microsoft/onnxruntime

Pull Requests Commits

[CUDA] Special case for K==0 in CUDA MatMul (#21525)

yuslepukhin committed 1 year ago

Verified c2911bbb

Replace usage of jcenter in React Native build.gradle files (#21714)

skottmckay committed 1 year ago

Verified 6af5394b

Fix neural-speed ci failure (#21694)

liqunfu committed 1 year ago

Verified 34394297

[js/webgpu] fix two errors of attention operator (#21687)

xhcao committed 1 year ago

Verified 9c6ee89f

move the A100 stage to main build (#21722)

mszhanyi committed 1 year ago

Verified 6db3d63a

enable qnn python arm64ec packaging (#21575)

jywu-msft committed 1 year ago

Verified a8462ffb

[DML EP] Update DML to 1.15.1 (#21695)

sumitsays committed 1 year ago

Verified c5592fdc

Security Fuzz Test Fixes (#21608)

jingyanwangms committed 1 year ago

Verified 154084ef

Web CI: make multi-browser test job optional (#21669)

fs-eire committed 1 year ago

Verified 6ae7e02d

[TensorRT EP] No workspace size limit to TRT memory pool (#21643)

chilo-ms committed 1 year ago

Verified 2abebb2a

Enable exporting for inference when loading from buffer without behavior changes (#21601)

carzh committed 1 year ago

Verified eeef0c8a

[Quant tool]: Improve symmetric quantization range update for Relu/Clip (#21573)

f2013519 committed 1 year ago

Verified 37be90c9

[QNN Quant tool] Fix validation of per-channel overrides for models with external data (#21656)

adrianlizarraga committed 1 year ago

Verified 390f0fd8

[JS/WebGPU] Add Dequantizelinear operator (#21642)

satyajandhyala committed 1 year ago

Verified 51b20441

[TensorRT EP] Add null_ptr check to avoid crash when running session which was failed to generate trt_engine previously (#21621)

yf711 committed 1 year ago

Verified 906ae77e

fix handling of multiple QuantizeLinear nodes (#21675)

saurabhkale17 committed 1 year ago

Verified 88788474

When quantize 4bit mamtul, force upgrade onnx domain opset to 21 (#21693)

fajin-corp committed 1 year ago

Verified 53a66f4e

Fix wrong per-tensor quantized weight type for matmul (#21347)

duanshengliu committed 1 year ago

Verified c6a73def

[CPU EP] Add block quantized Gather contrib op (#21630)

fajin-corp committed 1 year ago

Verified f30581ed

Fuse Pad even if Cast is present in-between (#21640)

sumitsays committed 1 year ago

Verified 702b2e28

[js/web] update the build script for webgpu to enable model dump by default (#19707)

fs-eire committed 1 year ago

Verified e6e4047a

[js/web] allow relative path matching (#21657)

fs-eire committed 1 year ago

Verified f4ec8525

update pipeline list for run_CIs_for_external_pr.py (#21665)

fs-eire committed 1 year ago

Verified ae2b4d31

[CUDA] Fix MHA mask (#21655)

tianleiwu committed 1 year ago

Verified 9334d4e3

Use zipped xcframework in nuget package (#21663)

skottmckay committed 1 year ago

Verified 410ae94e

Unblock migraphx and linux GPU training ci pipelines (#21662)

tianleiwu committed 1 year ago

Verified a46e49b4

[js/web] allow op test to use f16 type for inputs/outputs (#21664)

fs-eire committed 1 year ago

Verified 5e66fcc7

Match changes in gh-pages PR (#21628)

skottmckay committed 1 year ago

Verified d6160258

fix wrong check for tree ensemble regressor (#21595)

zhangxiang1993 committed 1 year ago

Verified c93b92a4

Pin transformer and optimum version (#21650)

mszhanyi committed 1 year ago

Verified 621b16f4

Older