microsoft/onnxruntime

Pull Requests Commits

reenable quant model tests (#13871)

mszhanyi committed 3 years ago

Verified ae2a9373

[DML EP] Add SkipLayerNormalization (#13849)

PatriceVignola committed 3 years ago

Verified 96d8d2c2

Extend vocab padding for logits MatMul for fp16 GPT2 GreedySearch (#13842)

hariharans29 committed 3 years ago

Verified 004a1538

Fix hipify to avoid nccl_service.h: No such file or directory (#13852)

cloudhan committed 3 years ago

Verified f79d3818

Upgrade pybind11 (#13853)

Changming Sun committed 3 years ago

Verified d12521d7

Use CacheTask to Accelerate MacOS build (#13859)

mszhanyi committed 3 years ago

Verified 78d18fbf

skip resnet50-int8 model test in training (#13856)

mszhanyi committed 3 years ago

Verified d2188fbf

Enable nuget packages for on device training (#13637)

askhade committed 3 years ago

Verified 65201e47

LayerNorm test fix (#13840)

ffwejoncy committed 3 years ago

Verified d5574e69

Misc minor bug fixes in transformer kernels (#13780)

hariharans29 committed 3 years ago

Verified 5f4e0c95

fix a wrong assert condition in benchmark_helper (#13821)

mindest committed 3 years ago

Verified f34ebbc8

Fix invocation of GetInputMemoryType. (#13828)

Pranav Sharma committed 3 years ago

Verified 335b62bd

[DML EP] Add an implementation for NonZero (#13768)

PatriceVignola committed 3 years ago

Verified b53bbe73

[oneDNN] Implemented Concat Op (#13646)

gaziqbal committed 3 years ago

Verified b9702587

[DML EP] Add support for LayerNorm (scale == nullptr) != (bias == nullptr) (#13818)

PatriceVignola committed 3 years ago

Verified c2d08fd7

[DML EP] Add mixed datatype support for DML's LayerNorm contrib op (#13734)

PatriceVignola committed 3 years ago

Verified a0b470bc

[quick fix] Build onnxruntime under DISABLE_ABSEIL (#13799)

ffwejoncy committed 3 years ago

Verified 82d123b6

Improve dependency management (#13523)

Changming Sun committed 3 years ago

Verified 04900f96

[DML EP] Add DML implementation for BiasGelu (#13795)

PatriceVignola committed 3 years ago

Verified e9b92fdf

Fix C26436 prefast errors (#13774)

nums11 committed 3 years ago

Verified e0dcbc38

[DML EP] Upgrade DML to 1.10.0 (#13796)

PatriceVignola committed 3 years ago

Verified 4128e44b

skip quantized model C# tests on GPU (#13782)

mszhanyi committed 3 years ago

Verified 777c474f

Improve DORT document (#13790)

wschin committed 3 years ago

Verified 7df8f842

[js/rn] support load model from buffer on Android (#12676)

fs-eire committed 3 years ago

Verified 77c97b6f

[DORT] Catch up with yesterday's PyTorch change (#13779)

wschin committed 3 years ago

Verified 639d2856

Move a function call outside a loop in ORTModule (#13771)

xadupre committed 3 years ago

Verified 441b30b2

Add DML support to the transformers benchmark.py script (#13776)

PatriceVignola committed 3 years ago

Verified 08ed09d2

Move C/C++ deps' URLs to deps.txt (#13769)

Changming Sun committed 3 years ago

Verified 29ed8811

Support exteranl weights in DML execution provider (#13740)

jeffbloo committed 3 years ago

Verified 571dc5a1

Fallback Pow op in layer norm to FP32 in TRT to avoid overflow (#13639)

stevenlix committed 3 years ago

Verified ce0025d3

Newer Older