microsoft/onnxruntime

Pull Requests Commits

Create Android Package pipeline (#7295)

guoyu-wang committed 5 years ago

Verified fce67e2b

User/sheilk/add api usage telemetry (#7320)

smk2007 committed 5 years ago

Verified b7c89ce7

Fix split op in the way it deals with the optional input (#7302)

hariharans29 committed 5 years ago

Verified 4971310d

matmul in dnnl (#7311)

jeyblu committed 5 years ago

Verified 61ba9ac1

yolov3 accuracy (#7235)

sfatimar committed 5 years ago

Verified 21c282ed

Zhalei/mlas test (#7213)

zhanghuanrong committed 5 years ago

Verified 6334c292

enable more unit tests for ROCM EP (#7307)

weixingzhang committed 5 years ago

Verified 75c0192e

NCHWc: Support "sizes" argument for Resize transform (#7290)

tracysh committed 5 years ago

Verified f27f5afd

Add Optype to type mismatch message (#7305)

jingyanwangms committed 5 years ago

Verified 2edf2955

Better error message when ORTModule used with torch.DataParallel (#7287)

baijumeswani committed 5 years ago

Verified b221a4fd

Polish Lamb Kernel (#7299)

weixingzhang committed 5 years ago

Verified c22963c2

Improve logged message for nodes that are forced to execute on CPU rather than some other EP (usually CUDA) (#7297)

hariharans29 committed 5 years ago

Verified 711cc99f

Polish Adam kernel (#7294)

weixingzhang committed 5 years ago

Verified 8ad5007f

change half gemm to use compute_32f as default (#7253)

tianleiwu committed 5 years ago

Verified 274e2fea

Support transpose by merge Reshape etc into direct xint8 operators. (#7265)

zhanghuanrong committed 5 years ago

Verified a4fdb4db

Narrow profiling scope (#7281)

RandySheriffH committed 5 years ago

Verified 42051c91

Enable CoreML EP for minimal extended mode (#7266)

guoyu-wang committed 5 years ago

Verified 370f9b88

Add support to dynamic positional/keyword input for ORTModule (#7189)

Thiago Crepaldi committed 5 years ago

Verified 7b4362c2

Fix codeql java warning (#7280)

guoyu-wang committed 5 years ago

Verified 4969431e

[Symbolic shape infer] fix scalar shape in Expand (#7285)

KeDengMS committed 5 years ago

Verified 0d49e539

NCHWc: avoid buffer reordering around Add nodes (#7279)

tracysh committed 5 years ago

Verified bc6ef809

Enable symbolic shape inference in ORTModule (#7282)

ytaous committed 5 years ago

Verified e14b291c

Suppress tracer warnings from onnx export in ORTModule (#7221)

baijumeswani committed 5 years ago

Verified d272c843

[OpenVINO-EP] Enabling save/Load blob feature (#7054)

MaajidKhan committed 5 years ago

Verified 27e77890

Add QGEMM benchmark (#7268)

chenfucn committed 5 years ago

Verified def4cc09

Restrict ConvGrad to __CUDA_ARCH__>=700 (#7278)

SherlockNoMad committed 5 years ago

Verified aa2c4651

ConvGrad CUDA Kernel Bugfix (#7273)

Lafi7e committed 5 years ago

Verified beb299e1

Support eval mode and torch.no_grad context in ORTModule and restructure ortmodule.py (#7162)

baijumeswani committed 5 years ago

Verified 844361bc

fix for using tensorrt:20.12 base image (#7264)

jywu-msft committed 5 years ago

Verified 025abf99

CUDA ConvGrad Kernel (#7227)

SherlockNoMad committed 5 years ago

Verified 4bc17ca0

Newer Older