Multi-class SVM with Spark 1.6?

Question

I ran a logical regression with multiple classes using Spark, but I would like to use SVM to cross-validate the results. It seems that Spark 1.6 only supports binary SVM classifications. Should I use other tools for this? H20, for example?

Answer 1

After some research, I found this branch, which was not integrated into Spark 1.6, which allowed me to run SVM on the problem of classifying several classes.

Many thanks to Bekbolatov.

: https://github.com/Bekbolatov/spark/commit/463d73323d5f08669d5ae85dc9791b036637c966