CODEWITHSUNDEEP
pythonmatplotliblinear-regression
celsowm
celsowm

Reputation: 404

Unexpected result trying to plot the regression line using axline without numpy

For studying purposes, I am trying to make a simple linear regression without external libs to calculate the slope and the intercept:

import matplotlib.pyplot as plt

x = [1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.1, 2.2]
y = [60,  62,  64,  66,  68,   70,  72, 74] 

n = len(x)
sx = sum(x)
sy = sum(y)
sxy = sum([x[i] * y[i] for i in range(n)])
sx2 = sum([x[i] ** 2 for i in range(n)])
 
b = (n * sxy - sx * sy) / (n * sx2 - sx ** 2)
a = (sy / n) - b * (sx / n)

def predict_peso(altura):
    return a + b * altura
 
altura_prev = 1.75
peso_prev = predict_peso(altura_prev)
plt.plot(altura_prev, peso_prev, marker="o", markeredgecolor="red",
markerfacecolor="green")

plt.scatter(x, y)
plt.show()

enter image description here

But when I try to draw the line using axline:

plt.axline(xy1=(0, a), slope=b, linestyle="--", color="k")

I got this result:

enter image description here

What can I do to fix that and draw the line properly?

Thanks in advance !

Upvotes: 0

Views: 97

Answers (1)

Michael Cao
Michael Cao

Reputation: 3609

The line is drawn correctly, but since you stated an xy1 point, the plot automatically zooms out to show it. We can return the view to the previous one by storing and then retrieving the xlim and ylim. Add this block after your plt.show()

x0, x1 = plt.xlim()
y0, y1 = plt.ylim()

plt.axline(xy1=(0, a), slope=b, linestyle="--", color="k")

plt.xlim(x0, x1)
plt.ylim(y0, y1)

Upvotes: 1

Related Questions