microsoft/onnxruntime

Pull Requests Commits

Merge branch 'ThreadPoolProfiler2' into snnn/python1

RandySheriffH committed 4 years ago

c9c4890d

Changming Sun committed 4 years ago

73dd1b14

Changming Sun committed 4 years ago

cbc9c30d

RandyShuai committed 4 years ago

c4a2aca0

use == instead of is for python 3.8

yufenglee committed 4 years ago

a356d38d

Changming Sun committed 4 years ago

7d007df4

Changming Sun committed 4 years ago

f64f8678

Changming Sun committed 4 years ago

bd519de5

Changming Sun committed 4 years ago

cb32083a

add python 3.8/3.9

Changming Sun committed 4 years ago

13e7ecb0

Fix training python packaging pipeline (#6613)

Changming Sun committed 4 years ago

Verified e70344e6

Update a few WinML model test filters for DML

jstoecker committed 4 years ago

Verified 1c727742

Support disabling training kernels as part of a reduced build (#6557)

Cian Hayes committed 4 years ago

Verified 8f14b8bd

Add engine decryption in TensorRT EP (#6612)

stevenlix committed 4 years ago

Verified e9d03983

Update CUDA build configs (#6598)

Changming Sun committed 4 years ago

Verified 0b89f931

Support double for operators ReduceMax, ReduceMin (#6265)

xadupre committed 4 years ago

Verified d3a2c8c1

enable omp for debug build

RandyShuai committed 4 years ago

ff063309

deprecate omp in ci

RandyShuai committed 4 years ago

6c5f50d0

Quantize model with QDQ format (#6541)

yufenglee committed 4 years ago

Verified 56e4e47f

Make kernel hash stable in type reduced build (#6603)

skottmckay committed 4 years ago

Verified c02ae61c

Fix layer_norm.cc on x86 (#6556)

Cian Hayes committed 4 years ago

Verified 16eed68a

Don't update the excluded ops/types unless args.update is true. Updating the exclusion info triggers rebuilding of all kernels using type reduction. (#6604)

skottmckay committed 4 years ago

Verified 13d7db9a

Fix bug with ORT format serialization of tensor attributes. (#6602)

skottmckay committed 4 years ago

Verified 0b1e21c6

[Mult-GPU inferencing] Add new API to get/set device id. Set correct device id in cuda allocator. (#6592)

Pranav Sharma committed 5 years ago

Verified 67ef6b1a

Fix TensorRT unnecessary file cache operations (#6601)

nietras committed 5 years ago

Verified 1dd920fa

Reduce CastMLFloat16ThroughFloat size (Scott's suggested changes), fix unused function warning. (#6597)

edgchen1 committed 5 years ago

Verified 19c130f5

Fix some coding conventions issues (#6583)

skottmckay committed 5 years ago

Verified 190b90a6

Generate error when an explicit stream argument is not provided in the <<<...>>> kernel launch syntax (#6599)

weixingzhang committed 5 years ago

Verified c86c21e0

Enable more ROCM ops that are sharing CUDA code. Some are needed for Turing NLG models.

jessebenson committed 5 years ago

d18aa45b

Fix build.gradle so it always targets Java 8 class files.

Craigacp committed 5 years ago

dbe31361

Older