szha commented on issue #11796: Batch_dot does not support FP16 well URL: https://github.com/apache/incubator-mxnet/issues/11796#issuecomment-405804549 Oops, wrong button. Relevant links: https://github.com/apache/incubator-mxnet/blob/master/src/operator/tensor/dot-inl.h#L1347-L1364 https://github.com/dmlc/mshadow/blob/master/mshadow/dot_engine-inl.h#L528-L539. While for float the strided gemm is used, the half_t type is calling regular gemm. Instead, the strided gemm in cublas can be used which supports half_t: https://docs.nvidia.com/cuda/cublas/#cublas-lt-t-gt-gemmbatched
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on 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
