ridgecv = linear_model.RidgeCV(alphas=(1e-10, 1e-9, 1e-8, 1e-7, 1e-6, 1e-5, 1e-4, 1e-3,1e-2,1e-1, 1, 10, 100, 1000, 10000),
                               normalize=True, fit_intercept=True)
ridgecv.fit(X=df[features], y=df[target])
weights = [(feat, ridgecv.coef_[i]) for i,feat in enumerate(features)]
print (len(weights), len(features))
selected_weights = [w for w in weights if abs(w[1]) > 0]
print ('Selected weights ', len(selecte_weights))

predicted1 = ridgecv.predict(df[features])
predicted2 = np.sum(df[w[0]]*w[1] for w in selecte_weights) + ridgecv.intercept_