[GitHub] [incubator-mxnet] arcadiaphy commented on issue #15574: fix naive engine for multi-threaded inference

arcadiaphy commented on issue #15574: fix naive engine for multi-threaded inference URL: https://github.com/apache/incubator-mxnet/pull/15574#issuecomment-513525766 @anirudh2290 Yes, in order to prevent async call from being pushed, using a single req_completed_ is not enough. We can add t

[GitHub] [incubator-mxnet] arcadiaphy commented on issue #15574: fix naive engine for multi-threaded inference

arcadiaphy commented on issue #15574: fix naive engine for multi-threaded inference URL: https://github.com/apache/incubator-mxnet/pull/15574#issuecomment-513072650 @apeforest @anirudh2290 Do we really need `req_completed_` in naive engine? I'm inclined to remove it now. -

[GitHub] [incubator-mxnet] arcadiaphy commented on issue #15574: fix naive engine for multi-threaded inference

arcadiaphy commented on issue #15574: fix naive engine for multi-threaded inference URL: https://github.com/apache/incubator-mxnet/pull/15574#issuecomment-512649728 @apeforest @ZhennanQin @marcoabreu @anirudh2290 Multi-threaded inference is very common in deploying, so we need to have a