Created
April 14, 2020 23:38
-
-
Save FavioVazquez/3640d62afb9b4a6481e42259a1f7d317 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| { | |
| "nbformat": 4, | |
| "nbformat_minor": 0, | |
| "metadata": { | |
| "colab": { | |
| "name": "Untitled1.ipynb", | |
| "provenance": [], | |
| "collapsed_sections": [] | |
| }, | |
| "kernelspec": { | |
| "name": "python3", | |
| "display_name": "Python 3" | |
| } | |
| }, | |
| "cells": [ | |
| { | |
| "cell_type": "code", | |
| "metadata": { | |
| "id": "FekgeDscCIlj", | |
| "colab_type": "code", | |
| "colab": {} | |
| }, | |
| "source": [ | |
| "import pandas as pd" | |
| ], | |
| "execution_count": 0, | |
| "outputs": [] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "metadata": { | |
| "id": "ARcdSkwfCJA6", | |
| "colab_type": "code", | |
| "colab": {} | |
| }, | |
| "source": [ | |
| "df = pd.read_csv(\"all_features.csv\")" | |
| ], | |
| "execution_count": 0, | |
| "outputs": [] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "metadata": { | |
| "id": "twd-n2SlCZvf", | |
| "colab_type": "code", | |
| "colab": { | |
| "base_uri": "https://localhost:8080/", | |
| "height": 1000 | |
| }, | |
| "outputId": "acf7c401-dc39-4e40-a7c0-d969bd0e8a2c" | |
| }, | |
| "source": [ | |
| "df.head()" | |
| ], | |
| "execution_count": 13, | |
| "outputs": [ | |
| { | |
| "output_type": "execute_result", | |
| "data": { | |
| "text/html": [ | |
| "<div>\n", | |
| "<style scoped>\n", | |
| " .dataframe tbody tr th:only-of-type {\n", | |
| " vertical-align: middle;\n", | |
| " }\n", | |
| "\n", | |
| " .dataframe tbody tr th {\n", | |
| " vertical-align: top;\n", | |
| " }\n", | |
| "\n", | |
| " .dataframe thead th {\n", | |
| " text-align: right;\n", | |
| " }\n", | |
| "</style>\n", | |
| "<table border=\"1\" class=\"dataframe\">\n", | |
| " <thead>\n", | |
| " <tr style=\"text-align: right;\">\n", | |
| " <th></th>\n", | |
| " <th>bedrooms</th>\n", | |
| " <th>bathrooms</th>\n", | |
| " <th>sqft_living</th>\n", | |
| " <th>sqft_lot</th>\n", | |
| " <th>waterfront</th>\n", | |
| " <th>view</th>\n", | |
| " <th>grade</th>\n", | |
| " <th>sqft_above</th>\n", | |
| " <th>sqft_basement</th>\n", | |
| " <th>yr_built</th>\n", | |
| " <th>yr_renovated</th>\n", | |
| " <th>zipcode</th>\n", | |
| " <th>lat</th>\n", | |
| " <th>long</th>\n", | |
| " <th>sqft_living15</th>\n", | |
| " <th>sqft_lot15</th>\n", | |
| " <th>KNeighbors(lat, long)</th>\n", | |
| " <th>sqft_living.1</th>\n", | |
| " <th>grade.1</th>\n", | |
| " <th>sqft_above.1</th>\n", | |
| " <th>sqft_living15.1</th>\n", | |
| " <th>bathrooms.1</th>\n", | |
| " <th>Number Of \"parking\" In 3.3952 KM Radius</th>\n", | |
| " <th>sqft_basement.1</th>\n", | |
| " <th>Number Of \"school\" In 3.0972 KM Radius</th>\n", | |
| " <th>Number Of \"post_box\" In 6.8997 KM Radius</th>\n", | |
| " <th>yr_renovated.1</th>\n", | |
| " <th>min(Average Total Market Value)</th>\n", | |
| " <th>sum(Average Total Market Value)</th>\n", | |
| " <th>max(Average Total Market Value)</th>\n", | |
| " <th>max(Average Tax Billed Amount)</th>\n", | |
| " <th>sum(Average Market Land Value)</th>\n", | |
| " <th>max(Average Market Land Value)</th>\n", | |
| " <th>sum(Average Tax Billed Amount)</th>\n", | |
| " <th>min(Average Market Land Value)</th>\n", | |
| " <th>min(Average Market Value Improvements)</th>\n", | |
| " <th>mean(Average Last Sale Amount)</th>\n", | |
| " <th>min(Average Last Sale Amount)</th>\n", | |
| " <th>var(Average Market Land Value)</th>\n", | |
| " <th>var(Average Tax Billed Amount)</th>\n", | |
| " <th>...</th>\n", | |
| " <th>The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months</th>\n", | |
| " <th>Percentage Divorced</th>\n", | |
| " <th>The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months</th>\n", | |
| " <th>Percentage of Homes With a Second Mortgage and Home Equity Loan</th>\n", | |
| " <th>Percentage Separated</th>\n", | |
| " <th>Number of Establishments</th>\n", | |
| " <th>First Quarter Payroll</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->0</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->1</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->2</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->3</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->4</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->5</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->6</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->7</th>\n", | |
| " <th>view->0</th>\n", | |
| " <th>view->1</th>\n", | |
| " <th>view->2</th>\n", | |
| " <th>view->3</th>\n", | |
| " <th>view->4</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->0</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->1</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->2</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->3</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->4</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->5</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->6</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->7</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->empty</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->0</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->1</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->2</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->3</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->4</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->5</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->6</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->7</th>\n", | |
| " <th>waterfront->0</th>\n", | |
| " <th>waterfront->1</th>\n", | |
| " <th>_TARGET_</th>\n", | |
| " </tr>\n", | |
| " </thead>\n", | |
| " <tbody>\n", | |
| " <tr>\n", | |
| " <th>0</th>\n", | |
| " <td>3</td>\n", | |
| " <td>1.00</td>\n", | |
| " <td>1180</td>\n", | |
| " <td>5650</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>7</td>\n", | |
| " <td>1180</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1955</td>\n", | |
| " <td>0</td>\n", | |
| " <td>98178</td>\n", | |
| " <td>47.5112</td>\n", | |
| " <td>-122.257</td>\n", | |
| " <td>1340</td>\n", | |
| " <td>5650</td>\n", | |
| " <td>353965.0</td>\n", | |
| " <td>1180</td>\n", | |
| " <td>7</td>\n", | |
| " <td>1180</td>\n", | |
| " <td>1340</td>\n", | |
| " <td>1.00</td>\n", | |
| " <td>9</td>\n", | |
| " <td>0</td>\n", | |
| " <td>25</td>\n", | |
| " <td>46</td>\n", | |
| " <td>0</td>\n", | |
| " <td>314603.12</td>\n", | |
| " <td>1139063.24</td>\n", | |
| " <td>412230.06</td>\n", | |
| " <td>4369.4263</td>\n", | |
| " <td>373499.24</td>\n", | |
| " <td>133953.23</td>\n", | |
| " <td>11474.3236</td>\n", | |
| " <td>105592.78</td>\n", | |
| " <td>209010.36</td>\n", | |
| " <td>218550.18</td>\n", | |
| " <td>199684.39</td>\n", | |
| " <td>2.681050e+08</td>\n", | |
| " <td>265443.251323</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.616020</td>\n", | |
| " <td>0.090615</td>\n", | |
| " <td>0.763580</td>\n", | |
| " <td>0.015080</td>\n", | |
| " <td>0.029380</td>\n", | |
| " <td>255</td>\n", | |
| " <td>16499</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>221900.0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>1</th>\n", | |
| " <td>3</td>\n", | |
| " <td>2.25</td>\n", | |
| " <td>2570</td>\n", | |
| " <td>7242</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>7</td>\n", | |
| " <td>2170</td>\n", | |
| " <td>400</td>\n", | |
| " <td>1951</td>\n", | |
| " <td>1991</td>\n", | |
| " <td>98125</td>\n", | |
| " <td>47.7210</td>\n", | |
| " <td>-122.319</td>\n", | |
| " <td>1690</td>\n", | |
| " <td>7639</td>\n", | |
| " <td>391522.5</td>\n", | |
| " <td>2570</td>\n", | |
| " <td>7</td>\n", | |
| " <td>2170</td>\n", | |
| " <td>1690</td>\n", | |
| " <td>2.25</td>\n", | |
| " <td>35</td>\n", | |
| " <td>400</td>\n", | |
| " <td>52</td>\n", | |
| " <td>128</td>\n", | |
| " <td>1991</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.635215</td>\n", | |
| " <td>0.098587</td>\n", | |
| " <td>0.760710</td>\n", | |
| " <td>0.022543</td>\n", | |
| " <td>0.016978</td>\n", | |
| " <td>1075</td>\n", | |
| " <td>91798</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>538000.0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>2</th>\n", | |
| " <td>2</td>\n", | |
| " <td>1.00</td>\n", | |
| " <td>770</td>\n", | |
| " <td>10000</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>6</td>\n", | |
| " <td>770</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1933</td>\n", | |
| " <td>0</td>\n", | |
| " <td>98028</td>\n", | |
| " <td>47.7379</td>\n", | |
| " <td>-122.233</td>\n", | |
| " <td>2720</td>\n", | |
| " <td>8062</td>\n", | |
| " <td>482775.0</td>\n", | |
| " <td>770</td>\n", | |
| " <td>6</td>\n", | |
| " <td>770</td>\n", | |
| " <td>2720</td>\n", | |
| " <td>1.00</td>\n", | |
| " <td>13</td>\n", | |
| " <td>0</td>\n", | |
| " <td>36</td>\n", | |
| " <td>39</td>\n", | |
| " <td>0</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.680070</td>\n", | |
| " <td>0.102450</td>\n", | |
| " <td>0.797507</td>\n", | |
| " <td>0.051937</td>\n", | |
| " <td>0.005570</td>\n", | |
| " <td>472</td>\n", | |
| " <td>26462</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>180000.0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>3</th>\n", | |
| " <td>4</td>\n", | |
| " <td>3.00</td>\n", | |
| " <td>1960</td>\n", | |
| " <td>5000</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>7</td>\n", | |
| " <td>1050</td>\n", | |
| " <td>910</td>\n", | |
| " <td>1965</td>\n", | |
| " <td>0</td>\n", | |
| " <td>98136</td>\n", | |
| " <td>47.5208</td>\n", | |
| " <td>-122.393</td>\n", | |
| " <td>1360</td>\n", | |
| " <td>5000</td>\n", | |
| " <td>601275.0</td>\n", | |
| " <td>1960</td>\n", | |
| " <td>7</td>\n", | |
| " <td>1050</td>\n", | |
| " <td>1360</td>\n", | |
| " <td>3.00</td>\n", | |
| " <td>4</td>\n", | |
| " <td>910</td>\n", | |
| " <td>17</td>\n", | |
| " <td>23</td>\n", | |
| " <td>0</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>...</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>NaN</td>\n", | |
| " <td>347</td>\n", | |
| " <td>17553</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>604000.0</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>4</th>\n", | |
| " <td>3</td>\n", | |
| " <td>2.00</td>\n", | |
| " <td>1680</td>\n", | |
| " <td>8080</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>8</td>\n", | |
| " <td>1680</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1987</td>\n", | |
| " <td>0</td>\n", | |
| " <td>98074</td>\n", | |
| " <td>47.6168</td>\n", | |
| " <td>-122.045</td>\n", | |
| " <td>1800</td>\n", | |
| " <td>7503</td>\n", | |
| " <td>523287.5</td>\n", | |
| " <td>1680</td>\n", | |
| " <td>8</td>\n", | |
| " <td>1680</td>\n", | |
| " <td>1800</td>\n", | |
| " <td>2.00</td>\n", | |
| " <td>44</td>\n", | |
| " <td>0</td>\n", | |
| " <td>30</td>\n", | |
| " <td>13</td>\n", | |
| " <td>0</td>\n", | |
| " <td>509856.38</td>\n", | |
| " <td>1687133.58</td>\n", | |
| " <td>588638.60</td>\n", | |
| " <td>6389.3823</td>\n", | |
| " <td>840044.00</td>\n", | |
| " <td>291288.10</td>\n", | |
| " <td>17177.8288</td>\n", | |
| " <td>257467.80</td>\n", | |
| " <td>252388.58</td>\n", | |
| " <td>364695.23</td>\n", | |
| " <td>343927.34</td>\n", | |
| " <td>3.812709e+08</td>\n", | |
| " <td>330414.563884</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.465560</td>\n", | |
| " <td>0.014840</td>\n", | |
| " <td>0.655650</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>446</td>\n", | |
| " <td>20811</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>0</td>\n", | |
| " <td>1</td>\n", | |
| " <td>0</td>\n", | |
| " <td>510000.0</td>\n", | |
| " </tr>\n", | |
| " </tbody>\n", | |
| "</table>\n", | |
| "<p>5 rows × 94 columns</p>\n", | |
| "</div>" | |
| ], | |
| "text/plain": [ | |
| " bedrooms bathrooms sqft_living ... waterfront->0 waterfront->1 _TARGET_\n", | |
| "0 3 1.00 1180 ... 1 0 221900.0\n", | |
| "1 3 2.25 2570 ... 1 0 538000.0\n", | |
| "2 2 1.00 770 ... 1 0 180000.0\n", | |
| "3 4 3.00 1960 ... 1 0 604000.0\n", | |
| "4 3 2.00 1680 ... 1 0 510000.0\n", | |
| "\n", | |
| "[5 rows x 94 columns]" | |
| ] | |
| }, | |
| "metadata": { | |
| "tags": [] | |
| }, | |
| "execution_count": 13 | |
| } | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "metadata": { | |
| "id": "wR6FUqoXCadg", | |
| "colab_type": "code", | |
| "colab": { | |
| "base_uri": "https://localhost:8080/", | |
| "height": 1000 | |
| }, | |
| "outputId": "7410a063-99c6-4209-ebb7-2a8aa9eb05ec" | |
| }, | |
| "source": [ | |
| "df.info()" | |
| ], | |
| "execution_count": 14, | |
| "outputs": [ | |
| { | |
| "output_type": "stream", | |
| "text": [ | |
| "<class 'pandas.core.frame.DataFrame'>\n", | |
| "RangeIndex: 21613 entries, 0 to 21612\n", | |
| "Data columns (total 94 columns):\n", | |
| " # Column Non-Null Count Dtype \n", | |
| "--- ------ -------------- ----- \n", | |
| " 0 bedrooms 21613 non-null int64 \n", | |
| " 1 bathrooms 21613 non-null float64\n", | |
| " 2 sqft_living 21613 non-null int64 \n", | |
| " 3 sqft_lot 21613 non-null int64 \n", | |
| " 4 waterfront 21613 non-null int64 \n", | |
| " 5 view 21613 non-null int64 \n", | |
| " 6 grade 21613 non-null int64 \n", | |
| " 7 sqft_above 21613 non-null int64 \n", | |
| " 8 sqft_basement 21613 non-null int64 \n", | |
| " 9 yr_built 21613 non-null int64 \n", | |
| " 10 yr_renovated 21613 non-null int64 \n", | |
| " 11 zipcode 21613 non-null int64 \n", | |
| " 12 lat 21613 non-null float64\n", | |
| " 13 long 21613 non-null float64\n", | |
| " 14 sqft_living15 21613 non-null int64 \n", | |
| " 15 sqft_lot15 21613 non-null int64 \n", | |
| " 16 KNeighbors(lat, long) 21613 non-null float64\n", | |
| " 17 sqft_living.1 21613 non-null int64 \n", | |
| " 18 grade.1 21613 non-null int64 \n", | |
| " 19 sqft_above.1 21613 non-null int64 \n", | |
| " 20 sqft_living15.1 21613 non-null int64 \n", | |
| " 21 bathrooms.1 21613 non-null float64\n", | |
| " 22 Number Of \"parking\" In 3.3952 KM Radius 21613 non-null int64 \n", | |
| " 23 sqft_basement.1 21613 non-null int64 \n", | |
| " 24 Number Of \"school\" In 3.0972 KM Radius 21613 non-null int64 \n", | |
| " 25 Number Of \"post_box\" In 6.8997 KM Radius 21613 non-null int64 \n", | |
| " 26 yr_renovated.1 21613 non-null int64 \n", | |
| " 27 min(Average Total Market Value) 14606 non-null float64\n", | |
| " 28 sum(Average Total Market Value) 14606 non-null float64\n", | |
| " 29 max(Average Total Market Value) 14606 non-null float64\n", | |
| " 30 max(Average Tax Billed Amount) 14606 non-null float64\n", | |
| " 31 sum(Average Market Land Value) 14606 non-null float64\n", | |
| " 32 max(Average Market Land Value) 14606 non-null float64\n", | |
| " 33 sum(Average Tax Billed Amount) 14606 non-null float64\n", | |
| " 34 min(Average Market Land Value) 14606 non-null float64\n", | |
| " 35 min(Average Market Value Improvements) 14606 non-null float64\n", | |
| " 36 mean(Average Last Sale Amount) 14606 non-null float64\n", | |
| " 37 min(Average Last Sale Amount) 14606 non-null float64\n", | |
| " 38 var(Average Market Land Value) 14606 non-null float64\n", | |
| " 39 var(Average Tax Billed Amount) 14606 non-null float64\n", | |
| " 40 var(Average Total Market Value) 14606 non-null float64\n", | |
| " 41 sum(Average Basement Finished Area) 14606 non-null float64\n", | |
| " 42 min(Average Number of Bedrooms) 14606 non-null float64\n", | |
| " 43 var(Average Market Value Improvements) 14606 non-null float64\n", | |
| " 44 avg_income 21613 non-null int64 \n", | |
| " 45 Median Family Income 20240 non-null float64\n", | |
| " 46 Mean Household Income 20240 non-null float64\n", | |
| " 47 Mean Monthly Owner Costs 20240 non-null float64\n", | |
| " 48 Median Monthly Mortgage and Owner Costs 20240 non-null float64\n", | |
| " 49 Median Monthly Owner Costs 20240 non-null float64\n", | |
| " 50 The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months 20240 non-null float64\n", | |
| " 51 Rent Mean 20240 non-null float64\n", | |
| " 52 Mean Monthly Mortgage and Owner Costs 20240 non-null float64\n", | |
| " 53 The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months 20240 non-null float64\n", | |
| " 54 The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months 20240 non-null float64\n", | |
| " 55 Percentage Divorced 20240 non-null float64\n", | |
| " 56 The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months 20240 non-null float64\n", | |
| " 57 Percentage of Homes With a Second Mortgage and Home Equity Loan 20240 non-null float64\n", | |
| " 58 Percentage Separated 20240 non-null float64\n", | |
| " 59 Number of Establishments 21613 non-null int64 \n", | |
| " 60 First Quarter Payroll 21613 non-null int64 \n", | |
| " 61 KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->0 21613 non-null int64 \n", | |
| " 62 KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->1 21613 non-null int64 \n", | |
| " 63 KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->2 21613 non-null int64 \n", | |
| " 64 KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->3 21613 non-null int64 \n", | |
| " 65 KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->4 21613 non-null int64 \n", | |
| " 66 KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->5 21613 non-null int64 \n", | |
| " 67 KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->6 21613 non-null int64 \n", | |
| " 68 KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->7 21613 non-null int64 \n", | |
| " 69 view->0 21613 non-null int64 \n", | |
| " 70 view->1 21613 non-null int64 \n", | |
| " 71 view->2 21613 non-null int64 \n", | |
| " 72 view->3 21613 non-null int64 \n", | |
| " 73 view->4 21613 non-null int64 \n", | |
| " 74 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->0 21613 non-null int64 \n", | |
| " 75 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->1 21613 non-null int64 \n", | |
| " 76 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->2 21613 non-null int64 \n", | |
| " 77 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->3 21613 non-null int64 \n", | |
| " 78 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->4 21613 non-null int64 \n", | |
| " 79 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->5 21613 non-null int64 \n", | |
| " 80 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->6 21613 non-null int64 \n", | |
| " 81 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->7 21613 non-null int64 \n", | |
| " 82 KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->empty 21613 non-null int64 \n", | |
| " 83 KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->0 21613 non-null int64 \n", | |
| " 84 KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->1 21613 non-null int64 \n", | |
| " 85 KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->2 21613 non-null int64 \n", | |
| " 86 KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->3 21613 non-null int64 \n", | |
| " 87 KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->4 21613 non-null int64 \n", | |
| " 88 KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->5 21613 non-null int64 \n", | |
| " 89 KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->6 21613 non-null int64 \n", | |
| " 90 KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->7 21613 non-null int64 \n", | |
| " 91 waterfront->0 21613 non-null int64 \n", | |
| " 92 waterfront->1 21613 non-null int64 \n", | |
| " 93 _TARGET_ 21613 non-null float64\n", | |
| "dtypes: float64(37), int64(57)\n", | |
| "memory usage: 15.5 MB\n" | |
| ], | |
| "name": "stdout" | |
| } | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "metadata": { | |
| "id": "WTyz3jdpDO_n", | |
| "colab_type": "code", | |
| "colab": { | |
| "base_uri": "https://localhost:8080/", | |
| "height": 1000 | |
| }, | |
| "outputId": "5b11e31b-611c-4314-fa73-30714810f23e" | |
| }, | |
| "source": [ | |
| "df.describe()" | |
| ], | |
| "execution_count": 15, | |
| "outputs": [ | |
| { | |
| "output_type": "execute_result", | |
| "data": { | |
| "text/html": [ | |
| "<div>\n", | |
| "<style scoped>\n", | |
| " .dataframe tbody tr th:only-of-type {\n", | |
| " vertical-align: middle;\n", | |
| " }\n", | |
| "\n", | |
| " .dataframe tbody tr th {\n", | |
| " vertical-align: top;\n", | |
| " }\n", | |
| "\n", | |
| " .dataframe thead th {\n", | |
| " text-align: right;\n", | |
| " }\n", | |
| "</style>\n", | |
| "<table border=\"1\" class=\"dataframe\">\n", | |
| " <thead>\n", | |
| " <tr style=\"text-align: right;\">\n", | |
| " <th></th>\n", | |
| " <th>bedrooms</th>\n", | |
| " <th>bathrooms</th>\n", | |
| " <th>sqft_living</th>\n", | |
| " <th>sqft_lot</th>\n", | |
| " <th>waterfront</th>\n", | |
| " <th>view</th>\n", | |
| " <th>grade</th>\n", | |
| " <th>sqft_above</th>\n", | |
| " <th>sqft_basement</th>\n", | |
| " <th>yr_built</th>\n", | |
| " <th>yr_renovated</th>\n", | |
| " <th>zipcode</th>\n", | |
| " <th>lat</th>\n", | |
| " <th>long</th>\n", | |
| " <th>sqft_living15</th>\n", | |
| " <th>sqft_lot15</th>\n", | |
| " <th>KNeighbors(lat, long)</th>\n", | |
| " <th>sqft_living.1</th>\n", | |
| " <th>grade.1</th>\n", | |
| " <th>sqft_above.1</th>\n", | |
| " <th>sqft_living15.1</th>\n", | |
| " <th>bathrooms.1</th>\n", | |
| " <th>Number Of \"parking\" In 3.3952 KM Radius</th>\n", | |
| " <th>sqft_basement.1</th>\n", | |
| " <th>Number Of \"school\" In 3.0972 KM Radius</th>\n", | |
| " <th>Number Of \"post_box\" In 6.8997 KM Radius</th>\n", | |
| " <th>yr_renovated.1</th>\n", | |
| " <th>min(Average Total Market Value)</th>\n", | |
| " <th>sum(Average Total Market Value)</th>\n", | |
| " <th>max(Average Total Market Value)</th>\n", | |
| " <th>max(Average Tax Billed Amount)</th>\n", | |
| " <th>sum(Average Market Land Value)</th>\n", | |
| " <th>max(Average Market Land Value)</th>\n", | |
| " <th>sum(Average Tax Billed Amount)</th>\n", | |
| " <th>min(Average Market Land Value)</th>\n", | |
| " <th>min(Average Market Value Improvements)</th>\n", | |
| " <th>mean(Average Last Sale Amount)</th>\n", | |
| " <th>min(Average Last Sale Amount)</th>\n", | |
| " <th>var(Average Market Land Value)</th>\n", | |
| " <th>var(Average Tax Billed Amount)</th>\n", | |
| " <th>...</th>\n", | |
| " <th>The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months</th>\n", | |
| " <th>Percentage Divorced</th>\n", | |
| " <th>The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months</th>\n", | |
| " <th>Percentage of Homes With a Second Mortgage and Home Equity Loan</th>\n", | |
| " <th>Percentage Separated</th>\n", | |
| " <th>Number of Establishments</th>\n", | |
| " <th>First Quarter Payroll</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->0</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->1</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->2</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->3</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->4</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->5</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->6</th>\n", | |
| " <th>KMeans_Clustering(bedrooms, bathrooms, sqft_living, sqft_lot, grade, sqft_above, sqft_basement, yr_built, yr_renovated, sqft_living15, sqft_lot15)->7</th>\n", | |
| " <th>view->0</th>\n", | |
| " <th>view->1</th>\n", | |
| " <th>view->2</th>\n", | |
| " <th>view->3</th>\n", | |
| " <th>view->4</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->0</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->1</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->2</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->3</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->4</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->5</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->6</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->7</th>\n", | |
| " <th>KMeans_Clustering(Percent of Houses With a Second Mortgage, Percentage of People With at Least High School Degree, Rent Median, Rent Mean, The empirical distribution value that an individual’s rent will be greater than 15% of their household income in the past 12 months, Total Population, Percentage Divorced, The empirical distribution value that an individual’s rent will be greater than 25% of their household income in the past 12 months, Percentage of Males With at Least High School Degree, Percentage of Homes With a Second Mortgage and Home Equity Loan, Percentage Separated, The empirical distribution value that an individual’s rent will be greater than 20% of their household income in the past 12 months, Median Monthly Owner Costs, Mean Family Income, The empirical distribution value that an individual’s rent will be greater than 10% of their household income in the past 12 months, Mean Household Income, The empirical distribution value that an individual’s rent will be greater than 35% of their household income in the past 12 months, Percentage of Homes With a Home Equity Loan, Median Monthly Mortgage and Owner Costs, The empirical distribution value that an individual’s rent will be greater than 30% of their household income in the past 12 months, Percentage of Homes With Some Type of Debt, Male Population, The empirical distribution value that an individual’s rent will be greater than 50% of their household income in the past 12 months, Median Family Income, The empirical distribution value that an individual’s rent will be greater than 40% of their household income in the past 12 months, Mean Monthly Owner Costs, Percentage of Females With at Least High School Degree, Female Population, Mean Monthly Mortgage and Owner Costs, Percentage Married)->empty</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->0</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->1</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->2</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->3</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->4</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->5</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->6</th>\n", | |
| " <th>KMeans_Clustering(Annual Payroll, Number of Employees, First Quarter Payroll, Number of Establishments)->7</th>\n", | |
| " <th>waterfront->0</th>\n", | |
| " <th>waterfront->1</th>\n", | |
| " <th>_TARGET_</th>\n", | |
| " </tr>\n", | |
| " </thead>\n", | |
| " <tbody>\n", | |
| " <tr>\n", | |
| " <th>count</th>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>2.161300e+04</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>2.161300e+04</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>14606.000000</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>14606.000000</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>1.460600e+04</td>\n", | |
| " <td>...</td>\n", | |
| " <td>20240.000000</td>\n", | |
| " <td>20240.000000</td>\n", | |
| " <td>20240.000000</td>\n", | |
| " <td>20240.000000</td>\n", | |
| " <td>20240.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>2.161300e+04</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.00000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>21613.000000</td>\n", | |
| " <td>2.161300e+04</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>mean</th>\n", | |
| " <td>3.370842</td>\n", | |
| " <td>2.114757</td>\n", | |
| " <td>2079.899736</td>\n", | |
| " <td>1.510697e+04</td>\n", | |
| " <td>0.007542</td>\n", | |
| " <td>0.234303</td>\n", | |
| " <td>7.656873</td>\n", | |
| " <td>1788.390691</td>\n", | |
| " <td>291.509045</td>\n", | |
| " <td>1971.005136</td>\n", | |
| " <td>84.402258</td>\n", | |
| " <td>98077.939805</td>\n", | |
| " <td>47.560053</td>\n", | |
| " <td>-122.213896</td>\n", | |
| " <td>1986.552492</td>\n", | |
| " <td>12768.455652</td>\n", | |
| " <td>5.293282e+05</td>\n", | |
| " <td>2079.899736</td>\n", | |
| " <td>7.656873</td>\n", | |
| " <td>1788.390691</td>\n", | |
| " <td>1986.552492</td>\n", | |
| " <td>2.114757</td>\n", | |
| " <td>28.002961</td>\n", | |
| " <td>291.509045</td>\n", | |
| " <td>29.608291</td>\n", | |
| " <td>65.058622</td>\n", | |
| " <td>84.402258</td>\n", | |
| " <td>4.991324e+05</td>\n", | |
| " <td>1.731990e+06</td>\n", | |
| " <td>6.170389e+05</td>\n", | |
| " <td>6668.517111</td>\n", | |
| " <td>7.556021e+05</td>\n", | |
| " <td>2.679120e+05</td>\n", | |
| " <td>17836.754765</td>\n", | |
| " <td>2.204429e+05</td>\n", | |
| " <td>2.781605e+05</td>\n", | |
| " <td>3.907671e+05</td>\n", | |
| " <td>3.607855e+05</td>\n", | |
| " <td>1.290823e+09</td>\n", | |
| " <td>5.767311e+05</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.588437</td>\n", | |
| " <td>0.088037</td>\n", | |
| " <td>0.733057</td>\n", | |
| " <td>0.031754</td>\n", | |
| " <td>0.013808</td>\n", | |
| " <td>903.476426</td>\n", | |
| " <td>2.478813e+05</td>\n", | |
| " <td>0.171286</td>\n", | |
| " <td>0.014945</td>\n", | |
| " <td>0.235460</td>\n", | |
| " <td>0.040763</td>\n", | |
| " <td>0.235830</td>\n", | |
| " <td>0.047148</td>\n", | |
| " <td>0.172720</td>\n", | |
| " <td>0.081849</td>\n", | |
| " <td>0.901726</td>\n", | |
| " <td>0.015361</td>\n", | |
| " <td>0.044557</td>\n", | |
| " <td>0.023597</td>\n", | |
| " <td>0.014760</td>\n", | |
| " <td>0.12548</td>\n", | |
| " <td>0.301439</td>\n", | |
| " <td>0.187989</td>\n", | |
| " <td>0.239671</td>\n", | |
| " <td>0.029658</td>\n", | |
| " <td>0.014667</td>\n", | |
| " <td>0.010595</td>\n", | |
| " <td>0.026975</td>\n", | |
| " <td>0.063527</td>\n", | |
| " <td>0.149493</td>\n", | |
| " <td>0.333364</td>\n", | |
| " <td>0.026558</td>\n", | |
| " <td>0.010827</td>\n", | |
| " <td>0.014667</td>\n", | |
| " <td>0.266321</td>\n", | |
| " <td>0.052561</td>\n", | |
| " <td>0.146208</td>\n", | |
| " <td>0.992458</td>\n", | |
| " <td>0.007542</td>\n", | |
| " <td>5.400881e+05</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>std</th>\n", | |
| " <td>0.930062</td>\n", | |
| " <td>0.770163</td>\n", | |
| " <td>918.440897</td>\n", | |
| " <td>4.142051e+04</td>\n", | |
| " <td>0.086517</td>\n", | |
| " <td>0.766318</td>\n", | |
| " <td>1.175459</td>\n", | |
| " <td>828.090978</td>\n", | |
| " <td>442.575043</td>\n", | |
| " <td>29.373411</td>\n", | |
| " <td>401.679240</td>\n", | |
| " <td>53.505026</td>\n", | |
| " <td>0.138564</td>\n", | |
| " <td>0.140828</td>\n", | |
| " <td>685.391304</td>\n", | |
| " <td>27304.179631</td>\n", | |
| " <td>2.671836e+05</td>\n", | |
| " <td>918.440897</td>\n", | |
| " <td>1.175459</td>\n", | |
| " <td>828.090978</td>\n", | |
| " <td>685.391304</td>\n", | |
| " <td>0.770163</td>\n", | |
| " <td>24.086282</td>\n", | |
| " <td>442.575043</td>\n", | |
| " <td>15.929420</td>\n", | |
| " <td>67.132014</td>\n", | |
| " <td>401.679240</td>\n", | |
| " <td>2.985804e+05</td>\n", | |
| " <td>1.019579e+06</td>\n", | |
| " <td>3.620719e+05</td>\n", | |
| " <td>3102.675799</td>\n", | |
| " <td>5.922501e+05</td>\n", | |
| " <td>2.092403e+05</td>\n", | |
| " <td>8292.144601</td>\n", | |
| " <td>1.744674e+05</td>\n", | |
| " <td>1.473483e+05</td>\n", | |
| " <td>2.839253e+05</td>\n", | |
| " <td>2.665214e+05</td>\n", | |
| " <td>3.291815e+09</td>\n", | |
| " <td>1.077895e+06</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.113196</td>\n", | |
| " <td>0.026654</td>\n", | |
| " <td>0.092420</td>\n", | |
| " <td>0.017389</td>\n", | |
| " <td>0.009169</td>\n", | |
| " <td>577.051946</td>\n", | |
| " <td>4.954311e+05</td>\n", | |
| " <td>0.376767</td>\n", | |
| " <td>0.121334</td>\n", | |
| " <td>0.424296</td>\n", | |
| " <td>0.197744</td>\n", | |
| " <td>0.424526</td>\n", | |
| " <td>0.211959</td>\n", | |
| " <td>0.378014</td>\n", | |
| " <td>0.274141</td>\n", | |
| " <td>0.297692</td>\n", | |
| " <td>0.122987</td>\n", | |
| " <td>0.206333</td>\n", | |
| " <td>0.151793</td>\n", | |
| " <td>0.120592</td>\n", | |
| " <td>0.33127</td>\n", | |
| " <td>0.458894</td>\n", | |
| " <td>0.390712</td>\n", | |
| " <td>0.426892</td>\n", | |
| " <td>0.169646</td>\n", | |
| " <td>0.120219</td>\n", | |
| " <td>0.102390</td>\n", | |
| " <td>0.162013</td>\n", | |
| " <td>0.243913</td>\n", | |
| " <td>0.356582</td>\n", | |
| " <td>0.471426</td>\n", | |
| " <td>0.160792</td>\n", | |
| " <td>0.103490</td>\n", | |
| " <td>0.120219</td>\n", | |
| " <td>0.442044</td>\n", | |
| " <td>0.223160</td>\n", | |
| " <td>0.353323</td>\n", | |
| " <td>0.086517</td>\n", | |
| " <td>0.086517</td>\n", | |
| " <td>3.671272e+05</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>min</th>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>290.000000</td>\n", | |
| " <td>5.200000e+02</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>290.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1900.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>98001.000000</td>\n", | |
| " <td>47.155900</td>\n", | |
| " <td>-122.519000</td>\n", | |
| " <td>399.000000</td>\n", | |
| " <td>651.000000</td>\n", | |
| " <td>1.776420e+05</td>\n", | |
| " <td>290.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>290.000000</td>\n", | |
| " <td>399.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>5.000000e+02</td>\n", | |
| " <td>1.500000e+03</td>\n", | |
| " <td>5.000000e+02</td>\n", | |
| " <td>17.675000</td>\n", | |
| " <td>1.500000e+03</td>\n", | |
| " <td>5.000000e+02</td>\n", | |
| " <td>52.545000</td>\n", | |
| " <td>5.000000e+02</td>\n", | |
| " <td>0.000000e+00</td>\n", | |
| " <td>1.000000e+02</td>\n", | |
| " <td>1.000000e+02</td>\n", | |
| " <td>0.000000e+00</td>\n", | |
| " <td>3.360000e-02</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.097050</td>\n", | |
| " <td>0.014840</td>\n", | |
| " <td>0.413500</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>69.000000</td>\n", | |
| " <td>3.424000e+03</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.00000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>7.500000e+04</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>25%</th>\n", | |
| " <td>3.000000</td>\n", | |
| " <td>1.750000</td>\n", | |
| " <td>1427.000000</td>\n", | |
| " <td>5.040000e+03</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>7.000000</td>\n", | |
| " <td>1190.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1951.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>98033.000000</td>\n", | |
| " <td>47.471000</td>\n", | |
| " <td>-122.328000</td>\n", | |
| " <td>1490.000000</td>\n", | |
| " <td>5100.000000</td>\n", | |
| " <td>3.390450e+05</td>\n", | |
| " <td>1427.000000</td>\n", | |
| " <td>7.000000</td>\n", | |
| " <td>1190.000000</td>\n", | |
| " <td>1490.000000</td>\n", | |
| " <td>1.750000</td>\n", | |
| " <td>7.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>18.000000</td>\n", | |
| " <td>17.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>3.071572e+05</td>\n", | |
| " <td>1.056681e+06</td>\n", | |
| " <td>3.752705e+05</td>\n", | |
| " <td>4756.130000</td>\n", | |
| " <td>3.779581e+05</td>\n", | |
| " <td>1.344959e+05</td>\n", | |
| " <td>12816.496800</td>\n", | |
| " <td>1.127331e+05</td>\n", | |
| " <td>1.837323e+05</td>\n", | |
| " <td>2.358284e+05</td>\n", | |
| " <td>2.177934e+05</td>\n", | |
| " <td>1.340566e+08</td>\n", | |
| " <td>1.961008e+05</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.526110</td>\n", | |
| " <td>0.075483</td>\n", | |
| " <td>0.686050</td>\n", | |
| " <td>0.019633</td>\n", | |
| " <td>0.009160</td>\n", | |
| " <td>472.000000</td>\n", | |
| " <td>3.505100e+04</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.00000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>3.219500e+05</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>50%</th>\n", | |
| " <td>3.000000</td>\n", | |
| " <td>2.250000</td>\n", | |
| " <td>1910.000000</td>\n", | |
| " <td>7.618000e+03</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>7.000000</td>\n", | |
| " <td>1560.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1975.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>98065.000000</td>\n", | |
| " <td>47.571800</td>\n", | |
| " <td>-122.230000</td>\n", | |
| " <td>1840.000000</td>\n", | |
| " <td>7620.000000</td>\n", | |
| " <td>4.720625e+05</td>\n", | |
| " <td>1910.000000</td>\n", | |
| " <td>7.000000</td>\n", | |
| " <td>1560.000000</td>\n", | |
| " <td>1840.000000</td>\n", | |
| " <td>2.250000</td>\n", | |
| " <td>18.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>28.000000</td>\n", | |
| " <td>34.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>4.394676e+05</td>\n", | |
| " <td>1.531614e+06</td>\n", | |
| " <td>5.462552e+05</td>\n", | |
| " <td>6059.696000</td>\n", | |
| " <td>6.360800e+05</td>\n", | |
| " <td>2.232265e+05</td>\n", | |
| " <td>16251.382100</td>\n", | |
| " <td>1.841822e+05</td>\n", | |
| " <td>2.467990e+05</td>\n", | |
| " <td>3.264923e+05</td>\n", | |
| " <td>2.992168e+05</td>\n", | |
| " <td>4.875428e+08</td>\n", | |
| " <td>3.704247e+05</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.593048</td>\n", | |
| " <td>0.090342</td>\n", | |
| " <td>0.754355</td>\n", | |
| " <td>0.028015</td>\n", | |
| " <td>0.011958</td>\n", | |
| " <td>719.000000</td>\n", | |
| " <td>7.386000e+04</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.00000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>4.500000e+05</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>75%</th>\n", | |
| " <td>4.000000</td>\n", | |
| " <td>2.500000</td>\n", | |
| " <td>2550.000000</td>\n", | |
| " <td>1.068800e+04</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>8.000000</td>\n", | |
| " <td>2210.000000</td>\n", | |
| " <td>560.000000</td>\n", | |
| " <td>1997.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>98118.000000</td>\n", | |
| " <td>47.678000</td>\n", | |
| " <td>-122.125000</td>\n", | |
| " <td>2360.000000</td>\n", | |
| " <td>10083.000000</td>\n", | |
| " <td>6.322650e+05</td>\n", | |
| " <td>2550.000000</td>\n", | |
| " <td>8.000000</td>\n", | |
| " <td>2210.000000</td>\n", | |
| " <td>2360.000000</td>\n", | |
| " <td>2.500000</td>\n", | |
| " <td>52.000000</td>\n", | |
| " <td>560.000000</td>\n", | |
| " <td>41.000000</td>\n", | |
| " <td>103.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>5.954931e+05</td>\n", | |
| " <td>2.065023e+06</td>\n", | |
| " <td>7.407536e+05</td>\n", | |
| " <td>7736.910600</td>\n", | |
| " <td>9.213458e+05</td>\n", | |
| " <td>3.250080e+05</td>\n", | |
| " <td>20772.840100</td>\n", | |
| " <td>2.643834e+05</td>\n", | |
| " <td>3.279511e+05</td>\n", | |
| " <td>4.557684e+05</td>\n", | |
| " <td>4.212920e+05</td>\n", | |
| " <td>1.272530e+09</td>\n", | |
| " <td>6.067995e+05</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.664628</td>\n", | |
| " <td>0.102450</td>\n", | |
| " <td>0.789697</td>\n", | |
| " <td>0.042650</td>\n", | |
| " <td>0.020972</td>\n", | |
| " <td>1264.000000</td>\n", | |
| " <td>2.369600e+05</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.00000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>0.000000</td>\n", | |
| " <td>6.450000e+05</td>\n", | |
| " </tr>\n", | |
| " <tr>\n", | |
| " <th>max</th>\n", | |
| " <td>33.000000</td>\n", | |
| " <td>8.000000</td>\n", | |
| " <td>13540.000000</td>\n", | |
| " <td>1.651359e+06</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>4.000000</td>\n", | |
| " <td>13.000000</td>\n", | |
| " <td>9410.000000</td>\n", | |
| " <td>4820.000000</td>\n", | |
| " <td>2015.000000</td>\n", | |
| " <td>2015.000000</td>\n", | |
| " <td>98199.000000</td>\n", | |
| " <td>47.777600</td>\n", | |
| " <td>-121.315000</td>\n", | |
| " <td>6210.000000</td>\n", | |
| " <td>871200.000000</td>\n", | |
| " <td>2.491448e+06</td>\n", | |
| " <td>13540.000000</td>\n", | |
| " <td>13.000000</td>\n", | |
| " <td>9410.000000</td>\n", | |
| " <td>6210.000000</td>\n", | |
| " <td>8.000000</td>\n", | |
| " <td>75.000000</td>\n", | |
| " <td>4820.000000</td>\n", | |
| " <td>64.000000</td>\n", | |
| " <td>211.000000</td>\n", | |
| " <td>2015.000000</td>\n", | |
| " <td>5.690511e+06</td>\n", | |
| " <td>1.924153e+07</td>\n", | |
| " <td>6.775511e+06</td>\n", | |
| " <td>60360.820000</td>\n", | |
| " <td>1.165971e+07</td>\n", | |
| " <td>4.081111e+06</td>\n", | |
| " <td>160743.274000</td>\n", | |
| " <td>3.497489e+06</td>\n", | |
| " <td>4.616882e+06</td>\n", | |
| " <td>7.718170e+06</td>\n", | |
| " <td>7.710338e+06</td>\n", | |
| " <td>1.135382e+11</td>\n", | |
| " <td>3.593186e+07</td>\n", | |
| " <td>...</td>\n", | |
| " <td>0.785890</td>\n", | |
| " <td>0.141148</td>\n", | |
| " <td>0.908680</td>\n", | |
| " <td>0.079740</td>\n", | |
| " <td>0.040370</td>\n", | |
| " <td>2999.000000</td>\n", | |
| " <td>2.777795e+06</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.00000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>1.000000</td>\n", | |
| " <td>7.700000e+06</td>\n", | |
| " </tr>\n", | |
| " </tbody>\n", | |
| "</table>\n", | |
| "<p>8 rows × 94 columns</p>\n", | |
| "</div>" | |
| ], | |
| "text/plain": [ | |
| " bedrooms bathrooms ... waterfront->1 _TARGET_\n", | |
| "count 21613.000000 21613.000000 ... 21613.000000 2.161300e+04\n", | |
| "mean 3.370842 2.114757 ... 0.007542 5.400881e+05\n", | |
| "std 0.930062 0.770163 ... 0.086517 3.671272e+05\n", | |
| "min 0.000000 0.000000 ... 0.000000 7.500000e+04\n", | |
| "25% 3.000000 1.750000 ... 0.000000 3.219500e+05\n", | |
| "50% 3.000000 2.250000 ... 0.000000 4.500000e+05\n", | |
| "75% 4.000000 2.500000 ... 0.000000 6.450000e+05\n", | |
| "max 33.000000 8.000000 ... 1.000000 7.700000e+06\n", | |
| "\n", | |
| "[8 rows x 94 columns]" | |
| ] | |
| }, | |
| "metadata": { | |
| "tags": [] | |
| }, | |
| "execution_count": 15 | |
| } | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "metadata": { | |
| "id": "1lbfhYwlDQnV", | |
| "colab_type": "code", | |
| "colab": { | |
| "base_uri": "https://localhost:8080/", | |
| "height": 35 | |
| }, | |
| "outputId": "d8237cb9-54fb-4550-9f6a-763f80ec82d6" | |
| }, | |
| "source": [ | |
| "df.shape" | |
| ], | |
| "execution_count": 17, | |
| "outputs": [ | |
| { | |
| "output_type": "execute_result", | |
| "data": { | |
| "text/plain": [ | |
| "(21613, 94)" | |
| ] | |
| }, | |
| "metadata": { | |
| "tags": [] | |
| }, | |
| "execution_count": 17 | |
| } | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "metadata": { | |
| "id": "-Z4jTUqHDTYU", | |
| "colab_type": "code", | |
| "colab": {} | |
| }, | |
| "source": [ | |
| "" | |
| ], | |
| "execution_count": 0, | |
| "outputs": [] | |
| } | |
| ] | |
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment