Data_Prep_with_Classification_Exercise_V2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Prep: Classification (exercise)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook we will use data on house sales in King County, Washington state to predict house prices using simple (one input) linear regression using Scikit-Learn. Scikit-Learn is a simple, efficient and well-designed tool for data analysis and machine learning. You will use it and seaborn to:\n",
    "\n",
    "    Use Python to load, visualize and prepare the data\n",
    "    Use SciKit-Learn functions to compute important summary statistics\n",
    "    Use matplotlib and Seaborn to plot and visualize the data\n",
    "    Fix nulls and zeros\n",
    "    Convert price to a form suitable for classification\n",
    "    Divide dataset into train and test subsets\n",
    "    Scale the features\n",
    "    Run: Random Forests, Support Vector Machines, and Neural Net\n",
    "\n",
    "In this notebook you will be provided with mostly complete code. The goal here is to use classification to model the dataset and gain experience with Scikit-Learn.  You will use classification to sort homes into two catagories and compare predicted versus actual results. Clearly this a simplified problem, but easy to understand for new data scientists.\n",
    "\n",
    "First, we need to load all of the appropriate libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.svm import SVC\n",
    "from sklearn import svm\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.impute import SimpleImputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing = pd.read_csv(\"kc_house_data2.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(21613, 21)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>date</th>\n",
       "      <th>price</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>sqft_living</th>\n",
       "      <th>sqft_lot</th>\n",
       "      <th>floors</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>view</th>\n",
       "      <th>...</th>\n",
       "      <th>grade</th>\n",
       "      <th>sqft_above</th>\n",
       "      <th>sqft_basement</th>\n",
       "      <th>yr_built</th>\n",
       "      <th>yr_renovated</th>\n",
       "      <th>zipcode</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>sqft_living15</th>\n",
       "      <th>sqft_lot15</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7129300520</td>\n",
       "      <td>20141013T000000</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1180</td>\n",
       "      <td>5650</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>7</td>\n",
       "      <td>1180</td>\n",
       "      <td>0</td>\n",
       "      <td>1955</td>\n",
       "      <td>0</td>\n",
       "      <td>98178</td>\n",
       "      <td>47.5112</td>\n",
       "      <td>-122.257</td>\n",
       "      <td>1340</td>\n",
       "      <td>5650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6414100192</td>\n",
       "      <td>20141209T000000</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.25</td>\n",
       "      <td>2570</td>\n",
       "      <td>7242</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>7</td>\n",
       "      <td>2170</td>\n",
       "      <td>400</td>\n",
       "      <td>1951</td>\n",
       "      <td>1991</td>\n",
       "      <td>98125</td>\n",
       "      <td>47.7210</td>\n",
       "      <td>-122.319</td>\n",
       "      <td>1690</td>\n",
       "      <td>7639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5631500400</td>\n",
       "      <td>20150225T000000</td>\n",
       "      <td>180000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>770</td>\n",
       "      <td>10000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>6</td>\n",
       "      <td>770</td>\n",
       "      <td>0</td>\n",
       "      <td>1933</td>\n",
       "      <td>0</td>\n",
       "      <td>98028</td>\n",
       "      <td>47.7379</td>\n",
       "      <td>-122.233</td>\n",
       "      <td>2720</td>\n",
       "      <td>8062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2487200875</td>\n",
       "      <td>20141209T000000</td>\n",
       "      <td>604000</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.00</td>\n",
       "      <td>1960</td>\n",
       "      <td>5000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>7</td>\n",
       "      <td>1050</td>\n",
       "      <td>910</td>\n",
       "      <td>1965</td>\n",
       "      <td>0</td>\n",
       "      <td>98136</td>\n",
       "      <td>47.5208</td>\n",
       "      <td>-122.393</td>\n",
       "      <td>1360</td>\n",
       "      <td>5000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1954400510</td>\n",
       "      <td>20150218T000000</td>\n",
       "      <td>510000</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.00</td>\n",
       "      <td>1680</td>\n",
       "      <td>8080</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>8</td>\n",
       "      <td>1680</td>\n",
       "      <td>0</td>\n",
       "      <td>1987</td>\n",
       "      <td>0</td>\n",
       "      <td>98074</td>\n",
       "      <td>47.6168</td>\n",
       "      <td>-122.045</td>\n",
       "      <td>1800</td>\n",
       "      <td>7503</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           id             date   price  bedrooms  bathrooms  sqft_living  \\\n",
       "0  7129300520  20141013T000000       0       3.0       1.00         1180   \n",
       "1  6414100192  20141209T000000       0       3.0       2.25         2570   \n",
       "2  5631500400  20150225T000000  180000       2.0       1.00          770   \n",
       "3  2487200875  20141209T000000  604000       4.0       3.00         1960   \n",
       "4  1954400510  20150218T000000  510000       3.0       2.00         1680   \n",
       "\n",
       "   sqft_lot  floors  waterfront  view  ...  grade  sqft_above  sqft_basement  \\\n",
       "0      5650     1.0           0     0  ...      7        1180              0   \n",
       "1      7242     2.0           0     0  ...      7        2170            400   \n",
       "2     10000     1.0           0     0  ...      6         770              0   \n",
       "3      5000     1.0           0     0  ...      7        1050            910   \n",
       "4      8080     1.0           0     0  ...      8        1680              0   \n",
       "\n",
       "   yr_built  yr_renovated  zipcode      lat     long  sqft_living15  \\\n",
       "0      1955             0    98178  47.5112 -122.257           1340   \n",
       "1      1951          1991    98125  47.7210 -122.319           1690   \n",
       "2      1933             0    98028  47.7379 -122.233           2720   \n",
       "3      1965             0    98136  47.5208 -122.393           1360   \n",
       "4      1987             0    98074  47.6168 -122.045           1800   \n",
       "\n",
       "   sqft_lot15  \n",
       "0        5650  \n",
       "1        7639  \n",
       "2        8062  \n",
       "3        5000  \n",
       "4        7503  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2b003b92e748>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(housing.corr(), linewidths=.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, let's drop the features that don't correlate well with price.  Although we don't do it in this exercise, you might also create new features like bathrooms squared.  Squaring a feature like bathrooms makes each incremental bathroom more valuable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2b003db63f60>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "drop_list = ['id', 'sqft_lot', 'condition', 'yr_built', 'yr_renovated', 'zipcode',\n",
    "             'sqft_lot15', 'lat', 'long', 'date']\n",
    "housing.drop(drop_list, axis=1, inplace=True)\n",
    "sns.heatmap(housing.corr(), linewidths=.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Clearing out the unused features leaves us with 10 potentially useful features. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>sqft_living</th>\n",
       "      <th>floors</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>view</th>\n",
       "      <th>grade</th>\n",
       "      <th>sqft_above</th>\n",
       "      <th>sqft_basement</th>\n",
       "      <th>sqft_living15</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1180</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>1180</td>\n",
       "      <td>0</td>\n",
       "      <td>1340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.25</td>\n",
       "      <td>2570</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>2170</td>\n",
       "      <td>400</td>\n",
       "      <td>1690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>180000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>770</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>770</td>\n",
       "      <td>0</td>\n",
       "      <td>2720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>604000</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.00</td>\n",
       "      <td>1960</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>1050</td>\n",
       "      <td>910</td>\n",
       "      <td>1360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>510000</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.00</td>\n",
       "      <td>1680</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>1680</td>\n",
       "      <td>0</td>\n",
       "      <td>1800</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    price  bedrooms  bathrooms  sqft_living  floors  waterfront  view  grade  \\\n",
       "0       0       3.0       1.00         1180     1.0           0     0      7   \n",
       "1       0       3.0       2.25         2570     2.0           0     0      7   \n",
       "2  180000       2.0       1.00          770     1.0           0     0      6   \n",
       "3  604000       4.0       3.00         1960     1.0           0     0      7   \n",
       "4  510000       3.0       2.00         1680     1.0           0     0      8   \n",
       "\n",
       "   sqft_above  sqft_basement  sqft_living15  \n",
       "0        1180              0           1340  \n",
       "1        2170            400           1690  \n",
       "2         770              0           2720  \n",
       "3        1050            910           1360  \n",
       "4        1680              0           1800  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2b003dd74d30>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "housing.plot(kind=\"scatter\", x=\"bedrooms\", y=\"price\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We saw this plot in the Regression session.  It looks linear; however, it appears that there distribution in not \"normal\" but skews towards lower price houses.  Here is a plot:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist = housing['price']\n",
    "sns.distplot(hist)\n",
    "housing['price'].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is clearly not a normal distribution.  Just looking at the plot, the mean appears to be around 500k.  Now we need to remove any nulls.  Luckily we have funtions to help with this,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "housing.fillna(value=5, inplace=True)\n",
    "housing.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ok, we filled in nulls arbitrarily with the numeric value 5.  This is close to a mean value and it is unlikely that four houses with wrong bedrooms nor one house with 5 floors is a problem.  Next we need to fill in zeros, i.e. cases where the price is zero.  Clearly selling a house for zero is an incorrect value, and will skew the results.  Let's fill them in with the mean."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing[housing.price == 0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Good, we only have four cases.  Rather than use the mean we could use the results from a linear regression model or use MICE, but with only four holes it is not worth the effort."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "col_sum = housing.sum(axis=0)\n",
    "print(col_sum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean = (col_sum.price)/(21613-4)\n",
    "print(mean)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can now replace the zeros with the mean.  As you can see from the first 5 rows we have fixed the first two rows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing.price.replace(to_replace=0, value=mean, inplace=True)\n",
    "housing.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For classification we want to assign each row to a category.  Therefore, we need to convert the price column into a label.  Our assignment here is to find house in the top 50% of the market so we will have to roughly divide the dataset in half.  Recall that our mean is around 500k."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "mybins = [2, 500000, 9000000]\n",
    "group_names = ['too_low', 'good']\n",
    "housing['price'] = pd.cut(housing['price'], bins=mybins, labels = group_names)\n",
    "housing['price'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Good, although we started with the mean, we see that there are more lower priced houses.  If we need 50/50 by quantity then we would need to figure what price is the median.  But this split is perfect for now.  The algorithms don't work well with labels, so our next step is too enclode each category.  Since we only have two categories, we only need 0 and 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_price = LabelEncoder()\n",
    "housing['price'] = new_price.fit_transform(housing['price'])\n",
    "housing['price'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "scrolled": true
   },
   "source": [
    "It might be interesting to plot the counts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.countplot(housing['price'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we are going to split the data set into features versus correct answers (i.e. encoded price)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = housing.drop('price', axis=1)\n",
    "y = housing['price']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some routines may complain about float64 versus int64 so let's convert everything to float64"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = X.astype(float)\n",
    "X.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = y.astype(float)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is the traditional split for creating some test data.  Luckily there is a handy routine for this!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2)\n",
    "print(len(X_train))\n",
    "print(len(X_test))\n",
    "print(len(y_train))\n",
    "print(len(y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the various features are not on the same scale they can bias the results.  Therefore it makes sense to scale the features via StandardScaler.  **Warning: if you scale the training set, then you need to scale the test set.** Notice that we fit and transform the training but only trainsform the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "myscaler = StandardScaler()\n",
    "X_train = myscaler.fit_transform(X_train)\n",
    "X_test = myscaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfc = RandomForestClassifier(n_estimators=200)\n",
    "rfc.fit(X_train, y_train)\n",
    "predict_rfc = rfc.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(classification_report(y_test, predict_rfc))\n",
    "print(confusion_matrix(y_test, predict_rfc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Support Vector Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "svc = SVC()\n",
    "svc.fit(X_train, y_train)\n",
    "predict_svc = svc.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(classification_report(y_test, predict_svc))\n",
    "print(confusion_matrix(y_test, predict_svc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mlp = MLPClassifier(hidden_layer_sizes=(10,10,10),max_iter=1000)\n",
    "mlp.fit(X_train, y_train)\n",
    "predict_mlp = mlp.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(classification_report(y_test, predict_mlp))\n",
    "print(confusion_matrix(y_test, predict_mlp))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}