Linear Regression
In [1]: # import libraries
import numpy as np
import pandas as pd
from sklearn.linear_model import LinearRegression
from sklearn import metrics
In [2]: # Read the dataset from csv file
df1 = pd.read_csv("C:/Users/DELL/Desktop/LR MLR [Link]")
[Link]()
Out[2]: ID CRIM ZN INDUS CHAS NOX RM AGE DIS RAD TAX PTRATIO B LSTA
0 1 7.52601 0.0 18.10 0 0.713 6.417 98.3 2.1850 24 666 20.2 304.21 19.3
1 2 0.01360 75.0 4.00 0 0.410 5.888 47.6 7.3197 3 469 21.1 396.90 14.8
2 3 0.05789 12.5 6.07 0 0.409 5.878 21.4 6.4980 4 345 18.9 396.21 8.1
3 4 0.05646 0.0 12.83 0 0.437 6.232 53.7 5.0141 5 398 18.7 386.40 12.3
4 5 2.14918 0.0 19.58 0 0.871 5.709 98.5 1.6232 5 403 14.7 261.95 15.7
In [4]: # Features
X = [Link](['MEDV', 'ID'], axis = 1)
# Target
y = df1['MEDV']
# Create a model
lr = LinearRegression()
# Fit the model
[Link](X, y)
# make predictions
pred = [Link](X)
In [5]: print('Intercept:', lr.intercept_)
Intercept: 41.040928943858106
In [6]: coeff_df = [Link](lr.coef_, [Link], columns=['Coefficient'])
coeff_df
Out[6]: Coefficient
CRIM -0.106472
ZN 0.059236
INDUS 0.051623
CHAS 3.002702
NOX -18.613446
RM 2.993018
AGE 0.004273
DIS -1.432040
RAD 0.300060
TAX -0.013191
PTRATIO -0.916840
B 0.008668
LSTAT -0.529803
In [8]: print('Mean Absolute Error:', metrics.mean_absolute_error(y, pred))
print('Mean Squared Error:', metrics.mean_squared_error(y, pred))
print('Root Mean Squared Error:', [Link](metrics.mean_squared_error(y, pred)))
print('R2:', [Link](metrics.r2_score(y, pred)))
Mean Absolute Error: 3.2249579483229565
Mean Squared Error: 21.53219881234212
Root Mean Squared Error: 4.640280035982971
R2: 0.8512690936204091
In [9]: features = [[0.06, 14, 6, 0, 0.5, 6, 22, 7, 4, 340, 19, 400, 8]]
print('The predicted price:', [Link](features)[0])
The predicted price: 19.41888951456918
In [ ]:
