microsoft/onnxruntime

Pull Requests Commits

rust bindings: Do not unnecessarily re-run build.rs (#17018)

j-baker committed 2 years ago

Verified 16eba537

Update cmake to 3.27 and upgrade Linux CUDA docker files from CentOS7 to UBI8 (#16856)

Changming Sun committed 2 years ago

Verified c6b0d185

[js/webgpu] Support slice int32 (#16968)

xhcao committed 2 years ago

Verified 026672e9

Fix insufficient space error in Android CI (#17423)

skottmckay committed 2 years ago

Verified e1a9f2ed

Reduce GPU memory for Whisper models converted to ONNX (#17378)

petermcaughan committed 2 years ago

Verified fa28359b

Introduce output type/shape validation (#17301)

yuslepukhin committed 2 years ago

Verified dbcc60be

Set proper nvcc threads to avoid OOM (#17419)

tianleiwu committed 2 years ago

Verified 8818a99c

Fix git working dir for ORT_BUILD_INFO (fixes #17197) (#17198)

1enn0 committed 2 years ago

Verified e3bb2a0c

Add ck's streamk and splitk gemm impl (#17280)

cloudhan committed 2 years ago

Verified 6ea3908d

[js/webgpu] Fix bug in conv2dByMatMul path (#17369)

qjia7 committed 2 years ago

Verified 5e747071

fix assert error in attention fusion script (#17375)

tianleiwu committed 2 years ago

Verified e7455751

output all parameters in the bert_perf_test tool (#17379)

tianleiwu committed 2 years ago

Verified e23f16ad

Change RuntimeError to ImportError (#17380)

baijumeswani committed 2 years ago

Verified 8b98ecad

Fix NNAPI optional input handling checks and unblock Android CI pipeline test failures (#17358)

YUNQIUGUO committed 2 years ago

Verified 16cfcd05

Flash Attention v2 MHA (#17227)

aciddelgado committed 2 years ago

Verified 44101e87

Refine build script for adding disable selected data types option (#17284)

YUNQIUGUO committed 2 years ago

Verified b5461950

Fix TRT EP's cuda graph feature (#17355)

chilo-ms committed 2 years ago

Verified 30a450dc

Change _mm512_setzero to _mm512_setzero_ps (#17362)

Changming Sun committed 2 years ago

Verified ae90b716

[WebNN EP] Fix compilation with newer flatbuffers (#17367)

Honry committed 2 years ago

Verified 3a538368

[js/webgpu] Adding support for abs with int32 type (#17359)

jchen351 committed 2 years ago

Verified e6049352

[js/webgpu] Add input/output shapes information to profiling (#17342)

qjia7 committed 2 years ago

Verified 352b745d

Fix a build warning: a constexpr function calls a non-constexpr function (#17363)

Changming Sun committed 2 years ago

Verified bbf28f09

Fuse ScaledSum and its backward BatchScale (#16517)

pengwa committed 2 years ago

Verified 58af36b4

Fix SkipLayerNorm fusion in transformer optimizer (#17320)

tianleiwu committed 2 years ago

Verified c11ed065

Add compiler cache in Linux GPU TensorRT CI. (#17348)

mszhanyi committed 2 years ago

Verified 507a40e1

Enable QDQ node unit support for Log op (#17354)

HectorSVC committed 2 years ago

Verified 47fe7fe9

only Flush once for the same stream in copyInputAcrossDevice() (#17303)

jslhcl committed 2 years ago

Verified 64f06d0b

[WebNN EP] Fix bug in interpreting ONNX's pads into WebNN's padding (#17325)

Honry committed 2 years ago

Verified 70e8c239

Fix a memleak in RunAsync python (#17326)

RandySheriffH committed 2 years ago

Verified 6c39641e

Update to nodejs version from 16 to 18.17.1 (#17351)

jchen351 committed 2 years ago

Verified 081c0692

Newer Older