microsoft/onnxruntime

Pull Requests Commits

Disable DML command list reuse for Xbox (#12063)

jstoecker committed 3 years ago

Verified 57ac3d0a

Eager mode: structure for supporting out= operators (#12066)

jamill committed 3 years ago

Verified ae88f435

remove --disable_iobinding for trt ep benchmark (#12053)

wangyems committed 3 years ago

Verified 8dc8d440

Make C# runtest.sh automatically set latest opset (#12039)

garymm committed 3 years ago

Verified 043816f8

Fix DML custom operators which set descriptor heap to command list (#12059)

jeffbloo committed 3 years ago

Verified 02b9b121

Add targets files for new .net6 frameworks (#12016)

skottmckay committed 3 years ago

Verified bfe1eca1

Eager mode: implement resize_ operation (#12004)

jamill committed 3 years ago

Verified 3e6b8d15

Extend lifetime of KernelDef when creating a standalone op (#12057)

RandySheriffH committed 3 years ago

Verified b858c2f7

Skip Constant Folding for ops producing an optional type output (#11839)

hariharans29 committed 3 years ago

Verified 2e27a7e3

Add eager support for aten:: equal. (#12020)

WilBrady committed 3 years ago

Verified 0fa2041f

Disable sequence-type tests since C# infra doesn't support well (#12037)

wschin committed 3 years ago

Verified 0ee0b8cf

Add FastGelu to kernel explorer for profiling. (#11995)

zhangyaobit committed 3 years ago

Verified da133ad3

Fix windows eager build break by pinning to torch version 1.11.0 (#12033)

WilBrady committed 3 years ago

Verified fdf12a5c

FP16_Optimizer Support for more Deepspeed Versions (#12046)

Lafi7e committed 3 years ago

Verified 04f7c2de

Move beamsearch shared initializers from subgraphs to main graph (#12025)

tianleiwu committed 3 years ago

Verified ecca6f4d

resolve comments

zhijxu-MS committed 3 years ago

9f260fb6

resolve comments

zhijxu committed 3 years ago

100aebbd

support optimizer opt for deepspeed 0.5.9

zhijxu-MS committed 3 years ago

2295b24c

update roialign cuda impl to onnx opset16 (#12036)

jywu-msft committed 3 years ago

Verified 102d01b2

Resize optimization for all architectures (#11956)

yihonglyu committed 3 years ago

Verified c8cd36da

Add warning about future computation change for ConvTranspose with auto_pad (#11984)

jcwchen committed 3 years ago

Verified 4eb54ff9

[TVM] handshake mechanism for support of TVMso EP (#11437)

vvchernov committed 3 years ago

Verified 8ba81466

Bump electron from 13.6.6 to 15.5.5 in /js/web (#11884)

dependabot[bot] committed 3 years ago

Verified c0dd9be7

[js/web][bugfix] fix negative axes for unsqueeze (#11944)

Yosshi999 committed 3 years ago

Verified 0702364d

convert_beam_search supports large gpt2 model (#11989)

tianleiwu committed 3 years ago

Verified 9be2b604

[DML EP] Pad operator: Handle negative pad counts (#11974)

sumitsays committed 3 years ago

Verified 4552dd38

Generalize native op creation (#11539)

RandySheriffH committed 3 years ago

Verified d5fcb432

DML EP ResNet50 opset 15 fails in ONNX checker for FusedBatchNormalization lacking training_mode attribute (#12010)

fdwr committed 3 years ago

Verified fc0143fe

[NNAPI EP] Update NNAPI headers (#11954)

edgchen1 committed 3 years ago

Verified f0459943

[C# Tests] Add support for double tensor output in TestPreTrainedModels. (#12008)

edgchen1 committed 3 years ago

Verified 466b2d9f

Newer Older