ml_src/iris-model-builder.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function, division"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Acquire Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_iris"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Load Iris Data\n",
    "iris_data = load_iris()\n",
    "features = iris_data.data  # X Matrix\n",
    "target = iris_data.target  # y Target Array\n",
    "\n",
    "# Names\n",
    "feature_names = iris_data.feature_names\n",
    "target_names = iris_data.target_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>target</th>\n",
       "      <th>flower</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>0</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>5.7</td>\n",
       "      <td>4.4</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74</th>\n",
       "      <td>6.4</td>\n",
       "      <td>2.9</td>\n",
       "      <td>4.3</td>\n",
       "      <td>1.3</td>\n",
       "      <td>1</td>\n",
       "      <td>versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>106</th>\n",
       "      <td>4.9</td>\n",
       "      <td>2.5</td>\n",
       "      <td>4.5</td>\n",
       "      <td>1.7</td>\n",
       "      <td>2</td>\n",
       "      <td>virginica</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>7.7</td>\n",
       "      <td>2.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>2.3</td>\n",
       "      <td>2</td>\n",
       "      <td>virginica</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>0</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143</th>\n",
       "      <td>6.8</td>\n",
       "      <td>3.2</td>\n",
       "      <td>5.9</td>\n",
       "      <td>2.3</td>\n",
       "      <td>2</td>\n",
       "      <td>virginica</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>4.8</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>4.9</td>\n",
       "      <td>2.4</td>\n",
       "      <td>3.3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>versicolor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>5.8</td>\n",
       "      <td>2.7</td>\n",
       "      <td>4.1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>versicolor</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal_length  sepal_width  petal_length  petal_width  target      flower\n",
       "0             5.1          3.5           1.4          0.2       0      setosa\n",
       "15            5.7          4.4           1.5          0.4       0      setosa\n",
       "74            6.4          2.9           4.3          1.3       1  versicolor\n",
       "106           4.9          2.5           4.5          1.7       2   virginica\n",
       "118           7.7          2.6           6.9          2.3       2   virginica\n",
       "3             4.6          3.1           1.5          0.2       0      setosa\n",
       "143           6.8          3.2           5.9          2.3       2   virginica\n",
       "30            4.8          3.1           1.6          0.2       0      setosa\n",
       "57            4.9          2.4           3.3          1.0       1  versicolor\n",
       "67            5.8          2.7           4.1          1.0       1  versicolor"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df = pd.DataFrame(features, columns=feature_names)\n",
    "iris_df.rename(columns={\n",
    "        \"sepal length (cm)\": \"sepal_length\",\n",
    "        \"sepal width (cm)\": \"sepal_width\",\n",
    "        \"petal length (cm)\": \"petal_length\",\n",
    "        \"petal width (cm)\": \"petal_width\"\n",
    "    }, inplace=True)\n",
    "iris_df[\"target\"] = target\n",
    "iris_df[\"flower\"] = iris_df[\"target\"].map(lambda val: target_names[val])\n",
    "iris_df.sample(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Explore Relationships"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAs4AAAGACAYAAAC9VuStAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x116a822e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 2, figsize=(12, 6))\n",
    "for idx, (x_col, y_col) in enumerate([(\"sepal_length\", \"petal_length\"), (\"sepal_width\", \"petal_width\")]):\n",
    "    for color, label in zip('bgr', target_names):\n",
    "        subset = iris_df[iris_df.flower == label]\n",
    "        ax.flat[idx].scatter(subset[x_col], subset[y_col], s=50, c=color, label=label)\n",
    "        ax.flat[idx].set_xlabel(x_col)\n",
    "        ax.flat[idx].set_ylabel(y_col)\n",
    "        ax.flat[idx].legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Build Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "\n",
    "def build_knn(features, target, **param_grid):\n",
    "    est = GridSearchCV(KNeighborsClassifier(), param_grid=param_grid, cv=5, scoring=\"accuracy\")\n",
    "    est.fit(features, target)\n",
    "    return est"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "knn_est = build_knn(features, target, n_neighbors=list(range(3, 10, 2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Score: 0.98\n",
      "Best Params {'n_neighbors': 7}\n"
     ]
    }
   ],
   "source": [
    "# Best Score\n",
    "print(\"Best Score:\", knn_est.best_score_)\n",
    "print(\"Best Params\", knn_est.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[mean: 0.96667, std: 0.02108, params: {'n_neighbors': 3},\n",
       " mean: 0.97333, std: 0.02494, params: {'n_neighbors': 5},\n",
       " mean: 0.98000, std: 0.01633, params: {'n_neighbors': 7},\n",
       " mean: 0.97333, std: 0.02494, params: {'n_neighbors': 9}]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_est.grid_scores_"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}