ptrendx commented on issue #17665: No speedup from using FP16 (4 times slower than PyTorch) URL: https://github.com/apache/incubator-mxnet/issues/17665#issuecomment-592646699 I believe dot ultimately calls this: https://github.com/apache/incubator-mxnet/blob/master/3rdparty/mshadow/mshadow/dot_engine-inl.h#L516 @oleg-trott About this point "For ConvNets, there is similarly little or no gain when using FP16 in MXNet (Unlike with PyTorch)" - do you have an example? We tested pretty much all of the networks from GluonCV and we do not see that behavior.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
