JuniorTheoryCommon

What does this program print?

Python

import numpy as np
from sklearn.linear_model import Lasso, Ridge

X = np.array([[1, 5], [2, 1], [3, 4], [4, 2], [5, 3]])
y = np.array([2, 4, 6, 8, 10])
ridge = Ridge(alpha=1.0).fit(X, y)
lasso = Lasso(alpha=1.0).fit(X, y)
print(int((ridge.coef_ == 0).sum()), int((lasso.coef_ == 0).sum()))

0 1

Expecting 0 0: the L1 penalty of Lasso sets the coefficient of the second column, which does not help predict y, exactly to zero.
Expecting 1 1: the L2 penalty of Ridge shrinks every coefficient toward zero but never makes one exactly zero.

What happens to the first coefficient of Lasso as alpha keeps growing?
Why should the features be scaled before either penalty is applied?

Back to questions Open topic in roadmap →Answer it yourself — Quiz

Back to questions

JuniorTheoryCommon

What does this program print?

Python

import numpy as np
from sklearn.linear_model import Lasso, Ridge

X = np.array([[1, 5], [2, 1], [3, 4], [4, 2], [5, 3]])
y = np.array([2, 4, 6, 8, 10])
ridge = Ridge(alpha=1.0).fit(X, y)
lasso = Lasso(alpha=1.0).fit(X, y)
print(int((ridge.coef_ == 0).sum()), int((lasso.coef_ == 0).sum()))

0 1

Expecting 0 0: the L1 penalty of Lasso sets the coefficient of the second column, which does not help predict y, exactly to zero.
Expecting 1 1: the L2 penalty of Ridge shrinks every coefficient toward zero but never makes one exactly zero.

What happens to the first coefficient of Lasso as alpha keeps growing?
Why should the features be scaled before either penalty is applied?

Back to questions Open topic in roadmap →Answer it yourself — Quiz