microsoft/onnxruntime

Pull Requests Commits

add qdq support of (un)squeeze and GlobalAveragePool (#10721)

yufenglee committed 4 years ago

Verified 7ab0c607

Skip SetName test on inbox build (#10699)

nums11 committed 4 years ago

Verified 9ad95bf0

POWER10: QGEMM optimization (#10642)

RajalakshmiSR committed 4 years ago

Verified 5d8c5409

Add ability to save calibration augmented models through external data format when model size exceeds 2Gb. (#10695)

mfuntowicz committed 4 years ago

Verified e5c6dc1f

[TVM EP] support of TVM Virtual Machine (#10341)

vvchernov committed 4 years ago

Verified 62cc9815

[js] release pipeline for web and react native (#10656)

hanbitmyths committed 4 years ago

Verified a7f6442c

Convert ConvActivationFusion transformer to a selector action transformer. (#10687)

edgchen1 committed 4 years ago

Verified 9e7d7a9e

check gpt-2 graph in converting beam search (#10712)

tianleiwu committed 4 years ago

Verified fa9090f2

Fix race condition in CUDA, ROCm, and TensorRT EP GetKernelRegistry() implementations. (#10200)

edgchen1 committed 4 years ago

Verified d07a2377

Print tensor snippet in dumping node Inputs/Outputs to StdOut (#10707)

tianleiwu committed 4 years ago

Verified 2fb2dae4

Add microbench to benchmark single operators. (#10678)

zhangyaobit committed 4 years ago

Verified a7738b52

[NNAPI QDQ] Add QDQ Concat (#10666)

guoyu-wang committed 4 years ago

Verified 19464614

Fix reshape allowzero with unknowndim (#10665)

BowenBao committed 4 years ago

Verified 6448ca64

set qdq as the default static quantization format (#10684)

yufenglee committed 4 years ago

Verified f652f70d

Update nuget icon (#10672)

mszhanyi committed 4 years ago

Verified f1b6f0be

Conv node bug, cached state was incoherent (#10041)

RyanUnderhill committed 4 years ago

Verified c1cf16ed

Upgrade emsdk to 3.1.3 (#10577)

fs-eire committed 4 years ago

Verified f4b2d3af

replace std::numeric_limits<T> by cub::FpLimits<T> (#10703)

tianleiwu committed 4 years ago

Verified c51b500c

Strided Tensor Support for Eager Mode (#10578)

Lafi7e committed 4 years ago

Verified 9a22b5d2

[java] Changes OrtEnvironment so it can't be closed by users (#10670)

Craigacp committed 4 years ago

Verified f8566085

Fix CUDA 10.2 compile error due to inlined_containers.h inclusion (#10702)

yuslepukhin committed 4 years ago

Verified e23a2245

Fix VLOG?_DEFAULT macros usability. (#10568)

cloudhan committed 4 years ago

Verified 3243c957

Avoid clang-tidy crashing due to readability-static-accessed-through-instance check bug (#10690)

cloudhan committed 4 years ago

Verified d1b2fb15

Add support of generating symmetric/asymmetric tensor's range for calibration (#10663)

chilo-ms committed 4 years ago

Verified d2d22f21

[iOS Packaging] Add full ORT build iOS package. (#10626)

edgchen1 committed 4 years ago

Verified ffde44cd

Add optional optimizer to remove leftover Q->DQ pairs after all other QDQ processing has completed (#10659)

skottmckay committed 4 years ago

Verified 1f6d8248

Convert com.microsoft::ATen into org.pytorch.aten::ATen onnx op (#10060)

Thiago Crepaldi committed 4 years ago

Verified e788cc2a

[java] Adding the graph description to the exposed model metadata. (#10318)

Craigacp committed 4 years ago

Verified e47434ea

Fix unsqueeze for opset 13 for ReduceMean Grad (#10668)

harshithapv committed 4 years ago

Verified 037f08f1

Add ability to customize ORT_CXX_API_THROW (#10688)

RyanUnderhill committed 4 years ago

Verified eb116595

Newer Older