Housing Price Regression Model Shootout

Housing Price Prediction Models

This nootbook contains a shootout between regression type models, attempting to predict housing prices based on a number of variables. Some of these variables include: Zoning, Lot Size, Pools, Bedrooms, etc.

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import os as os
from sklearn.metrics import mean_squared_error
from sklearn.metrics import mean_absolute_error
from sklearn.metrics import r2_score
from sklearn.model_selection import GridSearchCV
from numpy import arange
%matplotlib inline 
import sys
pd.set_option('display.max_colwidth', 1000)
pd.set_option('display.max_rows', 1000)

Read in AmesHousing.csv into a Dataframe Called Ames

#import data
Ames = pd.read_csv("AmesHousing.csv", sep=",", engine='python')
Ames.shape

Explore the Dimensions and Structures of the Dataframe

Ames.head()

#Check data types
Ames.dtypes.head()

#Remove the ID fields
Ames=Ames.drop(['Order', 'PID'],axis=1)
Ames.head()

# designate target variable name and move to front of dataframe
targetName = 'SalePrice'
targetSeries = Ames[targetName]
del Ames[targetName]
Ames.insert(0, targetName, targetSeries)
Ames.head()

Ames.describe()

Replace missing number values with variable mean

Replace missing object values with 'NONE'

Ames.isna().any().head()

#Fill NaN values of number variables with mean
for col in Ames.columns[1:]:
    dType = Ames[col].dtype
    if dType == float or dType == int:
        Ames[col].fillna(value=Ames[col].mean(), inplace=True)
    else:
        Ames[col].fillna(value='NONE', inplace=True)

Ames.isna().any().head()

#Success!