What algorithm does ml.classification.LogisticRegression use in spark?

Question

The ml.classification.LogisticRegression document points to the logistic regression in spark.mllib, which says:

We implemented two algorithms to solve logistic regression: mini-batch gradient descent and L-BFGS. We recommend L-BFGS over mini-batch gradient descent for faster convergence.

So in spark.mllib there are LogisticRegressionWithLBFGS and LogisticRegressionWithSGD. Which one is used in ml.classification.LogisticRegression?

Answer 1

它使用来自微风库的LBFGS算法进行L2正则化，使用OWLQN（LBFGS的扩展）进行L1和弹性网络正则化