onnxruntime/onnxruntime
mindest 4ffc1ff3b4
DMMHA: add unit tests; fix CPU, CUDA kernel (#22567)
### Description

Fixes:
(1) cpu kernel: applying scale before bias and mask like other MHA ops
(2) cpu kernel: correct offset during appending past to present.
(3) cuda kernel: apply mask if provided; fix output_qk offset.

Add DMMHA unit tests
2024-11-02 21:05:56 +08:00
..
contrib_ops DMMHA: add unit tests; fix CPU, CUDA kernel (#22567) 2024-11-02 21:05:56 +08:00
core DMMHA: add unit tests; fix CPU, CUDA kernel (#22567) 2024-11-02 21:05:56 +08:00
lora Accomodate BE platforms. Make sure we always write flatbuffers LE (#22375) 2024-10-11 09:14:44 -07:00
python Fix too strict assert in onnx_quantizer.py (#22283) 2024-11-01 19:08:35 -07:00
test DMMHA: add unit tests; fix CPU, CUDA kernel (#22567) 2024-11-02 21:05:56 +08:00
tool/etw Fix typos - 1st Wave (#21278) 2024-07-11 13:35:08 +08:00
wasm [WebNN EP] Add cache for MLContexts in the WebNNBackend (#22510) 2024-10-30 10:26:33 -07:00
__init__.py bumps up version in main from 1.20 -> 1.21 (#22482) 2024-10-17 12:32:35 -07:00
ReformatSource.ps1
ReformatSourcePython.bat
VSCodeCoverage.runsettings