microsoft/onnxruntime

Pull Requests Commits

C# ResNet50 v2 sample/tutorial (#4722)

marcusturewicz committed 5 years ago

Verified 37c45c3d

wangyems committed 5 years ago

Verified 61726e58

Remove docstring for removed parameter (#4734)

Sergii Dymchenko committed 5 years ago

Verified c334b573

Add the framework to support prepack (#4413)

yufenglee committed 5 years ago

Verified b22091dc

Support log sigmoid gradient (#4719)

zhijxu-MS committed 5 years ago

Verified 33fe7700

Revert "Remove code which is not thread-safe. (#4454)" (#4712)

wschin committed 5 years ago

Verified 7905c57f

Shape independent gradient builder for Concat (#4675)

ashbhandare committed 5 years ago

Verified fc2f36c6

[Android NNAPI EP] Enable test for BatchNormalization, enable dev_mode for Android, fix some issues in concat (#4715)

guoyu-wang committed 5 years ago

Verified 8507bc1f

Wire log(softmax) grad cuda kernel and add log(softmax) grad cpu kernel (#4726)

suffiank committed 5 years ago

Verified 4d39c6a6

ReshapeGrad optimization (#4708)

KeDengMS committed 5 years ago

Verified 9a73c8f4

Add initial Dockerfile for distributed training targets (#4578)

suffiank committed 5 years ago

Verified 005fa5c3

EnrichPyOpUT (#4681)

RandySheriffH committed 5 years ago

Verified e802b049

[Nuphar] added Gemm-to-MatMul conversion in model editor (#4691)

yangchen-MS committed 5 years ago

Verified 43142a82

Add .NET Core 3.0 nuget e2e pipeline tests (#4695)

smk2007 committed 5 years ago

Verified 5c5efa90

Localized Recompute for Gelu and AttentionDropout (#4402)

SherlockNoMad committed 5 years ago

Verified eb0f57f0

[Android NNAPI EP] change most of the exceptions to return Status (#4701)

guoyu-wang committed 5 years ago

Verified 0933148f

Remove the requirement of CUDA's version.txt (#4706)

Changming Sun committed 5 years ago

Verified 1e054739

Enable MatMul + Scale fusion (#4669)

edgchen1 committed 5 years ago

Verified 9d7284fc

Log telemetry for WinML Native API for setting intra op num usage (#4700)

Ryan Lai committed 5 years ago

Verified f9bd52f8

Stress-test and fix thread pool when work queues are full (#4690)

tlh20 committed 5 years ago

Verified 4bd9e8d0

Add 'Install ONNX' step to Windows GPU pipeline (#4696)

Changming Sun committed 5 years ago

Verified d0297f8d

Support int32 and int64 types for Tile CUDA kernel (#4684)

hariharans29 committed 5 years ago

Verified 49febba3

Add Named Dimension Override API to LearningModelSessionOptions (WinML) (#4606)

Ori Levari committed 5 years ago

Verified e6ef3653

resolves #3101 - fix nuget package restore for sdk-style projects (#4680)

yuslepukhin committed 5 years ago

Verified bb9b452a

Support easy way to request verbose logging in test runner (#4676)

hariharans29 committed 5 years ago

Verified 0828a900

Avoid building ONNX of every history ONNX versions in our CI (#4678)

Changming Sun committed 5 years ago

Verified 01ca6392

refine machine_info and output onnxruntime_tools version (#4679)

tianleiwu committed 5 years ago

Verified e70e9e2f

Added Dockerfile and build instructions for Jetson. Also set CUDA arch set automatically. (#4637)

borisfom committed 5 years ago

Verified 6958f49d

Support distill-bert fusion in transformers tool (#4631)

wangyems committed 5 years ago

Verified b1bfff34

Modify EmbedLayerNorm to support distill-bert (#4666)

wangyems committed 5 years ago

Verified 8cf2c1c4

Newer Older