Commits
Zhang Lei authored and GitHub committed 6efa9d9e106
Add more qordered int8 operators for CUDA provider (#12949) Attention, Quantize/Dequantize etc. Update QOrderedMatmul's schema, updated unittest. Verified test data for QOrdered Attention. Co-authored-by: Zhang Lei <phill.zhang@gmail.com> Co-authored-by: Lei Zhang <zhalei@microsoft.com>