MoisesHer opened a new pull request #18023: GPU gemms true fp16 (#17466) URL: https://github.com/apache/incubator-mxnet/pull/18023 * Temporal solution for fp16 accumulation in Bert gemms * Resolve alpha/beta type issue * add documentation for env variable MXNET_FC_TRUE_FP16 * Improve description of env variable * Add unitest checking environment variable * keep pseudo-fp16 if architecture does not support Float16Compute * Fix cpplint ## Description ## This PR allows true FP16 in CUBLAS gemms if environment variable MXNET_FC_TRUE_FP16 is set to true ## Checklist ## ### Essentials ### Please feel free to remove inapplicable items for your PR. - [x] Changes are complete (i.e. I finished coding on this PR) - [x] All changes have test coverage: - Unit tests are added for small changes to verify correctness (e.g. adding a new operator): tests/python/gpu/test_gluon_gpu:test_gemms_true_fp16 - [x] Code is well-documented: - [x] To the best of my knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change ### Changes ### - [x] General half_precision Gemm: allows true fp16 if MXNET_FC_TRUE_FP16 is set - [x] Transformer: allows true fp16 if MXNET_FC_TRUE_FP16 is set ## Comments ##
---------------------------------------------------------------- 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
