \n",
       "146           6.3          2.5           5.0          1.9  Iris-virginica   \n",
       "147           6.5          3.0           5.2          2.0  Iris-virginica   \n",
       "148           6.2          3.4           5.4          2.3  Iris-virginica   \n",
       "149           5.9          3.0           5.1          1.8  Iris-virginica   \n",
       "\n",
       "     species_num  \n",
       "145            2  \n",
       "146            2  \n",
       "147            2  \n",
       "148            2  \n",
       "149            2  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# all features and response need to be numbers for scikit-learn\n",
    "# map each iris species to a number\n",
    "iris['species_num'] = iris.species.map({'Iris-setosa':0, 'Iris-versicolor':1, 'Iris-virginica':2})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# store feature matrix in \"X\"\n",
    "feature_cols = ['sepal_length', 'sepal_width', 'petal_length', 'petal_width']\n",
    "X = iris[feature_cols]\n",
    "print (X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# alternative ways to create \"X\"\n",
    "X = iris.drop(['species', 'species_num'], axis=1)\n",
    "X = iris.loc[:, 'sepal_length':'petal_width']\n",
    "X = iris.iloc[:, 0:4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# store response vector in \"y\"\n",
    "y = iris.species_num\n",
    "print(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "<type 'numpy.ndarray'>\n"
     ]
    }
   ],
   "source": [
    "# check X's type\n",
    "print type(X)\n",
    "print type(X.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.series.Series'>\n",
      "<type 'numpy.ndarray'>\n"
     ]
    }
   ],
   "source": [
    "# check y's type\n",
    "print type(y)\n",
    "print type(y.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(150, 4)\n"
     ]
    }
   ],
   "source": [
    "# check X's shape (n = number of observations, p = number of features)\n",
    "print X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(150,)\n"
     ]
    }
   ],
   "source": [
    "# check y's shape (single dimension with length n)\n",
    "print y.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Step 2:** Decide on the estimator you want to to use and import that class\n",
    "- \"Estimator\" is scikit-learn's term for \"model\"\n",
    "- [See scikit-learn flowchart that aids estimator choice decisions](http://3.bp.blogspot.com/-dofu6J0sZ8o/UrctKb69QdI/AAAAAAAADfg/79ewPecn5XU/s1600/scikit-learn-flow-chart.jpg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Step 3:** \"Instantiate\" the \"estimator\"\n",
    "\n",
    "- \"Instantiate\" means \"make an instance of\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sklearn.neighbors.classification.KNeighborsClassifier"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# make an instance of a KNeighborsClassifier object\n",
    "knn = KNeighborsClassifier(n_neighbors=1)\n",
    "type(knn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Created an object that \"knows\" how to do K-nearest neighbors classification, and is just waiting for data\n",
    "- Name of the object does not matter\n",
    "- Can specify tuning parameters (aka \"hyperparameters\") during this step\n",
    "- All parameters not specified are set to their defaults"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
      "           metric_params=None, n_neighbors=1, p=2, weights='uniform')\n"
     ]
    }
   ],
   "source": [
    "print knn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Step 4:** Fit the model with data (aka \"model training\")\n",
    "\n",
    "- Model is \"learning\" the relationship between X and y in our \"training data\"\n",
    "- Process through which learning occurs varies by model\n",
    "- Occurs in-place"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "           metric_params=None, n_neighbors=1, p=2, weights='uniform')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.fit(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Once a model has been fit with data, it's called a \"fitted model\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### **Step 5:** Use the model to predict the response for a new observation\n",
    "\n",
    "- New observations are called \"out-of-sample\" data\n",
    "- Uses the information it learned during the model training process"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2])"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Since we used all our data to create the model, let's create a hypothetical new observation to see the prediction\n",
    "new_observation = [3, 5, 4, 2]\n",
    "knn.predict(new_observation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 2])"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Since we used all our data to create the model, let's create a hypothetical new observation to see the prediction\n",
    "new_observation = [[3, 5, 4, 2],[4,1, 5,7]]\n",
    "knn.predict(new_observation)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Returns a NumPy array, and we keep track of what the numbers \"mean\"\n",
    "- Can predict for multiple observations at once"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_new = [[3, 5, 4, 2], [5, 4, 3, 2]]\n",
    "knn.predict(X_new)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tuning a KNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1])"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# instantiate the model (using the value K=5)\n",
    "knn = KNeighborsClassifier(n_neighbors=5)\n",
    "\n",
    "# fit the model with data\n",
    "knn.fit(X, y)\n",
    "\n",
    "# predict the response for new observations\n",
    "knn.predict(X_new)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Question:** Which model produced the correct predictions for the two unknown irises?\n",
    "\n",
    "**Answer:** We don't know, because these are **out-of-sample observations**, meaning that we don't know the true response values. Our goal with supervised learning is to build models that generalize to out-of-sample data. However, we can't truly measure how well our models will perform on out-of-sample data.\n",
    "\n",
    "**Question:** Does that mean that we have to guess how well our models are likely to do?\n",
    "\n",
    "**Answer:** Thankfully, no. In the next class, we'll discuss **model evaluation procedures**, which allow us to use our existing labeled data to estimate how well our models are likely to perform on out-of-sample data. These procedures will help us to tune our models and choose between different types of models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0. ,  0.8,  0.2],\n",
       "       [ 0. ,  1. ,  0. ]])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculate predicted probabilities of class membership\n",
    "knn.predict_proba(X_new)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparing KNN with other models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Advantages of KNN:**\n",
    "\n",
    "- Simple to understand and explain\n",
    "- Model training is fast\n",
    "- Can be used for classification and regression\n",
    "\n",
    "**Disadvantages of KNN:**\n",
    "\n",
    "- Must store all of the training data\n",
    "- Prediction phase can be slow when n is large\n",
    "- Sensitive to irrelevant features\n",
    "- Sensitive to the scale of the data (for some distance metric options)\n",
    "- Accuracy is (generally) not competitive with the best supervised learning methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}