SimpleLinearRegression

 1 # coding: utf-8
 2 
 3 # In[3]:
 4 
 5 
 6 import numpy as np 
 7 import matplotlib.pyplot as plt
 8 
 9 def estimate_coefficients(x, y): 
10     # size of the dataset OR number of observations/points 
11     n = np.size(x) 
12   
13     # mean of x and y
14     # Since we are using numpy just calling mean on numpy is sufficient 
15     mean_x, mean_y = np.mean(x), np.mean(y) 
16   
17     # calculating cross-deviation and deviation about x 
18     SS_xy = np.sum(y*x - n*mean_y*mean_x) 
19     SS_xx = np.sum(x*x - n*mean_x*mean_x) 
20   
21     # calculating regression coefficients 
22     b_1 = SS_xy / SS_xx 
23     b_0 = mean_y - b_1*mean_x 
24   
25     return(b_0, b_1)
26 
27     # x,y are the location of points on graph
28     # color of the points change it to red blue orange play around
29 
30 
31 
32 def plot_regression_line(x, y, b): 
33     # plotting the points as per dataset on a graph
34     plt.scatter(x, y, color = "m",marker = "o", s = 30) 
35 
36     # predicted response vector 
37     y_pred = b[0] + b[1]*x 
38   
39     # plotting the regression line
40     plt.plot(x, y_pred, color = "g")
41   
42     # putting labels for x and y axis
43     plt.xlabel(‘Size‘) 
44     plt.ylabel(‘Cost‘) 
45   
46     # function to show plotted graph
47     plt.show()
48     
49 
50     
51 
52 
53 def main(): 
54     # Datasets which we create 
55     x = np.array([ 1,   2,   3,   4,   5,   6,   7,   8,    9,   10]) 
56     y = np.array([300, 350, 500, 700, 800, 850, 900, 900, 1000, 1200]) 
57   
58     # estimating coefficients 
59     b = estimate_coefficients(x, y) 
60     print("Estimated coefficients:\nb_0 = {} \nb_1 = {}".format(b[0], b[1])) 
61   
62     # plotting regression line 
63     plot_regression_line(x, y, b)
64 
65     
66 if __name__ == "__main__": 
67     main()