microsoft/onnxruntime

Pull Requests Commits

fs-eire committed 1 year ago

11892b50

Introduce custom external data loader

fs-eire committed 1 year ago

ca5c3d6d

Restructure MacOS framework package to fix malformed Framework errors (#21536)

vraspar committed 1 year ago

Verified 88c811b6

Fail tests gracefully for the minimal cuda build (#21391)

poweiw committed 1 year ago

Verified 2653226e

[WebNN EP] Support Dropout op (#21586)

Honry committed 1 year ago

Verified 8c641d71

[MIGraphX EP] Set External Data Path (#21598)

TedThemistokleous committed 1 year ago

Verified 45b7c41e

Security fuzz address sanitizer fix Bug (continue) (#21579)

tianleiwu committed 1 year ago

Verified 54d6614a

Adding CUDNN Frontend and use for CUDA NN Convolution (#19470)

JTischbein committed 1 year ago

Verified 13913542

[QNN EP] Support Conv + Clip/Relu fusion (#21537)

adrianlizarraga committed 1 year ago

Verified 0e708de4

Mlas int4 int8 with avx2/512 (#20687)

liqunfu committed 1 year ago

Verified b87e8edb

Add reduce kernels for bigger types (#21490)

neNasko1 committed 1 year ago

Verified d0a6f57d

[WebNN EP] Create MLGraphBuilder for every model builder (#21514)

Honry committed 1 year ago

Verified 8c2ee7b3

Bump torch from 1.13.1 to 2.2.0 in /tools/ci_build/github/windows/eager (#21505)

dependabot[bot] committed 1 year ago

Verified 3b73ef2b

Add CUDA custom op header files to Linux tarball (#21551)

Changming Sun committed 1 year ago

Verified 25722bb9

[QNN EP] Improve INT4 accuracy (#21582)

adrianlizarraga committed 1 year ago

Verified 4b8f6dcb

Fix quant_format argument for 4bit quantizer (#21581)

fajin-corp committed 1 year ago

Verified 8540ac4f

[WebNN EP] Fixed bug in ConvTranspose (#21569)

Honry committed 1 year ago

Verified a3883af7

[CUDA] Fix MultiHeadAttention thread safe and bias support (#21498)

tianleiwu committed 1 year ago

Verified c5f83896

Fix ETW trace logging crash in multithreading situations (#21566)

smk2007 committed 1 year ago

Verified b341c44c

[WebNN EP] Support ConvTranspose for TFLite backend (#21291)

Honry committed 1 year ago

Verified 1d4b1611

Utilize ext data location to reduce qd matmul memory usage (#21451)

fajin-corp committed 1 year ago

Verified e7aa1160

Extend Pad Fusion for AveragePool (#21556)

sumitsays committed 1 year ago

Verified 1637f22d

Enable FP16 Clip and Handle Bias in FP16 Depthwise Conv (#21493)

yihonglyu committed 1 year ago

Verified 530a2d7b

Remove references to the outdated CUDA EP factory method (#21549)

Changming Sun committed 1 year ago

Verified 82036b04

CoreML: Add ML Program Split Op (#21456)

vraspar committed 1 year ago

Verified 07d3be5b

[TensorRT EP] Update TRT OSS Parser to 10.2 (#21552)

yf711 committed 1 year ago

Verified 5d78b9a1

Keep QDQ nodes w/ nonpositive scale around MaxPool (#21182)

mcollinswisc committed 1 year ago

Verified 8417c325

Update labeling bot (#21548)

sophies927 committed 1 year ago

Verified d9858149

Propagate NaNs in the CPU min and max operators (#21492)

adamreeve committed 1 year ago

Verified 7543dd04

ORT- OVEP 1.19 PR-follow up (#21546)

preetha-intel committed 1 year ago

Verified c39f1c4f

Older