scikit-learn

import numpy as np
import matplotlib.pyplot as plt
import random, time, math

aa = np.array(range(10))
X = aa.reshape(len(aa), 1)
y = np.array(range(10))
print('\n', X, '\n\n', y)

# regression
from sklearn import linear_model
regr = linear_model.LinearRegression()
regr.fit(X, y)
print('\n', regr.coef_, regr.intercept_)

regr.fit(X, y*2)
print('\n', regr.coef_, regr.intercept_)

regr.fit(X, y+10)
print('\n', regr.coef_, regr.intercept_)


np.dot(X, regr.coef_)
X = np.array([10, 20, 30, 40]).reshape(4, 1)
np.dot(X, regr.coef_)
regr.predict(90)

# what happens with squaring
regr.fit(X, y**2)
print('\n', regr.coef_)
y2 = np.dot(X, regr.coef_)
# y2 is array([  0.,   9.,  18.,  27.,  36.,  45.,  54.,  63.,  72.,  81.])
# So lets plot the original and the fit (green)
plt.plot(aa, y**2), plt.plot(aa, y2),  plt.show()