microsoft/onnxruntime

Pull Requests Commits

Fix implicit-exception-spec-mismatch warning. (#7481) (#7483) (#7491)

chandru-r committed 4 years ago

Verified 39294a5e

Fix implicit-exception-spec-mismatch warning. (#7481)

chandru-r committed 4 years ago

Verified d9e8739e

Merge branch 'master' into eager_mode

souptc committed 4 years ago

7cc31bcc

Initial Objective-C API (#7366)

edgchen1 committed 4 years ago

Verified d21304ce

Add CMAKE_CUDA_ARCHITECTURES=52 to TensorRT CI pipelines (#7455)

Changming Sun committed 4 years ago

Verified 78e583d0

[wasm] fix memory info creation (#7461)

fs-eire committed 4 years ago

Verified c2418a1f

liqunfu committed 4 years ago

Verified 4cbd2cce

[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

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

Changming Sun committed 4 years ago

Verified b5592856

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

Add level two optimizations for constant propagation transformation. (#7410)

satyajandhyala committed 4 years ago

Verified 979d6315

TRT EP memory leak fix (#7415)

chilo-ms committed 4 years ago

Verified f1c3f3fc

[CoreML EP] Add Gemm/MatMul support (#7403)

guoyu-wang committed 4 years ago

Verified 043883b5

Older