microsoft/onnxruntime

Pull Requests Commits

Not fuse SkipLayerNorm when add has initializer input (#5123)

wangyems committed 5 years ago

Verified 89509f25

csharp build documentation (#5121)

askhade committed 5 years ago

Verified cd56ab19

Bump node-fetch from 2.6.0 to 2.6.1 in /nodejs

dependabot[bot] committed 5 years ago

15d431f3

Handle dummy mask in Attention operators (#5108)

tianleiwu committed 5 years ago

Verified ccfbc563

Solve name conflict in TensorRT engine caching (#5128)

stevenlix committed 5 years ago

Verified c794c88a

Enable onnxruntime_perf_test for ORT minimal build (#5126)

guoyu-wang committed 5 years ago

Verified 51f3d3af

Remove SparseTensor support from minimal build. (#5114)

skottmckay committed 5 years ago

Verified 59ee8ffb

Support Tensorflow benchmarking and onnx export in transformers tool (#5068)

wangyems committed 5 years ago

Verified 879751f3

Update Linux GPU build pipelines to CUDA 10.2 (#5120)

Changming Sun committed 5 years ago

Verified c5efb008

skip tests when model opset > released opset (#5096)

askhade committed 5 years ago

Verified a8557b3f

Add dll probe path so that the right DirectML.dll is loaded while running C# tests (#5104)

hariharans29 committed 5 years ago

Verified 782ccff2

Use CMake built-in function to compare NCCL version (#5118)

wschin committed 5 years ago

Verified 5618b9dd

Add transformers tools to python package (#5090)

tianleiwu committed 5 years ago

Verified c5d4ae04

[TensorRT] Align naming convention and remove redundant code (#5094)

mosdav committed 5 years ago

Verified 61051396

CMake fixes/tweaks for minimal builds and MinSizeRel builds (#5112)

skottmckay committed 5 years ago

Verified fae5915d

Fix a path problem in Dockerfile.manylinux2014_cuda10_2 (#5106)

Changming Sun committed 5 years ago

Verified a5530358

Disable some tests (#5103)

Changming Sun committed 5 years ago

Verified 47554a04

Remove IDeviceAllocator class as it doesn't extend IAllocator in any way. (#5067)

RyanUnderhill committed 5 years ago

Verified 3207de27

Move ort flatbuffers header to use enum class instead of enum (#5105)

guoyu-wang committed 5 years ago

Verified 5b6643ce

Enable onnx_test_runner for ort format (#5100)

guoyu-wang committed 5 years ago

Verified 43306153

Add store builds to nuget packaging (#5040)

tiagoshibata committed 5 years ago

Verified 62848c4d

Support Send and Recv for old NCCL versions (#5097)

wschin committed 5 years ago

Verified 9ba56dcf

Add re2 to memory leak checker whitelist (#5101)

Changming Sun committed 5 years ago

Verified 09a6ce6b

Not to call NVTX when not available (#5095)

wschin committed 5 years ago

Verified 934f30fc

Fix a few more signed/unsigned warnings. (#5098)

skottmckay committed 5 years ago

Verified 4b7aa16e

PipelinesForCuda11Cudnn8 (#4938)

RandySheriffH committed 5 years ago

Verified 5e10cde0

unify error handling in pipeline transformer (#5039)

xzhu1900 committed 5 years ago

Verified a90fae8c

Fix typo in DML native method call from the C# API (#5083)

hariharans29 committed 5 years ago

Verified 61151af3

Store/containerized apps support (#4651)

tiagoshibata committed 5 years ago

Verified f7c3e4fa

Use manylinux2014 for Linux CPU build (#5091)

Changming Sun committed 5 years ago

Verified 924ecb06

Newer Older