Multivariable Linear Regression Python | Multivariable Linear Regression Example | (Data Science ss : 10.5)

Multivariable Linear Regression Example

As we had see an example where we took single independent variable in our previous blog. Now lets take an example where we will take multiple independent variable & use your model to train & test them. Don’t worry I’ll not include math just practical example only.

Q.1) Let’s say you want an house in Mumbai but this time based on multiple factors you will buy the house such as House Area,Number of Bedrooms & Age of the house (How old that apartment is).

Key point : More Age of House Less the Price is & if the number of bedroom is high then the price should also be high.

Steps :

First of all go to my github & download the House_2.csv file.

https://github.com/Vegadhardik7/ALL_CSV

Save it in a suitable location & then open you jupyter notebook & let’s start to code

Remember after every step press shift+enter to execute that code.

Step1:

import pandas as pd

import numpy as np

import matplotlib.pyplot as plt

%matplotlib inline

Step2:

df=pd.read_csv(r"E:\MACHINE_LEARNING\House_2.csv")

df

output:

Step3:

#if you had noticed above we have NaN value which means that cell is empty

#Now let's see how to handle NaN values

#Let's insert the median value into the empty cells

med=df['Bedroom'].median()

med   

Output:

3.5

Step4:

#As you all know the number of the bedroom cant be 3.5 so we have to conver it into an integer

import math

bed=math.floor(med)

bed

Output:

3

Step5:

#Now we are filling all the NaN values with the median of that column

df.Bedroom=df.Bedroom.fillna(bed) #USe the fill NaN value permanently

df

Output:

Step6:

#Now formula : y=m1*x1+m2*x2+m3*x3+c

#m1,m2,m3         = Co-efficient

#Area,Bedroom,Age = Independent Variable(x) [Features]

#c                = Intercept

x=df[['Area','Bedroom','Age']]

y=df['Price']

from sklearn.linear_model import LinearRegression

reg=LinearRegression()

reg.fit(x,y) #.fit is used to train our Linear regression model

Output:

LinearRegression(copy_X=True, fit_intercept=True, 

n_jobs=None, normalize=False)

Step7:

reg.coef_                    # values of m1,m2,m3

Output:

array([ 1.66769396e+02,  6.57543763e+05, -1.04049744e+05])

Step8:

reg.intercept_               # values of c

Output:

1769010.7017902858

Step9:

reg.predict([[3450,3,5]])    #Area,Bedroom,Age       output is Price

Output:

array([3796747.68660065])

#Same this is done above this is just for your understanding

((1.66769396e+02)*(3450))+((6.57543763e+05)*3)+((-1.04049744e+05)*5)+ 1769010.7017902858

Output:

3796747.686990286

Here is the link of my Jupyter notebook go & check it out for you better understanding.

https://github.com/Vegadhardik7/ALL_CSV/blob/master/Multivariable_Linear_Regression_House_2.ipynb


Hey guys, do you found our content informative? please leave a comment below & don't forget to share it with your friends.