This PR is mainly for demo. For your comments, I think it's necessary to expose the number of threads for parallel inference. The computation in an executor is parallelized itself. This gives another option for parallelization along with parallelism in an executor. It's hard for a system to figure. Users should try and decide.
[ Full content available at: https://github.com/apache/incubator-mxnet/pull/12456 ] This message was relayed via gitbox.apache.org for [email protected]
