microsoft/onnxruntime

Pull Requests Commits

RandySheriffH committed 4 years ago

71edad5b

Extra test case for varying loop sizes

tlh20 committed 4 years ago

a853225b

tlh20 committed 4 years ago

1f374852

Remove unused code

tlh20 committed 4 years ago

6c4bce4a

tlh20 committed 4 years ago

37ea7aee

tlh20 committed 4 years ago

88f56297

tlh20 committed 4 years ago

d8717388

[JS] onnxruntime-web (#7394)

fs-eire committed 4 years ago

Verified 4ebc9c3b

NCHWc: ReorderInput improvements (#7442)

tracysh committed 4 years ago

Verified d13e5b2f

Partial graph execution perf improvements. (#7438)

codemzs committed 4 years ago

Verified 82108b18

Add pytorch version check before loading Python ONNX Runtime training module (#7377)

Thiago Crepaldi committed 4 years ago

Verified 0702a14e

Update build docker image cache cleanup build definition (#7452)

edgchen1 committed 4 years ago

Verified 4804ede5

Wait for dispatch done in RunParallelSection to fix random TP UT crash (#7443)

RandySheriffH committed 4 years ago

Verified 40568d88

Implement qlinear concat and unit test. (#7341)

zhanghuanrong committed 4 years ago

Verified ada0fbbd

Merge remote-tracking branch 'origin/master' into tiharr/dev-sticky-3

tlh20 committed 4 years ago

510eee7f

tlh20 committed 4 years ago

d8648562

Remove thread pool's cancel method and suppress some warnings (#7411)

Changming Sun committed 4 years ago

Verified b5592856

Merge remote-tracking branch 'origin/master' into tiharr/dev-sticky-3

tlh20 committed 4 years ago

b28596ab

SqueezeGrad Bugfix (#7412)

Lafi7e committed 4 years ago

Verified 368e4a32

Explicitly pass cuda stream to thrust function rather than use cuda default stream implicitly (#7414)

weixingzhang committed 4 years ago

Verified ca9b3f18

dnnl matmul tensor dimension check (#7383)

jeyblu committed 4 years ago

Verified b9cbbc41

Reduce perf gap between thread pool and omp (#7333)

RandySheriffH committed 4 years ago

Verified afe912d4

Add support for ORTModule to execute the graph when ONNX drops unused… (#7424)

Thiago Crepaldi committed 4 years ago

Verified 410a81b2

Add batch interface to floating point GEMM (#7323)

chenfucn committed 4 years ago

Verified f4f2cc1a

Add CI pipeline to publish Python training package targeting Rocm (#7417)

Suffian Khan committed 4 years ago

Verified 7a3c1787

Partial graph execution made simple. (#7324)

codemzs committed 4 years ago

Verified 34ebf7d3

Fix some warnings in our CUDA code (#7436)

Changming Sun committed 4 years ago

Verified 52082311

add gather elements (#7435)

Suffian Khan committed 4 years ago

Verified 8889e717

Build would fail when nccl is not under standard path (--nccl_home) (#7402)

weixingzhang committed 4 years ago

Verified ef727649

Revert the TRT change and move the build to a new pool (#7434)

Changming Sun committed 4 years ago

Verified 9f683bae

Older