microsoft/onnxruntime

Pull Requests Commits

Fix assert that the tensor's device type is 'cpu' #7248

ashari4 committed 4 years ago

Verified 56b22c1c

Add cuda kernels for GreaterOrEqual, LessOrEqual, Where; modify Clip to avoid memcpy (#7187)

ashbhandare committed 4 years ago

Verified e9ffcfa2

Update test_training_model.onnx to opset 12. (#7251)

Derek Murray committed 4 years ago

Verified c85657cf

MLAS: fix qgemm bus error with Android + ARM32 (#7250)

tracysh committed 4 years ago

Verified a9dbb511

Mem trt (#6868)

oliviajain committed 4 years ago

Verified fb40602e

Cleanup build.py (#7245)

Changming Sun committed 4 years ago

Verified 2fcd69d6

Update ContribOperators.md (#7246)

Changming Sun committed 4 years ago

Verified 5bd192c4

Make dW optional for convgrad (#7083)

pranav-prakash committed 4 years ago

Verified 3b16afc0

Update the build script for Android AAR package (#7229)

guoyu-wang committed 4 years ago

Verified c5973fbb

Add BERT-L perf regression test on MI100 and re-enable batch size test (#7240)

Suffian Khan committed 4 years ago

Verified 9f14af98

Add better model test error messaging (#7239)

Ryan Lai committed 4 years ago

Verified 10102c09

Fix Zip-Nuget-Java Packaging Pipeline (#7208)

askhade committed 4 years ago

Verified e7c5dcd5

Add detailed assertion error message (#7232)

jcwchen committed 4 years ago

Verified 3ee9b0ec

Update README.md (#7043)

mrshu committed 4 years ago

Verified 008065aa

Div mul fusion (#7183)

ashbhandare committed 4 years ago

Verified 2b851353

rename cuda_mem_limit and hip_mem_limit to gpu_mem_limit for both CUDA EP and ROCm EP (#7226)

weixingzhang committed 4 years ago

Verified 74ee24cf

Support for saving and loading pytorch compatible state dictionaries (#7220)

baijumeswani committed 4 years ago

Verified 68b12a61

handle optional input in quant topo sort (#7223)

yufenglee committed 4 years ago

Verified 8d737f97

HSA_NO_SCRATCH_RECLAIM and RCCL_ALLTOALL_KERNEL_DISABLE are not needed for ROCm 4.1 (#7224)

weixingzhang committed 4 years ago

Verified 59b57d83

move trt_profile to TensorrtFuncState and reuse it (#7195)

AhmadZakaria committed 4 years ago

Verified ba5f056b

enable more unit tests for ROCM EP (#7222)

weixingzhang committed 4 years ago

Verified ef88dc91

[NNAPI/CoreML EP] Add Onnx opset 14 support (#7211)

guoyu-wang committed 4 years ago

Verified afbbeaa3

Enable saving optimized models in OrtModule (#7214)

SherlockNoMad committed 4 years ago

Verified a98c2ebb

Add cupti path for python gpu packaging pipeline (#7200)

RandySheriffH committed 4 years ago

Verified ebde3209

Support SkipLayerNorm for ROCm EP (#7210)

weixingzhang committed 4 years ago

Verified 2d352056

update lamb and GatherGrad kernel for ROCm EP (#7184)

weixingzhang committed 4 years ago

Verified a3f17c8b

remove un-needed header file. (#7193)

weixingzhang committed 4 years ago

Verified 17f91ff4

Make IDataTransfer be directly shared with shared providers (#7215)

RyanUnderhill committed 4 years ago

Verified 5a6d4776

Check kernel def hashes (#7120)

edgchen1 committed 4 years ago

Verified 0ebeaf52

Make output 1 of ConcatTraining Optional and place on CPU (#7199)

ashbhandare committed 4 years ago

Verified 15c67ddb

Newer Older