Hi All, I found that LogisticRegressionWithLBFGS interface is not consistent with LogisticRegressionWithSGD in master and 1.1 release.
https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/mllib/classification/LogisticRegression.scala#L199 In the above code snippet, users can only construct a LogisticRegressionWithLBFGS without any user specified parameters. Although users can specific training parameters calling corresponding function of class LBFGS. But this behave different with LogisticRegressionWithSGD. Could anyone can tell me why we did not refactor the code to keep consistent interface? Thank you Yanbo