glm vs lasso v2.py

import numpy as np
from sklearn.linear_model import PoissonRegressor, Lasso, Ridge
import statsmodels.api as sm

X_array = np.asarray([[1, 2], [1, 3], [1, 4], [1, 3]])
y = np.asarray([2, 2, 3, 2])
Preg_alpha_1 = PoissonRegressor(alpha=1., fit_intercept=False).fit(X_array, y)
print('alpha 1 Poisson Reg', Preg_alpha_1.coef_)
Preg_alpha_2 = PoissonRegressor(alpha=2., fit_intercept=False).fit(X_array*4., y)
print('alpha 2 Poisson Reg', Preg_alpha_2.coef_)
Lreg_alpha_1 = Lasso(alpha=1., fit_intercept=False).fit(X_array, y)
print('alpha 1 Lasso OLS', Lreg_alpha_1.coef_)
Lreg_alpha_2 = Lasso(alpha=2., fit_intercept=False).fit(X_array/2, y)
print('alpha 2 Lasso OLS', Lreg_alpha_2.coef_)
Rreg_alpha_1 = Ridge(alpha=1., fit_intercept=False).fit(X_array, y)
print('alpha 1 Ridge OLS', Rreg_alpha_1.coef_)
Rreg_alpha_2 = Ridge(alpha=2., fit_intercept=False).fit(X_array*4, y)
print('alpha 1 Ridge OLS', Rreg_alpha_1.coef_)