microsoft/onnxruntime

Pull Requests Commits

dependabot[bot] committed 4 years ago

0b0e8ccf

Update our absl cmake files (#10762)

Changming Sun committed 4 years ago

Verified 283d0c47

Add micro-benchmark for FastGelu (#10744)

zhangyaobit committed 4 years ago

Verified 4c88fa59

upstream TVM. small code cleaning (#10515)

vvchernov committed 4 years ago

Verified 46d0b20a

[iOS packaging] Minor updates. (#10755)

edgchen1 committed 4 years ago

Verified 395a7242

Enable QDQ cleanup and NHWC optimizers in an extended minimal build. (#10729)

skottmckay committed 4 years ago

Verified e337f5fa

Pipeline changes to build full ORT package for Android (#10654)

guoyu-wang committed 4 years ago

Verified 7aa70685

Simplify QLinearConv registration so type reduction works with it. (#10747)

skottmckay committed 4 years ago

Verified 6072c6b6

Add an option to export ONNX graphs in ORTModule tests (#10579)

adk9 committed 4 years ago

Verified c2c85dd6

optimize web assembly build flags for multi-thread (#10759)

fs-eire committed 4 years ago

Verified 745fa588

Fix unused variable warning, move variable definitions closer to usages. (#10757)

edgchen1 committed 4 years ago

Verified c8ec7782

Change axis to 0D in cumsum tests. (#10715)

oliviajain committed 4 years ago

Verified ed87e1b7

A new pipeline to replace the existing WindowsAI packaging pipeline (#10646)

Changming Sun committed 4 years ago

Verified b3e96d61

Optimize BinaryElementWise and BiasGeluGrad kernels for AMD (#10594)

hubertlu-tw committed 4 years ago

Verified fe8d867e

Fix build with gcc 7.5 (#10567)

cloudhan committed 4 years ago

Verified 4c20f686

Add the missing status return in beam search (#10738)

feihugis committed 4 years ago

Verified 75160d67

Add option to force QDQIsInt8Allowed to return true when exporting to ORT format (#10719)

YUNQIUGUO committed 4 years ago

Verified a9dc50ba

Add restriction to first usage in allocation planner (#10724)

wangyems committed 4 years ago

Verified 44d08d80

Auto mixed precision conversion of GPT-2 onnx model (#10711)

tianleiwu committed 4 years ago

Verified 47ab0c20

add missing link to openvino (#10737)

oliviajain committed 4 years ago

Verified 7ebff2b2

orttraining packaging pipeline for rocm 5.0.1 (#10725)

baijumeswani committed 4 years ago

Verified f9b6eef0

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

Newer Older