Linreg

[aisthesis]

Simple tool for performing regularized linear regression.

[aisthesis]

Here's where I'm coming from on the formula: http://stackoverflow.com/questions/27476933/numpy-linear-regression-with-regularization

[aisthesis]

I'm pretty confident that the formula is correct. Here's my result using sklearn:

>>> from sklearn.linear_model import Ridge
>>> clf = Ridge(alpha=3., solver='svd')
>>> clf.fit(features['train'], labels['train'])
Ridge(alpha=3.0, copy_X=True, fit_intercept=True, max_iter=None,
   normalize=False, random_state=None, solver='svd', tol=0.001)
>>> clf.coef_
array([ 0.        ,  6.52208421,  3.82487688,  3.61985418,  2.18396316,
        2.10604481,  1.2807778 ,  1.28789997,  0.72492414])
>>> clf.intercept_
11.217589325366376
>>> predicted = clf.predict(features['test'])
>>> pn.learn.mse(predicted, labels['test'])
array([[ 7.14405231]])

[arkadasgupta]

i mostly have some minor comments. other than that it looks good to go

[arkadasgupta]

some replies on your comments. other than that good to go.