Skip to content

Instantly share code, notes, and snippets.

@machinelearning-blog

machinelearning-blog/Housing-Prices-with-California-Housing-Dataset.ipynb

Last active February 4, 2024 22:45
Show Gist options
  • Star 2 You must be signed in to star a gist
  • Fork 1 You must be signed in to fork a gist
  • Save machinelearning-blog/76b50b18c7db3408646cc8d18c50c20b to your computer and use it in GitHub Desktop.
Save machinelearning-blog/76b50b18c7db3408646cc8d18c50c20b to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
Housing-Prices-with-California-Housing-Dataset.ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Introduction\n",
"\n",
"In this notebook I explained how to tackle a machine learning from the beginning to the end, at the example of predicting housing prices. I wrote about a lot of tool & techniques, that are common in the machine learning landscape.\n",
"\n",
"After reading it, you will know everything you need, to build your own regression system. I got everything I cover in this notebook from the book \"Hands-On Machine Learning with Scikit_learn & Tensorflow\" from Aurelien Geron. I used the California Housing Dataset from the statlib repository.\n",
"\n",
"If you just want to have a quick look at what I did and what I've learned throughout writing this notebook, you can just read the [summary](#Summary) at the very end. \n",
"\n",
"\n",
"\n",
"## Table of Contents:\n",
"\n",
"- [Import Libraries](#importo) \n",
" \n",
"- [California housing Dataset](#data)\n",
"\n",
"- [Big Picture](#bigp)\n",
"\n",
"- [Train/Test Split](#split)\n",
"\n",
"- [Extensive Data Exploration](#explo)\n",
" - [Searching for Correlations](#corre)\n",
" \n",
" \n",
"\n",
"\n",
"- [Data Preparation](#prep)\n",
" - [How to process Categorical Attributes and Text](#categ)\n",
" - [Custom transfer](#custom) \n",
" - [Feature Scaling](#scale)\n",
" - [Pipelines for tranformation](#pipe)\n",
" \n",
" \n",
" \n",
"- [Train Models](#train)\n",
"\n",
"- [Fine Tuning](#tune)\n",
"\n",
"- [Evaluation](#evaluation)\n",
"\n",
"- [Summary](#summary)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='importo'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Import Libraries"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='import'></a>"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"_cell_guid": "872b0186-0c2c-41f8-8183-2804ee5a7e7d",
"_uuid": "320de3a5d061657e16529fde48ec12317519e900",
"collapsed": true
},
"outputs": [],
"source": [
"# Common imports\n",
"import pandas as pd\n",
"import numpy as np\n",
"\n",
"# data visualization\n",
"import matplotlib.pyplot as plt\n",
"from sklearn.model_selection import train_test_split\n",
"\n",
"# random seed to make output stable across runs\n",
"np.random.seed(42)\n",
"\n",
"# data visualization\n",
"%matplotlib inline\n",
"import matplotlib\n",
"import matplotlib.pyplot as plt\n",
"plt.rcParams['axes.labelsize'] = 14\n",
"plt.rcParams['xtick.labelsize'] = 12\n",
"plt.rcParams['ytick.labelsize'] = 12"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='data'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "1458144a-cf17-4964-9454-55631f3833fb",
"_uuid": "58a79c8848e62fcdcfbba9780d0d23c9e6893e3b"
},
"source": [
"# **California housing Dataset**\n",
"\n",
"The dataset we will use is the \"California Housing Prices\" dataset from the statlib repository, which is based on data from the 1990 census. This dataset offers great opportunities for learning. The prediciton task for this dataset wil be to predict housing prices based on several features."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Get the data:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import os\n",
"import tarfile\n",
"from six.moves import urllib\n",
"\n",
"root = \"https://raw.githubusercontent.com/ageron/handson-ml/master/\"\n",
"path = os.path.join(\"datasets\", \"housing\")\n",
"source = root + \"datasets/housing/housing.tgz\"\n",
"\n",
"def fetch_housing_data(housing_url=source, path=path):\n",
" if not os.path.isdir(path):\n",
" os.makedirs(path)\n",
" tgz_path = os.path.join(path, \"housing.tgz\")\n",
" urllib.request.urlretrieve(housing_url, tgz_path)\n",
" housing_tgz = tarfile.open(tgz_path)\n",
" housing_tgz.extractall(path=path)\n",
" housing_tgz.close()"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"fetch_housing_data()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import pandas as pd\n",
"\n",
"def load_data(housing_path=path):\n",
" csv = os.path.join(housing_path, \"housing.csv\")\n",
" return pd.read_csv(csv)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='bigpop'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Big Picture\n",
"\n",
"You are asked to build a machine learning model that predicts the median house value for a given \"district\". This data has metrics like population, median income, median housing price for each block group in califorania. Block groups are smallest geographical unit for which the US Census Bureu publishes data. I will just call them \"districts\" for now.\n",
"\n",
"Frame the problem:\n",
"\n",
"Is it supervised, unsupervised or reinforcement learning ? Is it a classification task, a regression task or something different ? \n",
"Our problem is clearly a supervised learning task, because you have labeled input data. It is also clearly a regression task since you have to predict a numeric value. "
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"_cell_guid": "b931c2d4-d061-4755-a778-50d2d25c7cd5",
"_uuid": "d5703416fad1ee2662fae3a120d72aeef6edaa7d"
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>longitude</th>\n",
" <th>latitude</th>\n",
" <th>housing_median_age</th>\n",
" <th>total_rooms</th>\n",
" <th>total_bedrooms</th>\n",
" <th>population</th>\n",
" <th>households</th>\n",
" <th>median_income</th>\n",
" <th>median_house_value</th>\n",
" <th>ocean_proximity</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>-122.23</td>\n",
" <td>37.88</td>\n",
" <td>41.0</td>\n",
" <td>880.0</td>\n",
" <td>129.0</td>\n",
" <td>322.0</td>\n",
" <td>126.0</td>\n",
" <td>8.3252</td>\n",
" <td>452600.0</td>\n",
" <td>NEAR BAY</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>-122.22</td>\n",
" <td>37.86</td>\n",
" <td>21.0</td>\n",
" <td>7099.0</td>\n",
" <td>1106.0</td>\n",
" <td>2401.0</td>\n",
" <td>1138.0</td>\n",
" <td>8.3014</td>\n",
" <td>358500.0</td>\n",
" <td>NEAR BAY</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>-122.24</td>\n",
" <td>37.85</td>\n",
" <td>52.0</td>\n",
" <td>1467.0</td>\n",
" <td>190.0</td>\n",
" <td>496.0</td>\n",
" <td>177.0</td>\n",
" <td>7.2574</td>\n",
" <td>352100.0</td>\n",
" <td>NEAR BAY</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>-122.25</td>\n",
" <td>37.85</td>\n",
" <td>52.0</td>\n",
" <td>1274.0</td>\n",
" <td>235.0</td>\n",
" <td>558.0</td>\n",
" <td>219.0</td>\n",
" <td>5.6431</td>\n",
" <td>341300.0</td>\n",
" <td>NEAR BAY</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>-122.25</td>\n",
" <td>37.85</td>\n",
" <td>52.0</td>\n",
" <td>1627.0</td>\n",
" <td>280.0</td>\n",
" <td>565.0</td>\n",
" <td>259.0</td>\n",
" <td>3.8462</td>\n",
" <td>342200.0</td>\n",
" <td>NEAR BAY</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" longitude latitude housing_median_age total_rooms total_bedrooms \\\n",
"0 -122.23 37.88 41.0 880.0 129.0 \n",
"1 -122.22 37.86 21.0 7099.0 1106.0 \n",
"2 -122.24 37.85 52.0 1467.0 190.0 \n",
"3 -122.25 37.85 52.0 1274.0 235.0 \n",
"4 -122.25 37.85 52.0 1627.0 280.0 \n",
"\n",
" population households median_income median_house_value ocean_proximity \n",
"0 322.0 126.0 8.3252 452600.0 NEAR BAY \n",
"1 2401.0 1138.0 8.3014 358500.0 NEAR BAY \n",
"2 496.0 177.0 7.2574 352100.0 NEAR BAY \n",
"3 558.0 219.0 5.6431 341300.0 NEAR BAY \n",
"4 565.0 259.0 3.8462 342200.0 NEAR BAY "
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing = load_data()\n",
"housing.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "163c2889-2e14-49e2-9e24-22378be91933",
"_uuid": "414836f4bebac63df7f781a53496d384e493f4f7"
},
"source": [
"Above you can see the top 5 rows of the dataset with the \"head()\" method. \n",
"\n",
"Each row represents one district. The dataset has 10 attributes: longitude, latitude, housing_median_age, total_rooms, total_bedrooms, population, households, median_income, median_house_value and ocean_proximity."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"_cell_guid": "b358bf1a-dda4-4a78-8f3e-151a5fba654a",
"_uuid": "de323020d946366a847f11e994ee5c0f7cb60ee3"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"<class 'pandas.core.frame.DataFrame'>\n",
"RangeIndex: 20640 entries, 0 to 20639\n",
"Data columns (total 10 columns):\n",
"longitude 20640 non-null float64\n",
"latitude 20640 non-null float64\n",
"housing_median_age 20640 non-null float64\n",
"total_rooms 20640 non-null float64\n",
"total_bedrooms 20433 non-null float64\n",
"population 20640 non-null float64\n",
"households 20640 non-null float64\n",
"median_income 20640 non-null float64\n",
"median_house_value 20640 non-null float64\n",
"ocean_proximity 20640 non-null object\n",
"dtypes: float64(9), object(1)\n",
"memory usage: 1.6+ MB\n"
]
}
],
"source": [
"housing.info()"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "7c590fa9-5623-4475-bfcd-8b9ef0d737f8",
"_uuid": "6aaeb407a71ab6608b4b60a6aa751d145908e5c9"
},
"source": [
"Now you can see a brief description of the data using the \"info()\" method above.\n",
"\n",
"Notice that there are 20,640 instances (entries) in the whole dataset. Also notice that the total_bedrooms attribute has only 20,433 non-null values which means that 207 districts don't have this feature. You can also see that ocean_proximity is not numerical and probably a categorical attribute. You can find out how many districts belong to each attribute by using the \"value_counts()\" method, which I used below."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"_cell_guid": "3e014e14-f988-454a-945c-0ea0e5a19656",
"_uuid": "083715a4c86ac5c09fd2def17e74e6e8c5db6a70"
},
"outputs": [
{
"data": {
"text/plain": [
"<1H OCEAN 9136\n",
"INLAND 6551\n",
"NEAR OCEAN 2658\n",
"NEAR BAY 2290\n",
"ISLAND 5\n",
"Name: ocean_proximity, dtype: int64"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing[\"ocean_proximity\"].value_counts()"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "e9c09c3b-c881-4a7e-8bb7-d87a88d25f46",
"_uuid": "8f897e5bb3228e78a61feb20c77a48b2437c6f8b"
},
"source": [
"Now I will use the \"describe()\" method to show a summary of the numerical attributes:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"_cell_guid": "f8ccf747-dc71-4690-bbda-5fb9d0e7aefd",
"_uuid": "9f50c3e7fbe4b846ff8109b85628321b4675462b"
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>longitude</th>\n",
" <th>latitude</th>\n",
" <th>housing_median_age</th>\n",
" <th>total_rooms</th>\n",
" <th>total_bedrooms</th>\n",
" <th>population</th>\n",
" <th>households</th>\n",
" <th>median_income</th>\n",
" <th>median_house_value</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>count</th>\n",
" <td>20640.000000</td>\n",
" <td>20640.000000</td>\n",
" <td>20640.000000</td>\n",
" <td>20640.000000</td>\n",
" <td>20433.000000</td>\n",
" <td>20640.000000</td>\n",
" <td>20640.000000</td>\n",
" <td>20640.000000</td>\n",
" <td>20640.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>mean</th>\n",
" <td>-119.569704</td>\n",
" <td>35.631861</td>\n",
" <td>28.639486</td>\n",
" <td>2635.763081</td>\n",
" <td>537.870553</td>\n",
" <td>1425.476744</td>\n",
" <td>499.539680</td>\n",
" <td>3.870671</td>\n",
" <td>206855.816909</td>\n",
" </tr>\n",
" <tr>\n",
" <th>std</th>\n",
" <td>2.003532</td>\n",
" <td>2.135952</td>\n",
" <td>12.585558</td>\n",
" <td>2181.615252</td>\n",
" <td>421.385070</td>\n",
" <td>1132.462122</td>\n",
" <td>382.329753</td>\n",
" <td>1.899822</td>\n",
" <td>115395.615874</td>\n",
" </tr>\n",
" <tr>\n",
" <th>min</th>\n",
" <td>-124.350000</td>\n",
" <td>32.540000</td>\n",
" <td>1.000000</td>\n",
" <td>2.000000</td>\n",
" <td>1.000000</td>\n",
" <td>3.000000</td>\n",
" <td>1.000000</td>\n",
" <td>0.499900</td>\n",
" <td>14999.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25%</th>\n",
" <td>-121.800000</td>\n",
" <td>33.930000</td>\n",
" <td>18.000000</td>\n",
" <td>1447.750000</td>\n",
" <td>296.000000</td>\n",
" <td>787.000000</td>\n",
" <td>280.000000</td>\n",
" <td>2.563400</td>\n",
" <td>119600.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>50%</th>\n",
" <td>-118.490000</td>\n",
" <td>34.260000</td>\n",
" <td>29.000000</td>\n",
" <td>2127.000000</td>\n",
" <td>435.000000</td>\n",
" <td>1166.000000</td>\n",
" <td>409.000000</td>\n",
" <td>3.534800</td>\n",
" <td>179700.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>75%</th>\n",
" <td>-118.010000</td>\n",
" <td>37.710000</td>\n",
" <td>37.000000</td>\n",
" <td>3148.000000</td>\n",
" <td>647.000000</td>\n",
" <td>1725.000000</td>\n",
" <td>605.000000</td>\n",
" <td>4.743250</td>\n",
" <td>264725.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>max</th>\n",
" <td>-114.310000</td>\n",
" <td>41.950000</td>\n",
" <td>52.000000</td>\n",
" <td>39320.000000</td>\n",
" <td>6445.000000</td>\n",
" <td>35682.000000</td>\n",
" <td>6082.000000</td>\n",
" <td>15.000100</td>\n",
" <td>500001.000000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" longitude latitude housing_median_age total_rooms \\\n",
"count 20640.000000 20640.000000 20640.000000 20640.000000 \n",
"mean -119.569704 35.631861 28.639486 2635.763081 \n",
"std 2.003532 2.135952 12.585558 2181.615252 \n",
"min -124.350000 32.540000 1.000000 2.000000 \n",
"25% -121.800000 33.930000 18.000000 1447.750000 \n",
"50% -118.490000 34.260000 29.000000 2127.000000 \n",
"75% -118.010000 37.710000 37.000000 3148.000000 \n",
"max -114.310000 41.950000 52.000000 39320.000000 \n",
"\n",
" total_bedrooms population households median_income \\\n",
"count 20433.000000 20640.000000 20640.000000 20640.000000 \n",
"mean 537.870553 1425.476744 499.539680 3.870671 \n",
"std 421.385070 1132.462122 382.329753 1.899822 \n",
"min 1.000000 3.000000 1.000000 0.499900 \n",
"25% 296.000000 787.000000 280.000000 2.563400 \n",
"50% 435.000000 1166.000000 409.000000 3.534800 \n",
"75% 647.000000 1725.000000 605.000000 4.743250 \n",
"max 6445.000000 35682.000000 6082.000000 15.000100 \n",
"\n",
" median_house_value \n",
"count 20640.000000 \n",
"mean 206855.816909 \n",
"std 115395.615874 \n",
"min 14999.000000 \n",
"25% 119600.000000 \n",
"50% 179700.000000 \n",
"75% 264725.000000 \n",
"max 500001.000000 "
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing.describe()"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "664b0cd9-56e8-4a22-9e78-b5bede35194c",
"_uuid": "e9b663bf2af0a6172bae7e716bd7d4b4cbcf3cc4"
},
"source": [
"I hope the count, mean, min and max rows are self explanatory for you. Be aware of the fact, that all null values are ignored here (count of total_bedrooms is 20,433 not 20,640). The std row shows you the standard deviation, which quantifies the amount of variation between the values. The 25%, 50%, 75% are the percentiles. For an example 75% of the districts have housing_median_age of lower than 37 while 50% are lower than 29 and 25% are lower than 18. These are often called the 25th percentile (or 1st quartile), the median and the 75th percentile (or 3rd quartile).\n",
"\n",
"Another good way to explore a given dataset is to plot a histogrom of each numerical attribute. A histogram shows you the number of instances (vertical axes) and the number of the given value range (horizontal axes)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"_cell_guid": "abe6a2ce-7e7b-4e3b-88e6-e9afe761251b",
"_uuid": "6f1d2724f48dd644a4d732e9e64e4999d76336e4"
},
"outputs": [
{
"data": {
"text/plain": [
"array([[<matplotlib.axes._subplots.AxesSubplot object at 0x114c0dc50>,\n",
" <matplotlib.axes._subplots.AxesSubplot object at 0x1149e5a20>,\n",
" <matplotlib.axes._subplots.AxesSubplot object at 0x114a22898>],\n",
" [<matplotlib.axes._subplots.AxesSubplot object at 0x114a5d898>,\n",
" <matplotlib.axes._subplots.AxesSubplot object at 0x114a93898>,\n",
" <matplotlib.axes._subplots.AxesSubplot object at 0x114a938d0>],\n",
" [<matplotlib.axes._subplots.AxesSubplot object at 0x114b01358>,\n",
" <matplotlib.axes._subplots.AxesSubplot object at 0x114b39358>,\n",
" <matplotlib.axes._subplots.AxesSubplot object at 0x110e9f278>]], dtype=object)"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAABJkAAANhCAYAAABErGrRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n9v98LjRWTwNklQAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x10ead8710>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"%matplotlib inline\n",
"housing.hist(bins=50, figsize=(20, 15))"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "0ea36f7a-b503-4e6d-a79e-6ab7ce0ea0fe",
"_uuid": "0e08f0a577cd1a5be4e7f28ebd39ac978234a0aa"
},
"source": [
"What you should notice about the histograms above:\n",
"\n",
"* The attributes have varying scales, which we will discuss later on in this post.\n",
"* Many of the histograms are \"tail heavy\" which means that they extend further to the right of the median than to the left which makes it harder for an algortihm to detect patterns. We will transform the later on.\n",
"* The median_income attribute is not in US Dollar because the data has been scaled and capped at 15.0001 and at 0.4999. This is called a \"preprocessed attribute\" and is common within machine learning, but you should understand how the data was preprocessed.\n",
"* The housing_median_age and the median_house_value attributes are also capped. That the median_house_value is capped could be a serious problem, because this is your label (what you want to predict) and your model could learn that the price never goes beyond that limit. In this case we only have the option to remove the capped one or to collect the right labels for them.\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='split'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "340ad737-3d30-424a-a84d-397e404e27c6",
"_uuid": "4b3399e428021578725afc4e0deb640911c8a8ab"
},
"source": [
"# Train/Test Split\n",
"\n",
"It is important that we now set a part of the data aside. Your brain is an amazing pattern detection system and therefore extremly prune to overfitting. If you would also look at the test set during the visual explaration, which we will do now, you may find some patterns unconsciously within the data, which let you select a particular algorithm that leads to an overfitted model. Because of that you could launch a system that performs not as well on new data than expected.\n",
"\n",
"\n",
"We use the straight forward \"train_test_split()\" method from sklearn to split our data into train and test subsets. For further information you can look at the [documentation](http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html)."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"_cell_guid": "d6b045b4-082e-461a-97c6-24b774889c63",
"_uuid": "5eb0a7ae42cf906b48cffd64b42c365d7c5383f4"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"16512 Train Instances + 4128 Test Instances\n"
]
}
],
"source": [
"train_set, test_set = train_test_split(housing, test_size=0.2, random_state=42)\n",
"print(len(train_set), \"Train Instances +\", len(test_set), \"Test Instances\")"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "145751be-9210-40b3-abce-633ed1f77685",
"_uuid": "e06df4e0dfac8b984094cac5f8c1b0909690f587",
"collapsed": true
},
"source": [
"Now we splitted our data purely random, which is fine if you have a large dataset, but if it is not, you could have a sampling bias. When a company decides to call 10,000 people because of a survey, they want to make sure that these people represent the whole population. For an example the US population consists out of 49% male and 51% female, so a well conducted survey would try to maintain this ratio, which is called **stratified sampling**. The population is divided into homogeneous subgroups, called **strata** and the right number of instances is sampled from each **stratum** to makes sure that the data really represents the population. \n",
"\n",
"Imagine that you talked with real estate experts and they tell you, that the median_icome attribute is very important, when you want to predict housing prices. If that's the case, you want a test set that is really representative of the income categories of the dataset. Because median_income is continous, you need to convert it into a categorical attribute. If you look again at the holograms we analyzed detailed above, you may notice that most median income values are clustered around 20,000 - 50,000 but some go far beyond 60,000. It is very important that you don't have too many strata and that each stratum should have a sufficient number of instances. If this is not the case, the estimate of the stratums importance may be biased and your model could think that a stratum is less important. \n",
"\n",
"The code below transforms the median_income attribute into a categorical one by dividing the median income by 1.5 to limit the number of income categories and rounds it up using \"np.ceil()\" to have discrete categories. It merges all the categories that are greater than 5 into category 5. The categories are represented in the histogram below the code."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"_cell_guid": "d53ecae1-1a12-437c-8b12-6e1c6df4be27",
"_uuid": "c13ccd16196c87c7fdfb8f2e9cffec426b7ba541",
"collapsed": true
},
"outputs": [],
"source": [
"housing[\"income_categories\"] = np.ceil(housing[\"median_income\"] / 1.5)\n",
"housing[\"income_categories\"].where(housing[\"income_categories\"] < 5, 5.0, inplace=True)"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"_cell_guid": "094d446a-d9bf-4a39-9991-aa0e30baf674",
"_uuid": "b093861132c2c0e2a1af08c73aa3d74d775a24f3"
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAD/CAYAAAAe7cpbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSVLr/wFQCpeDgCVb9wAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x114e5a668>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.hist(housing[\"income_categories\"])\n",
"fig = plt.gcf()\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "5815dac5-5e9c-41d5-85fb-ef6eb14c0627",
"_uuid": "a6928ef0c5717217cd3f2131b3a253c8df1c23e4"
},
"source": [
"Lats but not least you need to do stratified sampling based on the income categories. You can use sklearn's \"StratifiedShuffleSplit\" class:"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"_cell_guid": "01aeb05a-04ea-437b-8eba-c8029ccf7b20",
"_uuid": "becf62d21a87a344032be9e10475e7f3e1e021b8",
"collapsed": true
},
"outputs": [],
"source": [
"from sklearn.model_selection import StratifiedShuffleSplit\n",
"\n",
"split = StratifiedShuffleSplit(n_splits=1, test_size=0.2, random_state=42)\n",
"for train_index, test_index in split.split(housing, housing[\"income_categories\"]):\n",
" strat_train_set = housing.loc[train_index]\n",
" strat_test_set = housing.loc[test_index]"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "c784c4d8-9709-4886-9593-c546e697cca0",
"_uuid": "7b46e4760db044bad7aae73d1bdaea2732cfae84"
},
"source": [
"The code below just generates the table and compares the different sampling types."
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"_cell_guid": "d6a8b443-b958-45e2-987e-e4165e6849f1",
"_uuid": "fa0f4a0bcc6cc65d39e0180557c4066b2d309647",
"collapsed": true
},
"outputs": [],
"source": [
"def income_categories_proportions(data):\n",
" return data[\"income_categories\"].value_counts() / len(data)\n",
"\n",
"train_set, test_set = train_test_split(housing, test_size=0.2, random_state=42)\n",
"\n",
"compare_props = pd.DataFrame({\n",
" \"Overall\": income_categories_proportions(housing),\n",
" \"Stratified\": income_categories_proportions(strat_test_set),\n",
" \"Random\": income_categories_proportions(test_set),\n",
"}).sort_index()\n"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"_cell_guid": "d9bb91f8-51fc-4b87-b8bc-a77eb7096c04",
"_uuid": "ebe18d6828da04fb13194fa81250f36f554d83b8"
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Overall</th>\n",
" <th>Random</th>\n",
" <th>Stratified</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>1.0</th>\n",
" <td>0.039826</td>\n",
" <td>0.040213</td>\n",
" <td>0.039729</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2.0</th>\n",
" <td>0.318847</td>\n",
" <td>0.324370</td>\n",
" <td>0.318798</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3.0</th>\n",
" <td>0.350581</td>\n",
" <td>0.358527</td>\n",
" <td>0.350533</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4.0</th>\n",
" <td>0.176308</td>\n",
" <td>0.167393</td>\n",
" <td>0.176357</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5.0</th>\n",
" <td>0.114438</td>\n",
" <td>0.109496</td>\n",
" <td>0.114583</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Overall Random Stratified\n",
"1.0 0.039826 0.040213 0.039729\n",
"2.0 0.318847 0.324370 0.318798\n",
"3.0 0.350581 0.358527 0.350533\n",
"4.0 0.176308 0.167393 0.176357\n",
"5.0 0.114438 0.109496 0.114583"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"compare_props"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "6a432f0a-9929-4e2f-b071-d6f293ad185e",
"_uuid": "69f8046e5f4340e7251896eb66f18e435c7f2df2"
},
"source": [
"Now we want to remove the income_categoreis attribute because we don't need it anymore."
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"_cell_guid": "0e11ebca-3c15-4ebe-a06c-cb462981d4b4",
"_uuid": "3a7e5f7d7183ef0b3c696976183457ae2146ceef",
"collapsed": true
},
"outputs": [],
"source": [
"housing = housing.drop(['income_categories'], axis=1)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='explo'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "5228a9b2-8c62-4a54-a022-9d3cbbdcb27d",
"_uuid": "ecd77ea95758b7f85b6f161c0c213d7f1dfc30f2"
},
"source": [
"# Extensive Data Exploration\n",
"\n",
"Now it is time for exploring the data. First of all, we want to visualize the geographical data with latitude and longitude. A good way to do this is to create a scatterplot of all the districts. It is important that you set alpha equal to 0.1, because then the scatterplot has a high density and therefore it is much easier to visualize."
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"_cell_guid": "21966a98-0cc8-4089-97d4-12f3105a8f64",
"_uuid": "ce2d50f66c002f902492d379722ddabc274afabf"
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x114e85e48>"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAERCAYAAACD9ivUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x114b2a9e8>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", alpha=0.1)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "d736b0dc-f5ba-4657-aadc-d1f8e6d97f41",
"_uuid": "4c6a62beff2ed8de88604de8565496fbbb96e4f8"
},
"source": [
"If you are fimiliar with California you can clearly see, the high density in the Bay Area, Los Angelos, San Diego and in the Central Valley around Sacramento and Fresno. Our brains are very good at finding patterns on pictures but you often need to play around with the parameters to make the important patterns really stand out.\n",
"\n",
"Now we will look at the housing prices at the scatterplot below. The color represents the price and the radius of each circle represents the districts population."
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"_cell_guid": "195b0445-60a9-4d54-88a1-a450e77a5569",
"_uuid": "299026b0a3e3840db42cac502ebb1b8f96a03e5b"
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.legend.Legend at 0x1a1f425588>"
]
},
"execution_count": 18,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAmoAAAG1CAYAAABJUtgYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQKlQExERESlQ/w9C2urQJ4MDCgAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x114bbeef0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", alpha=0.4,\n",
" s=housing[\"population\"]/100, label=\"population\", figsize=(10,7),\n",
" c=\"median_house_value\", cmap=plt.get_cmap(\"jet\"), colorbar=True,\n",
" sharex=False)\n",
"plt.legend()"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "95789a78-fca9-4014-bb1d-190f1516ea64",
"_uuid": "b03ad5cc59fff5f1cba1682c75afa056ab3f026f"
},
"source": [
"On the scatterplot we can see, that the housing prices are related to the location (close to the ocean) and to the population density, but we know that the housing prices of costoal districts are not that high in Northern California, so we can't make that rule as simple as that. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='corre'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"### Searching for Correlations:\n",
"\n",
"The housing dataset isn't that large and therefore we can easily compute the correlations between every attribute using the \"corr()\" method. We will start by looking how much each attribute is correlated to the median house value."
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"_cell_guid": "cc59638c-4994-4446-a64c-ad5b5505f201",
"_uuid": "2491c4069e9ea2e1033d9a10aa32f81829f6e4bb"
},
"outputs": [
{
"data": {
"text/plain": [
"median_house_value 1.000000\n",
"median_income 0.688075\n",
"total_rooms 0.134153\n",
"housing_median_age 0.105623\n",
"households 0.065843\n",
"total_bedrooms 0.049686\n",
"population -0.024650\n",
"longitude -0.045967\n",
"latitude -0.144160\n",
"Name: median_house_value, dtype: float64"
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"correlations = housing.corr()\n",
"correlations[\"median_house_value\"].sort_values(ascending=False)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "e453cbc4-7b97-409f-8e07-d19e32d98729",
"_uuid": "9258d68633ec5b98a8f67fefac0d3a7a19961e25"
},
"source": [
"The coefficient of the correlation ranges from 1 to -1. The closer it is to 1 the more correlated it is and vice versa. Correlations that are close to 0, means that there is no correlation, neither negative or positive. You can see that the median_income is correlated the most with the median house value. Because of that, we will generate a more detailed scatterplot below:"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"_cell_guid": "0a02380a-7c96-46dd-bb6f-ee828c10ef48",
"_uuid": "ac969f7b9f57fc87f891e17a4bf3ce78bf77766f"
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x1a1f503b38>"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAaYAAAERCAYAAAAwvgPOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmi6b46fLJlYAAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x1a1f4236a0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"housing.plot(kind=\"scatter\", x=\"median_income\", y=\"median_house_value\", alpha=0.1)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "ca1bddd0-260e-402c-8e55-c97d78edc45e",
"_uuid": "841750a90fdeeb986e0e04dc49ec0f84bf191a63"
},
"source": [
"The scatterplot reveals, that the correlation is indeed very strong because we can clearly see an upward trend and the points are not to dispersed. We can also clearly see the price-cap, we talked about earlier, at 500 000 as a horizontal line. Other less obvious lines are around 450 000, 350 000 and 280 000. We may have to remove the corresponding districts to prevent the model from learning to reproduce these data faults. \n",
"\n",
"I hope that the previous explanation and visulizations made you more comfortable with the concepts of exploring data to gain insights. We identified a few very important faults in the data that we need to clean up before we can put the data into the machine learning model, that we will build later on. We also found some interesting correlations between the different attributes and we recognized the tail-heavy distribution, that we will also clean up later on. Of course a lot of things are different at every project you will work on, but the general guidelines will be the same and therefore you now already have a good understanding of some of those guidelines. \n",
"\n",
"Before we now actually prepare the data to fed it into the model, we should think about combinating a few attributes. For example, the number of rooms within a district is of course not very helpful, if you don't know how many households are within that district. You want the number of roms per household. The number of bedrooms isn't that helpful for the same reason, but it would make sense to compare it with the total number of rooms within a household. Also the population per household would be an interesting attribute. I will create these new attributes in the code below and then we will look at the correlation matrix again."
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"_cell_guid": "eac6f05d-4568-4ca4-a46e-7ff086b62c2f",
"_uuid": "e368da3928bbf5468ae625ab4f7a828945f2d649",
"collapsed": true
},
"outputs": [],
"source": [
"housing[\"rooms_per_household\"] = housing[\"total_rooms\"]/housing[\"households\"]\n",
"housing[\"bedrooms_per_room\"] = housing[\"total_bedrooms\"]/housing[\"total_rooms\"]\n",
"housing[\"population_per_household\"]=housing[\"population\"]/housing[\"households\"]"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"_cell_guid": "f2e63821-0d29-4fc2-b594-c3fcab1943bf",
"_uuid": "2f5f333e0da6931df9a73c176dcc1e7de6516587"
},
"outputs": [
{
"data": {
"text/plain": [
"median_house_value 1.000000\n",
"median_income 0.688075\n",
"rooms_per_household 0.151948\n",
"total_rooms 0.134153\n",
"housing_median_age 0.105623\n",
"households 0.065843\n",
"total_bedrooms 0.049686\n",
"population_per_household -0.023737\n",
"population -0.024650\n",
"longitude -0.045967\n",
"latitude -0.144160\n",
"bedrooms_per_room -0.255880\n",
"Name: median_house_value, dtype: float64"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"corr_matrix = housing.corr()\n",
"corr_matrix[\"median_house_value\"].sort_values(ascending=False)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "42f4cea0-499f-4480-9d6b-6d91e400f9fc",
"_uuid": "c65d667cd148dc626c697e7821a8e3e961b663bb"
},
"source": [
"Now we see, that the bedrooms_per_room attribute is more correlated with the median house value than the total number of rooms or bedrooms. Houses with a lower bedroom/room ratio tend to be more expensive. The rooms_per_household attribute is also better than the total number of rooms in a district. Obviously the larger the house, the higher the price.\n",
"\n",
"The part of exploring the data does not have to be fully thorough and accurated. The reason why we do it, is to gain a few important insights, that we can use when we build our model. Building proper Machine Learning models is a very iterative process, which means you gain some insights, build a model and then you try to gain more insights and update the model that it works better and so on. You will continue to do this, till you reached a satisfying accuracy and will probably try out several algorithms and paratemer adjustments on your way."
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "99700f05-7bb5-4152-bc2c-d65d89ea2f49",
"_uuid": "88282eed403c15f5efd7bcbca13c236c2d526aed"
},
"source": [
"<a id='prep'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "aefd5a9d-3542-4a97-a8df-6b1f85ac892f",
"_uuid": "b6fcf8279e5c379fc1a1e697ffe322211922f284"
},
"source": [
"# Data Preparation\n",
"\n",
"Now it is time to prepare the data so that our model can process it. We will write functions that do this instead of doing it manually. The reason for this is, that you can reuse these on a new dataset and at new projects you will work on. But first, let's revert to a clean training set by copying start_train_set and let's separate the predictors and the labels since we don't necessarily want to apply the same tranformations to the predictors and the target values."
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"_cell_guid": "2ff0a247-2391-43ee-96a1-79f40e1e6cb4",
"_uuid": "c4ef4efd75b37051101d159606a53c8193bb6bac",
"collapsed": true
},
"outputs": [],
"source": [
"housing = strat_train_set.drop(\"median_house_value\", axis=1) # drop labels for training set\n",
"housing_labels = strat_train_set[\"median_house_value\"].copy()"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "3a24b48a-fcfd-44be-a401-3fb8ecf727c3",
"_uuid": "81fcf6453c16fe41636d918f35d920525dca8105"
},
"source": [
"We noticed earlier that the total_bedrooms attribute has some missing values. Most Machine Learning algorithms can't work with datasets that have missing values. \n",
"\n",
"There are 3 ways to solve this problem:\n",
"\n",
"1.) You could remove the whole attribute\n",
"\n",
"2.) You could get rid of the districts that contain missing values.\n",
"\n",
"3.) You could replace them with zeros, the median or the mean\n",
"\n",
"We chose option 3 and will compute the median on the training set. Sklearn provides you with \"Imputer\" to do this. You first need to specify an Imputer instance, that specifies that you want to replace each attributes missing values with the median of that attribute. Because the median can only be computed on numerical attributes, we need to make a copy of the data without the ocean_proximity attribute that contains text and no numbers."
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"_cell_guid": "3140b527-f0d6-4f09-ab6b-16bc3c0d1019",
"_uuid": "8052b22a26805ebf055fb9352a333f7b0ce6e4f2",
"collapsed": true
},
"outputs": [],
"source": [
"median = housing[\"total_bedrooms\"].median()\n",
"\n",
"from sklearn.preprocessing import Imputer\n",
"\n",
"imputer = Imputer(strategy=\"median\")\n",
"\n",
"housing_num = housing.drop('ocean_proximity', axis=1)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "5d810bbc-b433-432c-b5df-d079ff8ca4b1",
"_uuid": "147fa5bd957cdf9e29cf1c963efd1ec4935ddb61"
},
"source": [
"Now we can fit the Imputer instance to the training data using the \"fit()\" method."
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"_cell_guid": "ee6c8f05-a663-4b59-aed6-6d698aa97a74",
"_uuid": "5784bdb13f25d4c4f65b1e0091e4567ccd7255ff"
},
"outputs": [
{
"data": {
"text/plain": [
"Imputer(axis=0, copy=True, missing_values='NaN', strategy='median', verbose=0)"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"imputer.fit(housing_num)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "d49f7485-94b4-41dd-906d-a5ccb456e2a8",
"_uuid": "34153a60b5926f3930bd5ddfbd29fe8d4831dd8e"
},
"source": [
"The imputer has simply computed the median of each attribute and stored the result in the statistics variable. We apllied the imputer to all numerical attributes, because it's safer if your system will go live sometimes.\n",
"\n",
"Now we can use the trained imputer to replace the missing values by the learned medians. This results in a plain numpy array but we will transform it back into a pandas dataframe. We will do these 2 steps in the code below."
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"_cell_guid": "635cd93c-532f-4798-bc7a-0cef5edb77cc",
"_uuid": "901400f20de5de10fc4cf143d2b67b3f1527e6a9",
"collapsed": true
},
"outputs": [],
"source": [
"X = imputer.transform(housing_num)\n",
"housing_tr = pd.DataFrame(X, columns=housing_num.columns, index = list(housing.index.values))\n",
"housing_tr = housing_tr.drop(['income_categories'], axis=1)"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {
"_cell_guid": "3ed2b9fd-0696-4cbd-b6a2-574f4459342d",
"_uuid": "e921a6ba2e7277335a9fd96025282b1b22488213"
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>longitude</th>\n",
" <th>latitude</th>\n",
" <th>housing_median_age</th>\n",
" <th>total_rooms</th>\n",
" <th>total_bedrooms</th>\n",
" <th>population</th>\n",
" <th>households</th>\n",
" <th>median_income</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>4629</th>\n",
" <td>-118.30</td>\n",
" <td>34.07</td>\n",
" <td>18.0</td>\n",
" <td>3759.0</td>\n",
" <td>433.0</td>\n",
" <td>3296.0</td>\n",
" <td>1462.0</td>\n",
" <td>2.2708</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6068</th>\n",
" <td>-117.86</td>\n",
" <td>34.01</td>\n",
" <td>16.0</td>\n",
" <td>4632.0</td>\n",
" <td>433.0</td>\n",
" <td>3038.0</td>\n",
" <td>727.0</td>\n",
" <td>5.1762</td>\n",
" </tr>\n",
" <tr>\n",
" <th>17923</th>\n",
" <td>-121.97</td>\n",
" <td>37.35</td>\n",
" <td>30.0</td>\n",
" <td>1955.0</td>\n",
" <td>433.0</td>\n",
" <td>999.0</td>\n",
" <td>386.0</td>\n",
" <td>4.6328</td>\n",
" </tr>\n",
" <tr>\n",
" <th>13656</th>\n",
" <td>-117.30</td>\n",
" <td>34.05</td>\n",
" <td>6.0</td>\n",
" <td>2155.0</td>\n",
" <td>433.0</td>\n",
" <td>1039.0</td>\n",
" <td>391.0</td>\n",
" <td>1.6675</td>\n",
" </tr>\n",
" <tr>\n",
" <th>19252</th>\n",
" <td>-122.79</td>\n",
" <td>38.48</td>\n",
" <td>7.0</td>\n",
" <td>6837.0</td>\n",
" <td>433.0</td>\n",
" <td>3468.0</td>\n",
" <td>1405.0</td>\n",
" <td>3.1662</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" longitude latitude housing_median_age total_rooms total_bedrooms \\\n",
"4629 -118.30 34.07 18.0 3759.0 433.0 \n",
"6068 -117.86 34.01 16.0 4632.0 433.0 \n",
"17923 -121.97 37.35 30.0 1955.0 433.0 \n",
"13656 -117.30 34.05 6.0 2155.0 433.0 \n",
"19252 -122.79 38.48 7.0 6837.0 433.0 \n",
"\n",
" population households median_income \n",
"4629 3296.0 1462.0 2.2708 \n",
"6068 3038.0 727.0 5.1762 \n",
"17923 999.0 386.0 4.6328 \n",
"13656 1039.0 391.0 1.6675 \n",
"19252 3468.0 1405.0 3.1662 "
]
},
"execution_count": 27,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"\n",
"sample_incomplete_rows = housing[housing.isnull().any(axis=1)].head()\n",
"housing_tr.loc[sample_incomplete_rows.index.values]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='categ'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "9501dbe4-8a1e-4064-9429-af02eb564c67",
"_uuid": "4613124b317e8cc1c245d0c9d32fa61393914429"
},
"source": [
"### How to process Categorical Attributes and Text\n",
"\n",
"Like I allready mentioned most of the machine learning algorithms can just work with numerical data. The ocean_proximity attribute is still a categorical feature and we will convert it now. We can use Pandas' factorize() method to convert this string categorical feature to an integer categorical feature.\n"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {
"_cell_guid": "be3d19fd-71f0-40ae-b5ad-99cae55d9f79",
"_uuid": "cb2cf1a14e79870fe7497d3287e32a7ea3282adc"
},
"outputs": [
{
"data": {
"text/plain": [
"array([0, 0, 1, 2, 0, 2, 0, 2, 0, 0])"
]
},
"execution_count": 28,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_cat = housing['ocean_proximity']\n",
"housing_cat_encoded, housing_categories = housing_cat.factorize()\n",
"housing_cat_encoded[:10]"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"_cell_guid": "56a47263-86a2-4172-bcfc-04b56b042f13",
"_uuid": "d7bb8acd1d3bb422931cedbaf4a322c9bd9f5c9a"
},
"outputs": [
{
"data": {
"text/plain": [
"Index(['<1H OCEAN', 'NEAR OCEAN', 'INLAND', 'NEAR BAY', 'ISLAND'], dtype='object')"
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_categories"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "340075ee-9d89-433f-bdd3-d4192db9e937",
"_uuid": "0ecc4178f097eeecbbc8a1ecd4a7a549a5712716"
},
"source": [
"One issues with this solution is that our algorithm will assume that two nearby values are more similar than two distant values, which is obviously not the case. For axample category 0 and 4 are more similar than categories 0 and 1. A common solution for this problem is to create one binary attribute per category: one attribute equal to 1 when the ctaegory is \"<1H OCEAN\" and 0 otherwise, another attribute equal to 1 when the category is \"INLAND\" and 0 otherwise, and so on. This is called \"one-hot-encoding\", because only one attribute will be equal to 1 (hot) and the others will be 0 (cold). \n",
"\n",
"We will do exactly this in the code cell below. We use the OneHotEncoder encoder, that sklearn provides, to convert numeric categorical values into one-hot vectors. Note that \"fit_transform()\" expects a 2D array but \"housing_cat\" is a 1D array. Therefore we need to reshape it."
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {
"_cell_guid": "198d183d-728d-46ea-a59f-289b87b6dc36",
"_uuid": "d2c108903dc6930444cbcf10d90ed444df66b31f"
},
"outputs": [
{
"data": {
"text/plain": [
"<16512x5 sparse matrix of type '<class 'numpy.float64'>'\n",
"\twith 16512 stored elements in Compressed Sparse Row format>"
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.preprocessing import OneHotEncoder\n",
"\n",
"encoder = OneHotEncoder()\n",
"housing_cat_1hot = encoder.fit_transform(housing_cat_encoded.reshape(-1,1))\n",
"housing_cat_1hot"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "0ab71854-2b1b-464a-99e4-9ef4bbc3392e",
"_uuid": "bc570a35e2523bfb59616774ff33f6cd46c0c00d"
},
"source": [
"The output is a so called **Scipy sparse matrix** and not a numpy array. A sparse matrix only stores the location of the non zero elements and doesn't stores a matrix full of zeros except for one per row. This saves a lot of memory and is very useful when you have categorical attributes with thousands of categories. For simplicity reasons we want it to be a usual numpy array and therefore we convert it in the code cell below, using the \"toarray()\" method."
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {
"_cell_guid": "475af798-ef4e-4c42-a410-d5b13f9a8941",
"_uuid": "bdd8b86c2ecc5758d9d6316ba4930e7658ee5da1"
},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 1., 0., 0., 0., 0.],\n",
" [ 1., 0., 0., 0., 0.],\n",
" [ 0., 1., 0., 0., 0.],\n",
" ..., \n",
" [ 0., 0., 1., 0., 0.],\n",
" [ 1., 0., 0., 0., 0.],\n",
" [ 0., 0., 0., 1., 0.]])"
]
},
"execution_count": 31,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_cat_1hot.toarray()"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "3412f021-1a4f-406c-b33c-782331dee702",
"_uuid": "307aade2d86e471ff0af3405e52f616c7c764567"
},
"source": [
"Note that there is a way that combines the two steps in one shot ( from categories to integer categories and then from integer categories to one-hot vectors). You can use Scikit-Learn's upcoming CategoricalEncoder class, which will soon be added to Scikit-Learn, and in the meantime you can use the code below (copied from Pull Request #9151 of Aurelien Geron)."
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {
"_cell_guid": "72f7fad4-8fdc-4c72-98a2-b04c3df5aa0f",
"_uuid": "87ddb259aa7906c0638092c8c9acccfb4baf72bb",
"collapsed": true
},
"outputs": [],
"source": [
"# copied from PR #9151.\n",
"# Just run this cell or copy it to your code. You don't need to understand it right now.\n",
"\n",
"from sklearn.base import BaseEstimator, TransformerMixin\n",
"from sklearn.utils import check_array\n",
"from sklearn.preprocessing import LabelEncoder\n",
"from scipy import sparse\n",
"\n",
"class CategoricalEncoder(BaseEstimator, TransformerMixin):\n",
" \"\"\"Encode categorical features as a numeric array.\n",
" The input to this transformer should be a matrix of integers or strings,\n",
" denoting the values taken on by categorical (discrete) features.\n",
" The features can be encoded using a one-hot aka one-of-K scheme\n",
" (``encoding='onehot'``, the default) or converted to ordinal integers\n",
" (``encoding='ordinal'``).\n",
" This encoding is needed for feeding categorical data to many scikit-learn\n",
" estimators, notably linear models and SVMs with the standard kernels.\n",
" Read more in the :ref:`User Guide <preprocessing_categorical_features>`.\n",
" Parameters\n",
" ----------\n",
" encoding : str, 'onehot', 'onehot-dense' or 'ordinal'\n",
" The type of encoding to use (default is 'onehot'):\n",
" - 'onehot': encode the features using a one-hot aka one-of-K scheme\n",
" (or also called 'dummy' encoding). This creates a binary column for\n",
" each category and returns a sparse matrix.\n",
" - 'onehot-dense': the same as 'onehot' but returns a dense array\n",
" instead of a sparse matrix.\n",
" - 'ordinal': encode the features as ordinal integers. This results in\n",
" a single column of integers (0 to n_categories - 1) per feature.\n",
" categories : 'auto' or a list of lists/arrays of values.\n",
" Categories (unique values) per feature:\n",
" - 'auto' : Determine categories automatically from the training data.\n",
" - list : ``categories[i]`` holds the categories expected in the ith\n",
" column. The passed categories are sorted before encoding the data\n",
" (used categories can be found in the ``categories_`` attribute).\n",
" dtype : number type, default np.float64\n",
" Desired dtype of output.\n",
" handle_unknown : 'error' (default) or 'ignore'\n",
" Whether to raise an error or ignore if a unknown categorical feature is\n",
" present during transform (default is to raise). When this is parameter\n",
" is set to 'ignore' and an unknown category is encountered during\n",
" transform, the resulting one-hot encoded columns for this feature\n",
" will be all zeros.\n",
" Ignoring unknown categories is not supported for\n",
" ``encoding='ordinal'``.\n",
" Attributes\n",
" ----------\n",
" categories_ : list of arrays\n",
" The categories of each feature determined during fitting. When\n",
" categories were specified manually, this holds the sorted categories\n",
" (in order corresponding with output of `transform`).\n",
" Examples\n",
" --------\n",
" Given a dataset with three features and two samples, we let the encoder\n",
" find the maximum value per feature and transform the data to a binary\n",
" one-hot encoding.\n",
" >>> from sklearn.preprocessing import CategoricalEncoder\n",
" >>> enc = CategoricalEncoder(handle_unknown='ignore')\n",
" >>> enc.fit([[0, 0, 3], [1, 1, 0], [0, 2, 1], [1, 0, 2]])\n",
" ... # doctest: +ELLIPSIS\n",
" CategoricalEncoder(categories='auto', dtype=<... 'numpy.float64'>,\n",
" encoding='onehot', handle_unknown='ignore')\n",
" >>> enc.transform([[0, 1, 1], [1, 0, 4]]).toarray()\n",
" array([[ 1., 0., 0., 1., 0., 0., 1., 0., 0.],\n",
" [ 0., 1., 1., 0., 0., 0., 0., 0., 0.]])\n",
" See also\n",
" --------\n",
" sklearn.preprocessing.OneHotEncoder : performs a one-hot encoding of\n",
" integer ordinal features. The ``OneHotEncoder assumes`` that input\n",
" features take on values in the range ``[0, max(feature)]`` instead of\n",
" using the unique values.\n",
" sklearn.feature_extraction.DictVectorizer : performs a one-hot encoding of\n",
" dictionary items (also handles string-valued features).\n",
" sklearn.feature_extraction.FeatureHasher : performs an approximate one-hot\n",
" encoding of dictionary items or strings.\n",
" \"\"\"\n",
"\n",
" def __init__(self, encoding='onehot', categories='auto', dtype=np.float64,\n",
" handle_unknown='error'):\n",
" self.encoding = encoding\n",
" self.categories = categories\n",
" self.dtype = dtype\n",
" self.handle_unknown = handle_unknown\n",
"\n",
" def fit(self, X, y=None):\n",
" \"\"\"Fit the CategoricalEncoder to X.\n",
" Parameters\n",
" ----------\n",
" X : array-like, shape [n_samples, n_feature]\n",
" The data to determine the categories of each feature.\n",
" Returns\n",
" -------\n",
" self\n",
" \"\"\"\n",
"\n",
" if self.encoding not in ['onehot', 'onehot-dense', 'ordinal']:\n",
" template = (\"encoding should be either 'onehot', 'onehot-dense' \"\n",
" \"or 'ordinal', got %s\")\n",
" raise ValueError(template % self.handle_unknown)\n",
"\n",
" if self.handle_unknown not in ['error', 'ignore']:\n",
" template = (\"handle_unknown should be either 'error' or \"\n",
" \"'ignore', got %s\")\n",
" raise ValueError(template % self.handle_unknown)\n",
"\n",
" if self.encoding == 'ordinal' and self.handle_unknown == 'ignore':\n",
" raise ValueError(\"handle_unknown='ignore' is not supported for\"\n",
" \" encoding='ordinal'\")\n",
"\n",
" X = check_array(X, dtype=np.object, accept_sparse='csc', copy=True)\n",
" n_samples, n_features = X.shape\n",
"\n",
" self._label_encoders_ = [LabelEncoder() for _ in range(n_features)]\n",
"\n",
" for i in range(n_features):\n",
" le = self._label_encoders_[i]\n",
" Xi = X[:, i]\n",
" if self.categories == 'auto':\n",
" le.fit(Xi)\n",
" else:\n",
" valid_mask = np.in1d(Xi, self.categories[i])\n",
" if not np.all(valid_mask):\n",
" if self.handle_unknown == 'error':\n",
" diff = np.unique(Xi[~valid_mask])\n",
" msg = (\"Found unknown categories {0} in column {1}\"\n",
" \" during fit\".format(diff, i))\n",
" raise ValueError(msg)\n",
" le.classes_ = np.array(np.sort(self.categories[i]))\n",
"\n",
" self.categories_ = [le.classes_ for le in self._label_encoders_]\n",
"\n",
" return self\n",
"\n",
" def transform(self, X):\n",
" \"\"\"Transform X using one-hot encoding.\n",
" Parameters\n",
" ----------\n",
" X : array-like, shape [n_samples, n_features]\n",
" The data to encode.\n",
" Returns\n",
" -------\n",
" X_out : sparse matrix or a 2-d array\n",
" Transformed input.\n",
" \"\"\"\n",
" X = check_array(X, accept_sparse='csc', dtype=np.object, copy=True)\n",
" n_samples, n_features = X.shape\n",
" X_int = np.zeros_like(X, dtype=np.int)\n",
" X_mask = np.ones_like(X, dtype=np.bool)\n",
"\n",
" for i in range(n_features):\n",
" valid_mask = np.in1d(X[:, i], self.categories_[i])\n",
"\n",
" if not np.all(valid_mask):\n",
" if self.handle_unknown == 'error':\n",
" diff = np.unique(X[~valid_mask, i])\n",
" msg = (\"Found unknown categories {0} in column {1}\"\n",
" \" during transform\".format(diff, i))\n",
" raise ValueError(msg)\n",
" else:\n",
" # Set the problematic rows to an acceptable value and\n",
" # continue `The rows are marked `X_mask` and will be\n",
" # removed later.\n",
" X_mask[:, i] = valid_mask\n",
" X[:, i][~valid_mask] = self.categories_[i][0]\n",
" X_int[:, i] = self._label_encoders_[i].transform(X[:, i])\n",
"\n",
" if self.encoding == 'ordinal':\n",
" return X_int.astype(self.dtype, copy=False)\n",
"\n",
" mask = X_mask.ravel()\n",
" n_values = [cats.shape[0] for cats in self.categories_]\n",
" n_values = np.array([0] + n_values)\n",
" indices = np.cumsum(n_values)\n",
"\n",
" column_indices = (X_int + indices[:-1]).ravel()[mask]\n",
" row_indices = np.repeat(np.arange(n_samples, dtype=np.int32),\n",
" n_features)[mask]\n",
" data = np.ones(n_samples * n_features)[mask]\n",
"\n",
" out = sparse.csc_matrix((data, (row_indices, column_indices)),\n",
" shape=(n_samples, indices[-1]),\n",
" dtype=self.dtype).tocsr()\n",
" if self.encoding == 'onehot-dense':\n",
" return out.toarray()\n",
" else:\n",
" return out"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "c5987945-432d-4046-8bbd-e51b33ad5cff",
"_uuid": "7e72083e960e4badf8e4c46f5fd534bd8cee5221"
},
"source": [
"The CategoricalEncoder expects a 2D array, containing one or more categorical input features. We need to reshape housing_cat to a 2D array:"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {
"_cell_guid": "b63a9d4f-65fc-46bc-b274-19308c2d4e61",
"_uuid": "3fd292830950fbd47ad6cc15d0c63d597b332130"
},
"outputs": [
{
"data": {
"text/plain": [
"<16512x5 sparse matrix of type '<class 'numpy.float64'>'\n",
"\twith 16512 stored elements in Compressed Sparse Row format>"
]
},
"execution_count": 33,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"cat_encoder = CategoricalEncoder()\n",
"housing_cat_reshaped = housing_cat.values.reshape(-1, 1)\n",
"housing_cat_1hot = cat_encoder.fit_transform(housing_cat_reshaped)\n",
"housing_cat_1hot"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "7c888e51-b4ad-49f7-afdf-a25843a46d3f",
"_uuid": "0f907c918b612e7c7ed437ef448ecb89fc91ac88"
},
"source": [
"We use toarray() again to get a dense array:"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {
"_cell_guid": "8eff92bc-7253-418a-90dd-e4d20e678c64",
"_uuid": "9935fd3baab104cb5d881a44a0ccc8caa09499eb"
},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 1., 0., 0., 0., 0.],\n",
" [ 1., 0., 0., 0., 0.],\n",
" [ 0., 0., 0., 0., 1.],\n",
" ..., \n",
" [ 0., 1., 0., 0., 0.],\n",
" [ 1., 0., 0., 0., 0.],\n",
" [ 0., 0., 0., 1., 0.]])"
]
},
"execution_count": 34,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_cat_1hot.toarray()"
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {
"_cell_guid": "5076a2e0-b19b-4654-bc2e-3bbce01cbeb0",
"_uuid": "aabe7ef4437f97c88b127ec170e3ace239aa7dd8"
},
"outputs": [
{
"data": {
"text/plain": [
"[array(['<1H OCEAN', 'INLAND', 'ISLAND', 'NEAR BAY', 'NEAR OCEAN'], dtype=object)]"
]
},
"execution_count": 35,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"cat_encoder.categories_"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='custom'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "93539dfc-1f69-43ae-a337-2d6f29dc78d5",
"_uuid": "a3018b73f1e63b090cda2dba0165e7fa1ea77bd4"
},
"source": [
"### Custom transfer\n",
"\n",
"Sklearn provides us with many different useful built in functions that help us to preprocess data and a lot of other stuff. But you will need to write your own for tasks such as custom cleanup operations or combining specific attributes. You want that the functions you write are working flawless togehter with sklearn functionalities like pipelines for example. Since Sklearn relies on [duck typing](http://davidasboth.com/2016/11/14/duck-typing/), all you need to do is to create a class and implement the three following mehtods: fit(), transform() and fit_transform(). You get the last one simply by adding **TransformerMixin** as a base class. And if you add **BaseEstimator** as a base class, and avoid \"*args\" and \"*kargs\" in your code, you will get two extra methods (get_params() and set_params()), which will we useful to automatically tune your hyperparameters. To illustrate this, I wrote a small transformer class below that adds the combined attributes:"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {
"_cell_guid": "e8b7af42-9e7c-4440-bb45-975a918cfaba",
"_uuid": "4857801d503f632f6a51d1a08223df4aeed8eea9",
"collapsed": true
},
"outputs": [],
"source": [
"from sklearn.base import BaseEstimator, TransformerMixin\n",
"\n",
"rooms_ix, bedrooms_ix, population_ix, household_ix = 3, 4, 5, 6\n",
"\n",
"class CombinedAttributesAdder(BaseEstimator, TransformerMixin):\n",
" def __init__(self, add_bedrooms_per_room = True): \n",
" self.add_bedrooms_per_room = add_bedrooms_per_room\n",
" def fit(self, X, y=None):\n",
" return self \n",
" def transform(self, X, y=None):\n",
" rooms_per_household = X[:, rooms_ix] / X[:, household_ix]\n",
" population_per_household = X[:, population_ix] / X[:, household_ix]\n",
" if self.add_bedrooms_per_room:\n",
" bedrooms_per_room = X[:, bedrooms_ix] / X[:, rooms_ix]\n",
" return np.c_[X, rooms_per_household, population_per_household,\n",
" bedrooms_per_room]\n",
" else:\n",
" return np.c_[X, rooms_per_household, population_per_household]\n",
"\n",
"attr_adder = CombinedAttributesAdder(add_bedrooms_per_room=False)\n",
"housing_extra_attribs = attr_adder.transform(housing.values)"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {
"_cell_guid": "f8c9148c-e699-4c6c-b2da-2f83f92f53e3",
"_uuid": "41185ec11a15179e37cc6763712efea7537c0ee9"
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>longitude</th>\n",
" <th>latitude</th>\n",
" <th>housing_median_age</th>\n",
" <th>total_rooms</th>\n",
" <th>total_bedrooms</th>\n",
" <th>population</th>\n",
" <th>households</th>\n",
" <th>median_income</th>\n",
" <th>ocean_proximity</th>\n",
" <th>income_categories</th>\n",
" <th>rooms_per_household</th>\n",
" <th>population_per_household</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>-121.89</td>\n",
" <td>37.29</td>\n",
" <td>38</td>\n",
" <td>1568</td>\n",
" <td>351</td>\n",
" <td>710</td>\n",
" <td>339</td>\n",
" <td>2.7042</td>\n",
" <td>&lt;1H OCEAN</td>\n",
" <td>2</td>\n",
" <td>4.62537</td>\n",
" <td>2.0944</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>-121.93</td>\n",
" <td>37.05</td>\n",
" <td>14</td>\n",
" <td>679</td>\n",
" <td>108</td>\n",
" <td>306</td>\n",
" <td>113</td>\n",
" <td>6.4214</td>\n",
" <td>&lt;1H OCEAN</td>\n",
" <td>5</td>\n",
" <td>6.00885</td>\n",
" <td>2.70796</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>-117.2</td>\n",
" <td>32.77</td>\n",
" <td>31</td>\n",
" <td>1952</td>\n",
" <td>471</td>\n",
" <td>936</td>\n",
" <td>462</td>\n",
" <td>2.8621</td>\n",
" <td>NEAR OCEAN</td>\n",
" <td>2</td>\n",
" <td>4.22511</td>\n",
" <td>2.02597</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>-119.61</td>\n",
" <td>36.31</td>\n",
" <td>25</td>\n",
" <td>1847</td>\n",
" <td>371</td>\n",
" <td>1460</td>\n",
" <td>353</td>\n",
" <td>1.8839</td>\n",
" <td>INLAND</td>\n",
" <td>2</td>\n",
" <td>5.23229</td>\n",
" <td>4.13598</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>-118.59</td>\n",
" <td>34.23</td>\n",
" <td>17</td>\n",
" <td>6592</td>\n",
" <td>1525</td>\n",
" <td>4459</td>\n",
" <td>1463</td>\n",
" <td>3.0347</td>\n",
" <td>&lt;1H OCEAN</td>\n",
" <td>3</td>\n",
" <td>4.50581</td>\n",
" <td>3.04785</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" longitude latitude housing_median_age total_rooms total_bedrooms population \\\n",
"0 -121.89 37.29 38 1568 351 710 \n",
"1 -121.93 37.05 14 679 108 306 \n",
"2 -117.2 32.77 31 1952 471 936 \n",
"3 -119.61 36.31 25 1847 371 1460 \n",
"4 -118.59 34.23 17 6592 1525 4459 \n",
"\n",
" households median_income ocean_proximity income_categories \\\n",
"0 339 2.7042 <1H OCEAN 2 \n",
"1 113 6.4214 <1H OCEAN 5 \n",
"2 462 2.8621 NEAR OCEAN 2 \n",
"3 353 1.8839 INLAND 2 \n",
"4 1463 3.0347 <1H OCEAN 3 \n",
"\n",
" rooms_per_household population_per_household \n",
"0 4.62537 2.0944 \n",
"1 6.00885 2.70796 \n",
"2 4.22511 2.02597 \n",
"3 5.23229 4.13598 \n",
"4 4.50581 3.04785 "
]
},
"execution_count": 37,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_extra_attribs = pd.DataFrame(housing_extra_attribs, columns=list(housing.columns)+[\"rooms_per_household\", \"population_per_household\"])\n",
"housing_extra_attribs.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "9ded6ce5-818a-4ff5-9f8f-512afa067c88",
"_uuid": "c42a4d4390d49f1ca50a4797f91f19b3bc11fa79"
},
"source": [
"Here the transformer has one hyperparameter, which is set to true by default and allows you to easily find out whether adding this attribute helps the machine learning model or not. You can add any hyperparameter to gate any data preparation step you are not sure about. The more you automate the data prepration steps, the more combinations you can try out, which therefore makes it much more likely that you find a great one and save a lot of time."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='scale'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "32e6535f-2445-4478-b5f1-64512ae9397e",
"_uuid": "0e99fdfc72779ce61fd20e80a1a7ff40b564dc4c"
},
"source": [
"### Feature Scaling\n",
"\n",
"Feature scaling is one of the most important transformations you need to apply, since nearly all machine learning algorithms perform bad when the input numerical attributes have widely varying scales, which is the case at our current dataset. For example, the median incomes range from o to 15, but the total number of rooms from 6 to 39,320. Note that scaling the target values is not required.\n",
"\n",
"There are two common ways:\n",
"\n",
"**1.) min-max scaling**\n",
"\n",
"**2.) standardization**\n",
"\n",
"With min-max sclaing the values are shifted and rescaled that they range from 0 to 1. This can be done by subtracting the **min value** and dividing by the **max** minus the **min**. Sklearn's transformer called **MinMaxScaler**, which has a feature_range hyperparameter, that let's you change the range to whatever you want to (instead of 0 to 1), can do this for you. \n",
"\n",
"Standardization subtracts the mean value and divides it by the variance. It does not bound the values into a specific range, which can be good or bad. Neural Networks for example often need input values that range from 0 to1. The good thing is, that standardization is less effected by outliers. If for example, you suppose a district has a median income of 100 (by mistake), min-max scaling would would map the values that range from 0 to 15 down to 0 to 0.15, whereas standardization would give you a much more accurate outcome. With sklearn you can do standardization with the **StandardScaler** tramsformer. And always just do transformations with the training set only, to prevent overfitting. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='pipe'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "d13eee6b-d0c5-4f02-a935-9570eebc4e19",
"_uuid": "2f49c34d87f5e7f3b351dd41041a3ff2e7c4283e"
},
"source": [
"### Pipelines for tranformation\n",
"\n",
"There are a lot of data tranformation steps that also need to be done in the right order. To help you with that, sklearn provides the **Pipeline** class. In the code below, you can see a small pipeline of attributes:"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {
"_cell_guid": "22d3b112-d96b-4cca-8e06-2467a1b75ed5",
"_uuid": "583d9f30e29693244e74f37564f7ccad5ec7413a",
"collapsed": true
},
"outputs": [],
"source": [
"from sklearn.pipeline import Pipeline\n",
"from sklearn.preprocessing import StandardScaler\n",
"\n",
"num_pipeline = Pipeline([\n",
" ('imputer', Imputer(strategy=\"median\")),\n",
" ('attribs_adder', CombinedAttributesAdder()),\n",
" ('std_scaler', StandardScaler()),\n",
" ])\n",
"\n",
"housing_num_tr = num_pipeline.fit_transform(housing_num)"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {
"_cell_guid": "42977775-382e-491f-924c-b43ad6a3355b",
"_uuid": "0f4ca41f8eefb6457945bc4f1a3dc555218da03b"
},
"outputs": [
{
"data": {
"text/plain": [
"array([[-1.15604281, 0.77194962, 0.74333089, ..., -0.31205452,\n",
" -0.08649871, 0.15531753],\n",
" [-1.17602483, 0.6596948 , -1.1653172 , ..., 0.21768338,\n",
" -0.03353391, -0.83628902],\n",
" [ 1.18684903, -1.34218285, 0.18664186, ..., -0.46531516,\n",
" -0.09240499, 0.4222004 ],\n",
" ..., \n",
" [ 1.58648943, -0.72478134, -1.56295222, ..., 0.3469342 ,\n",
" -0.03055414, -0.52177644],\n",
" [ 0.78221312, -0.85106801, 0.18664186, ..., 0.02499488,\n",
" 0.06150916, -0.30340741],\n",
" [-1.43579109, 0.99645926, 1.85670895, ..., -0.22852947,\n",
" -0.09586294, 0.10180567]])"
]
},
"execution_count": 39,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_num_tr"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "516966e7-d290-4308-936f-3f0e2040475e",
"_uuid": "721f45e58e2d10455574af4d9549baba7cdf11bb"
},
"source": [
"The Pipeline constructor takes a list of name/estimator pairs, which define a sequence of steps. Note that all instead of the last estimator must be transformers, which means that they must have a fit_transform() method. Also note that if you call the pipelines fit() method, it calls fit_transform() on all transformers. In our example, the last estimator is a StandardScaler (for standardization), so the pipeline has a transform() method that applies all the transform to the data in sequence. \n",
"\n",
"Now it would be nice if we could feed a Pandas DataFrame directly into our pipeline, instead of having to first manually extract the numerical columns into a NumPy array. Unfortunately sklearn provides nothing that does this for us, so we have to write it by ourself:"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {
"_cell_guid": "c50fc2f1-ab30-4f0f-8e1f-6de508dc88c7",
"_uuid": "5da733d528a5cef71d62479043321bf060d5904b",
"collapsed": true
},
"outputs": [],
"source": [
"from sklearn.base import BaseEstimator, TransformerMixin\n",
"\n",
"class DataFrameSelector(BaseEstimator, TransformerMixin):\n",
" def __init__(self, attribute_names):\n",
" self.attribute_names = attribute_names\n",
" def fit(self, X, y=None):\n",
" return self\n",
" def transform(self, X):\n",
" return X[self.attribute_names].values"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "3459f693-b61e-4872-89a3-16b3d9437e1a",
"_uuid": "b1ace1f620365a8b9b89284f8725ea7b8299c575"
},
"source": [
"This will tranform our data by selecting the desired attributes, dropping the others and converting the resulting dataframe into a NumPy array. With this in mind you can easily write a pipeline that takes a Pandas dataframe and handle only the numerical values. Just start the pipeline with a **DataFrameSelector** to pick only the numerical attributes. And you can just as easily write another one for the categorical attributes, by simply selecting categorical attributes using a **DataFrameSelector** and then applying a **LabelBinarizer**."
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {
"_cell_guid": "55f536e3-8e73-41de-a1f4-e8ac1aae0e04",
"_uuid": "da1d634421e100d5f26cff578a71102da279e645",
"collapsed": true
},
"outputs": [],
"source": [
"num_attribs = list(housing_num)\n",
"cat_attribs = [\"ocean_proximity\"]\n",
"\n",
"num_pipeline = Pipeline([\n",
" ('selector', DataFrameSelector(num_attribs)),\n",
" ('imputer', Imputer(strategy=\"median\")),\n",
" ('attribs_adder', CombinedAttributesAdder()),\n",
" ('std_scaler', StandardScaler()),\n",
" ])\n",
"\n",
"cat_pipeline = Pipeline([\n",
" ('selector', DataFrameSelector(cat_attribs)),\n",
" ('cat_encoder', CategoricalEncoder(encoding=\"onehot-dense\")),\n",
" ])"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "46a11875-7c21-4f69-8915-513d753c6522",
"_uuid": "7a14b2a724aa5af46558004cfebcc85fe192257e"
},
"source": [
"But there is also a way to join these two pipelines into a single pipeline, using sklearns **FeatureUnion**. You can give it a list of transformer-pipelines. A complete pipeline that handles numerical and categorical attributes looks like this:"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {
"_cell_guid": "cee7c2f1-38dc-4e6f-a527-43a8a5e0c5c4",
"_uuid": "7ca5540c4d8f862b8d91312e4f1b107c03430b53",
"collapsed": true
},
"outputs": [],
"source": [
"from sklearn.pipeline import FeatureUnion\n",
"\n",
"full_pipeline = FeatureUnion(transformer_list=[\n",
" (\"num_pipeline\", num_pipeline),\n",
" (\"cat_pipeline\", cat_pipeline),\n",
" ])"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "25fd1373-968d-408a-97e3-fb7a9fffaffd",
"_uuid": "4b4b0fa9f9100530af1323c90d815e4ee1d4f375"
},
"source": [
"The code below runs the whole pipeline:"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {
"_cell_guid": "0078dc3a-82dc-4548-8d5d-0a4a94a5bba9",
"_uuid": "a8a6ca6f2ee6fe8ffc473f5ea80226c6f6513514"
},
"outputs": [
{
"data": {
"text/plain": [
"array([[-1.15604281, 0.77194962, 0.74333089, ..., 0. ,\n",
" 0. , 0. ],\n",
" [-1.17602483, 0.6596948 , -1.1653172 , ..., 0. ,\n",
" 0. , 0. ],\n",
" [ 1.18684903, -1.34218285, 0.18664186, ..., 0. ,\n",
" 0. , 1. ],\n",
" ..., \n",
" [ 1.58648943, -0.72478134, -1.56295222, ..., 0. ,\n",
" 0. , 0. ],\n",
" [ 0.78221312, -0.85106801, 0.18664186, ..., 0. ,\n",
" 0. , 0. ],\n",
" [-1.43579109, 0.99645926, 1.85670895, ..., 0. ,\n",
" 1. , 0. ]])"
]
},
"execution_count": 43,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_prepared = full_pipeline.fit_transform(housing)\n",
"housing_prepared"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "a8354675-fff1-4e54-997c-ffc83ed6c15d",
"_uuid": "45afd75232073cfd44cbb9947e6c0577217a0eac"
},
"source": [
"<a id='train'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "832d8138-38a0-4289-900b-68e6f0323308",
"_uuid": "291cce033be3d5c2f9a56648fb6a97139f6072ac"
},
"source": [
"# Train Models\n",
"\n",
"We looked at the big picture of your problem and framed it, we explored and visualized the data and preprocessed it. Now it is time to select and train a Machine Learning model. The hard part were the previous steps. What we do now, is going to be much simpler and easier.\n",
"\n",
"First, let's test whether a Linear Regression model gives us a satisfying result:"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {
"_cell_guid": "45e9e929-2891-4eb3-820b-af5097baaa5e",
"_uuid": "89d71dd1adb221b3b641a88b722d3ab8b3edd1db"
},
"outputs": [
{
"data": {
"text/plain": [
"LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)"
]
},
"execution_count": 44,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.linear_model import LinearRegression\n",
"\n",
"lin_reg = LinearRegression()\n",
"lin_reg.fit(housing_prepared, housing_labels)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "446a26cf-63ec-4a59-98b4-52f9f7e30ca1",
"_uuid": "0cda946bd4d1ae544a875f46de6cb3b981c651be"
},
"source": [
"Let's try it out on the training set:"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {
"_cell_guid": "55302166-c96f-4ac6-90fc-837e8022d433",
"_uuid": "73d4afe9ec9801303c2074838e9bdb98b33807d5"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Predictions: [ 203682.37379543 326371.39370781 204218.64588245 58685.4770482\n",
" 194213.06443039]\n"
]
}
],
"source": [
"some_data = housing.iloc[:5]\n",
"some_labels = housing_labels.iloc[:5]\n",
"some_data_prepared = full_pipeline.transform(some_data)\n",
"\n",
"print(\"Predictions:\", lin_reg.predict(some_data_prepared))"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {
"_cell_guid": "c9e63f21-1ec7-43e7-97fb-f6677f6958e4",
"_uuid": "a2e27e117fd8b35ff367410bcfbbb395dc6c01c8"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Labels: [286600.0, 340600.0, 196900.0, 46300.0, 254500.0]\n"
]
}
],
"source": [
"print(\"Labels:\", list(some_labels))"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "1871b2e3-59cd-441c-b84b-56cf826c3d70",
"_uuid": "585c5b83d4060e3f43e8d9369f52d737667ee758"
},
"source": [
"Let's use [RMSE (Root Mean Squared Error)](http://www.statisticshowto.com/rmse/) to judge the quality of our predictions:"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {
"_cell_guid": "af680258-93ae-4463-92c1-a6d5a5417ea2",
"_uuid": "295d543d6a5027fc050567dafd08b8e1dda0bab7"
},
"outputs": [
{
"data": {
"text/plain": [
"68376.642954599374"
]
},
"execution_count": 47,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.metrics import mean_squared_error\n",
"\n",
"housing_predictions = lin_reg.predict(housing_prepared)\n",
"lin_mse = mean_squared_error(housing_labels, housing_predictions)\n",
"lin_rmse = np.sqrt(lin_mse)\n",
"lin_rmse"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "5dada588-2f5e-413e-91f3-9c26a55a9823",
"_uuid": "e01e74ae47d5f458e3638d72b60b03679b09219c"
},
"source": [
"This is clearly not a great score. Since most districts median_housing_values range between 120,000 and 265,000 dollar, a prediction erro of $68,376 is not very satisfying and also an example of a model underfitting the data. This either means that the features do not provide enough information to make proper predictions, or that the model is just not powerful enough.\n",
"\n",
"\n",
"The main ways to fix underfitting are:\n",
"\n",
"1.) feed the model with better features\n",
"\n",
"2.) select a more powerful model\n",
"\n",
"3.) reduce the constraints on the model\n",
"\n",
"\n",
"First let's try out a more powerful model since we just only tested one."
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "35d4334a-ec83-42fd-ad13-53bbc32dbf50",
"_uuid": "c1b6320853f924538978e88895d1ce9e5ad8ece1"
},
"source": [
"Let's use a DecisionTreeRegressor, which can find complex nonlinear relationships in the data:"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {
"_cell_guid": "a964cbab-baa4-4878-9e1f-8222d3bc8fde",
"_uuid": "9146b7f3909466b8008718b7687ed98971d35218"
},
"outputs": [
{
"data": {
"text/plain": [
"DecisionTreeRegressor(criterion='mse', max_depth=None, max_features=None,\n",
" max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
" min_impurity_split=None, min_samples_leaf=1,\n",
" min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
" presort=False, random_state=42, splitter='best')"
]
},
"execution_count": 48,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.tree import DecisionTreeRegressor\n",
"\n",
"tree_reg = DecisionTreeRegressor(random_state=42)\n",
"tree_reg.fit(housing_prepared, housing_labels)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "973b0725-cce5-43e6-8be7-2d5a24aa1d26",
"_uuid": "de4510ba8497a6e968d847c45641e2665fa4c932"
},
"source": [
"Let's try it out on the training set:"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {
"_cell_guid": "4ea7c153-689b-451e-9a16-0e7ebd37e146",
"_uuid": "e9206d446ceef3a47464531fb5be3b0c3c73c744"
},
"outputs": [
{
"data": {
"text/plain": [
"0.0"
]
},
"execution_count": 49,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_predictions = tree_reg.predict(housing_prepared)\n",
"tree_mse = mean_squared_error(housing_labels, housing_predictions)\n",
"tree_rmse = np.sqrt(tree_mse)\n",
"tree_rmse"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "b01f8b5d-7346-40c4-80ec-d027a20a4df1",
"_uuid": "6ce1c963be18b578b1bd6b5118b78033285fdf30"
},
"source": [
"This gives you no error at all, which means that we strongly overfitted our data. How can we be sure ? As we allready discussed earlier, you don't want to use the test set until you are confident about your model. But how can we test how our model performs if we can't use the test data ? One way to do this is using **K-Fold Cross-Validation**, which uses part of the training set for training and a part for validation. The following code randomly splits the training set into 10 subset called **folds**. Then it trains and evaluates 10 times, using every fold once for either training or validation:"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {
"_cell_guid": "09b2032d-162e-4e34-8d15-161023d593c6",
"_uuid": "2c56d86357bf1e0734605c41d7feaf426e420f72",
"collapsed": true
},
"outputs": [],
"source": [
"from sklearn.model_selection import cross_val_score\n",
"\n",
"scores = cross_val_score(tree_reg, housing_prepared, housing_labels,\n",
" scoring=\"neg_mean_squared_error\", cv=10)\n",
"tree_rmse_scores = np.sqrt(-scores)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "17da0dc6-764b-4a17-82b2-450851dc62f0",
"_uuid": "f52859334863993e13a5e837d3b83da5ddcecbc2"
},
"source": [
"Let's look at the result:"
]
},
{
"cell_type": "code",
"execution_count": 51,
"metadata": {
"_cell_guid": "5e8b2180-d92b-4b45-b684-87dd52ca99b8",
"_uuid": "dda1a1dac00f76e624d06d24a9371264a61b69ce"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Scores: [ 70307.54079165 67270.94561829 71361.89886493 68889.91784739\n",
" 70992.94150631 74140.66907294 70607.72083897 70576.13836762\n",
" 76433.16342183 70270.96266671]\n",
"Mean: 71085.1898997\n",
"Standard deviation: 2431.96102969\n"
]
}
],
"source": [
"def display_scores(scores):\n",
" print(\"Scores:\", scores)\n",
" print(\"Mean:\", scores.mean())\n",
" print(\"Standard deviation:\", scores.std())\n",
"\n",
"display_scores(tree_rmse_scores)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "3efada63-547d-4713-84ca-e34a90aa53b6",
"_uuid": "4667df44f910c4e6c59e00aac96a47ee6fc47c1d"
},
"source": [
"Now we have a real estimate of how the decision tree performs. In fact, it does even perform worse than our previous linear regression model.\n",
"Note that cross validation not only gives you an estimate of your models performance, but also of how precise this estimate is (standard deviation). Here we have a prediction error of 71,085 dollar. The standard deviation is 2,431 Dollar, which means that our prediction error could be $2,431 more or less.\n",
"\n",
"Now let's compute the scores for our linear regression model again, but using K-Fold Cross-Validation:"
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {
"_cell_guid": "90ff6617-e614-4049-8509-80880478f1b0",
"_uuid": "d608fb205e09c007c6437f14274c891acf8bff2b"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Scores: [ 66877.52325028 66608.120256 70575.91118868 74179.94799352\n",
" 67683.32205678 71103.16843468 64782.65896552 67711.29940352\n",
" 71080.40484136 67687.6384546 ]\n",
"Mean: 68828.9994845\n",
"Standard deviation: 2662.76157061\n"
]
}
],
"source": [
"lin_scores = cross_val_score(lin_reg, housing_prepared, housing_labels,\n",
" scoring=\"neg_mean_squared_error\", cv=10)\n",
"lin_rmse_scores = np.sqrt(-lin_scores)\n",
"display_scores(lin_rmse_scores)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "8929efac-fcfb-4b4e-8784-22f60ccf8627",
"_uuid": "c7f97ea61372686c35a978fc0faaf7ba7eaff028"
},
"source": [
"Now we defenitely know that our decision tree model is overfitting and performs worse than our linear regression model. Now let's try one last model, the RandomForestRegressor. Random Forest works by training many Decision Trees on random subsets of the features and then it averages their predictions. Building a model on top of other models is called **Essemble Learning** and often pushes Machine Learning algortihms further."
]
},
{
"cell_type": "code",
"execution_count": 53,
"metadata": {
"_cell_guid": "19058264-5b31-44eb-b626-6cbc180e907b",
"_uuid": "795c25750fb3a77ada89b37f01497bac425c4db3"
},
"outputs": [
{
"data": {
"text/plain": [
"RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
" max_features='auto', max_leaf_nodes=None,\n",
" min_impurity_decrease=0.0, min_impurity_split=None,\n",
" min_samples_leaf=1, min_samples_split=2,\n",
" min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,\n",
" oob_score=False, random_state=42, verbose=0, warm_start=False)"
]
},
"execution_count": 53,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.ensemble import RandomForestRegressor\n",
"\n",
"forest_reg = RandomForestRegressor(random_state=42)\n",
"forest_reg.fit(housing_prepared, housing_labels)"
]
},
{
"cell_type": "code",
"execution_count": 54,
"metadata": {
"_cell_guid": "c1dc128f-706a-4f38-a4ae-d09f7b05c6c4",
"_uuid": "b9ce122250dd136b035b3ca743d068f4ca2861e7"
},
"outputs": [
{
"data": {
"text/plain": [
"22112.540875989125"
]
},
"execution_count": 54,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"housing_predictions = forest_reg.predict(housing_prepared)\n",
"forest_mse = mean_squared_error(housing_labels, housing_predictions)\n",
"forest_rmse = np.sqrt(forest_mse)\n",
"forest_rmse"
]
},
{
"cell_type": "code",
"execution_count": 55,
"metadata": {
"_cell_guid": "19a5dd39-e489-4ca7-88f2-6f3c168d447d",
"_uuid": "6c8d83fe4836948c129cc8ea04f5c0cf54675e07"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Scores: [ 51513.5500248 48821.99364978 53451.63889205 54910.55855153\n",
" 50405.68166817 56772.90691171 51909.21824137 50516.14893695\n",
" 55699.80788403 53107.7805515 ]\n",
"Mean: 52710.9285312\n",
"Standard deviation: 2414.72717912\n"
]
}
],
"source": [
"forest_scores = cross_val_score(forest_reg, housing_prepared, housing_labels,\n",
" scoring=\"neg_mean_squared_error\", cv=10)\n",
"forest_rmse_scores = np.sqrt(-forest_scores)\n",
"display_scores(forest_rmse_scores)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "f642f6f6-c0e8-4491-81cc-3c4c46346fef",
"_uuid": "f40651bacad031fbefde0736d5e6f211578d8bd3"
},
"source": [
"This is a lot better. Note that the score on the training set is still much lower, than on the validation set, which indicates that the model is overfitting the training set and that we should optimize the model to solve this problem. However, you shouldn't fine tune a model too much before you haven't tried out various other machine learning models from different categories of algorithms, without spending to much time tweaking it. You could use different Support Vector Machines with different Kernels, maybe a neural network and so on. The goal is that you have a rough estimate about, what algorithms performs the best and then you should spent time fine tuning them for a better accuracy. Note that you should save every model you tested using pythons [pickle](https://wiki.python.org/moin/UsingPickle) module or sklearn [joblib](http://scikit-learn.org/stable/modules/model_persistence.html)."
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "b954133c-f963-49be-908c-f32ec6470d10",
"_uuid": "c0e7c69fad1c82da547d467824406c05ed286b99"
},
"source": [
"<a id='tune'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "ad5b5c4c-2693-483b-bb7d-97aa1f5efacc",
"_uuid": "3a609c52cbbac2c7f98762f90077bdba7a759e3a"
},
"source": [
"# Fine Tuning\n",
"\n",
"Now let's assume you played around with different algorithms and have a rough estimate over the best performing ones. We will now look at Grid Search as an example to fine tune your model.\n",
"\n",
"\n",
"### Grid Search\n",
"Instead of trying out different hyperparametrs manually until you find some great combinations, you can let Grid Search do that for you. All you need to do is tell it which hyperparameters you want to experiment with and what values. The code below searches for the best combination of hyperparameter values for the RandomForestRegressor: "
]
},
{
"cell_type": "code",
"execution_count": 56,
"metadata": {
"_cell_guid": "86f21756-80c7-4287-ace3-093b46a14867",
"_uuid": "e5fc2ff148232897dbb4b00e87e526ad5ac47bae"
},
"outputs": [
{
"data": {
"text/plain": [
"GridSearchCV(cv=5, error_score='raise',\n",
" estimator=RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
" max_features='auto', max_leaf_nodes=None,\n",
" min_impurity_decrease=0.0, min_impurity_split=None,\n",
" min_samples_leaf=1, min_samples_split=2,\n",
" min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,\n",
" oob_score=False, random_state=42, verbose=0, warm_start=False),\n",
" fit_params=None, iid=True, n_jobs=1,\n",
" param_grid=[{'n_estimators': [3, 10, 30], 'max_features': [2, 4, 6, 8]}, {'bootstrap': [False], 'n_estimators': [3, 10], 'max_features': [2, 3, 4]}],\n",
" pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',\n",
" scoring='neg_mean_squared_error', verbose=0)"
]
},
"execution_count": 56,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.model_selection import GridSearchCV\n",
"\n",
"param_grid = [\n",
" {'n_estimators': [3, 10, 30], 'max_features': [2, 4, 6, 8]},\n",
" {'bootstrap': [False], 'n_estimators': [3, 10], 'max_features': [2, 3, 4]},\n",
" ]\n",
"\n",
"forest_reg = RandomForestRegressor(random_state=42)\n",
"grid_search = GridSearchCV(forest_reg, param_grid, cv=5,\n",
" scoring='neg_mean_squared_error')\n",
"grid_search.fit(housing_prepared, housing_labels)"
]
},
{
"cell_type": "code",
"execution_count": 57,
"metadata": {
"_cell_guid": "6d0b2082-3813-4ade-8ed6-517b07349e62",
"_uuid": "a8b09592912609ceaf7db150caef258eb89c7b7b"
},
"outputs": [
{
"data": {
"text/plain": [
"{'max_features': 6, 'n_estimators': 30}"
]
},
"execution_count": 57,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"grid_search.best_params_"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "282c3026-b195-4791-b0f8-19c0752a53d7",
"_uuid": "b7cb87b4bf6a7e476a37f35fbcba8cdb7d1dd667"
},
"source": [
"This tells us that the best solution would be by setting the max_features to 8 and the n_estimators to 30."
]
},
{
"cell_type": "code",
"execution_count": 58,
"metadata": {
"_cell_guid": "8204b052-a95b-4488-aeb3-2e24d2a2125b",
"_uuid": "5446afb99fc2c7aab51253832dc6a887db6bbd4e"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"64265.8521009 {'max_features': 2, 'n_estimators': 3}\n",
"55882.9848402 {'max_features': 2, 'n_estimators': 10}\n",
"53472.5297705 {'max_features': 2, 'n_estimators': 30}\n",
"61320.6172102 {'max_features': 4, 'n_estimators': 3}\n",
"53834.6661703 {'max_features': 4, 'n_estimators': 10}\n",
"51273.2598733 {'max_features': 4, 'n_estimators': 30}\n",
"59851.1600773 {'max_features': 6, 'n_estimators': 3}\n",
"53108.4926792 {'max_features': 6, 'n_estimators': 10}\n",
"50804.4906775 {'max_features': 6, 'n_estimators': 30}\n",
"59225.2197785 {'max_features': 8, 'n_estimators': 3}\n",
"52883.782581 {'max_features': 8, 'n_estimators': 10}\n",
"50942.1591379 {'max_features': 8, 'n_estimators': 30}\n",
"62801.357247 {'bootstrap': False, 'max_features': 2, 'n_estimators': 3}\n",
"54452.7056215 {'bootstrap': False, 'max_features': 2, 'n_estimators': 10}\n",
"61122.9494918 {'bootstrap': False, 'max_features': 3, 'n_estimators': 3}\n",
"53014.3329645 {'bootstrap': False, 'max_features': 3, 'n_estimators': 10}\n",
"60252.6537668 {'bootstrap': False, 'max_features': 4, 'n_estimators': 3}\n",
"52716.5505964 {'bootstrap': False, 'max_features': 4, 'n_estimators': 10}\n"
]
}
],
"source": [
"cvres = grid_search.cv_results_\n",
"for mean_score, params in zip(cvres[\"mean_test_score\"], cvres[\"params\"]):\n",
" print(np.sqrt(-mean_score), params)"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "a8f530b5-53af-45e6-a8cd-243e3232002f",
"_uuid": "9f467d653b4eae6b5916e57aa96ef666588e1ea5"
},
"source": [
"The RMSE score for this combinationis 50,804. Now you have succesfully fine tuned your model ! Alternatives to Grid Search would be [Randomized Search](http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RandomizedSearchCV.html) (good for exploring many hyperparamter combinations) and [Ensemble Methods](https://www.toptal.com/machine-learning/ensemble-methods-machine-learning). "
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "09641c86-26a5-4bdd-8e0c-adf3f7988ed4",
"_uuid": "d8126e40e896953613eeb434b9d6c097823cc70d"
},
"source": [
"<a id='evaluation'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "dbc1f83f-d67e-4ffb-aade-789de04593bd",
"_uuid": "37b1e04219808f343f0d889fcb1a29c2d706df96"
},
"source": [
"# Evaluation\n",
"\n",
"Now we can evaluate the final model on the test set. You just have to get the labels from your test set, run your full_pipeline to transform the data and eveluate the model:"
]
},
{
"cell_type": "code",
"execution_count": 59,
"metadata": {
"_cell_guid": "a1ab9cbc-b6f5-4843-a31a-4d08b1902dd2",
"_uuid": "e747dc4169c852cd315f8495ccf69f728fd553b3",
"collapsed": true
},
"outputs": [],
"source": [
"final_model = grid_search.best_estimator_\n",
"\n",
"X_test = strat_test_set.drop(\"median_house_value\", axis=1)\n",
"y_test = strat_test_set[\"median_house_value\"].copy()\n",
"\n",
"X_test_prepared = full_pipeline.transform(X_test)\n",
"final_predictions = final_model.predict(X_test_prepared)\n",
"\n",
"final_mse = mean_squared_error(y_test, final_predictions)\n",
"final_rmse = np.sqrt(final_mse)"
]
},
{
"cell_type": "code",
"execution_count": 60,
"metadata": {
"_cell_guid": "d00aced3-0abe-4f4c-aede-0d7cb9471ee9",
"_uuid": "caca5a23be93bfc7f57cf519660e640401590bf2"
},
"outputs": [
{
"data": {
"text/plain": [
"49036.586427241818"
]
},
"execution_count": 60,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"final_rmse"
]
},
{
"cell_type": "markdown",
"metadata": {
"_cell_guid": "1aa96c68-538c-4f9b-a3ee-cd19dc56bb49",
"_uuid": "0b01cbed6b058c05cadf6d045e6819ab02f36a1b"
},
"source": [
"We now have a final prediciton error of $49,036 and learned a lot of things on the way. Note that usually the performance on the test set is slightly worse, because your system is fine tuned to perform well on the training set. \n",
"\n",
"Hopefully all of this gave you a good idea of what a machine learning project looks like and a blueprint for you to build great systems in the future. As I already mentioned, most of the work has to be done before you select and train an algorithm. Note that it is propably better to know four or five machine learning algorithms very well rather than spending all your time exploring advanced algorithms and not enough time on the overall process."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id='summary'></a>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Summary\n",
"\n",
"Now we have a pretty good understanding of how supervised regression works and learned a massive amout of things on the way.\n",
"\n",
"We learned how to frame a problem, plot histograms & draw conclusions from it, split data into training & testing subsets and what tailheavyness is & how to solve it. \n",
"\n",
"We now know why stratified sampling can be very important and we used it on the income_categories feature, after we created it. We also plotted scatterplots, computed a correlation coefficient table, discussed scaled & capped features and even wrote functions for feature engineering. \n",
"\n",
"In terms of data preparation, we talked about missing values and filled some. We processed categorical attributes and text, using One hot encoding and reshaped arrays along the way. We did some feature scaling and learned that most machine learning algorithms perform bad, when the input numerical attributes have widely varying scales. \n",
"\n",
"Furthermore we used the sklearn pipeline class, joined two pipelines into a single one, used RMSE to evaluate our models and learned a lot about over- and underfitting. \n",
"\n",
"On top of that we used K-fold cross validation and tuned our hyperparameters with grid search. \n",
"\n",
"Along the project we used a Linear Regression model, a DecisionTreeRegressor and a RandomForestRegressor."
]
}
],
"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": "3.6.3"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment