microsoft/onnxruntime

Pull Requests Commits

add a way in JS to query memory usage info

fs-eire committed 1 year ago

a81572de

Pin transformer and optimum version (#21650)

mszhanyi committed 1 year ago

Verified 621b16f4

Improve speed in combining per-channel data (#21563)

duanshengliu committed 1 year ago

Verified b95aa056

Fix usability checker CoreML config file path. (#21626)

edgchen1 committed 1 year ago

Verified 4ad87ca2

[QNN EP] Update QNN SDK to 2.25 (#21623)

adrianlizarraga committed 1 year ago

Verified 0acefc79

Fix docker image layer caching to avoid redundant docker building and transient connection exceptions. (#21612)

mszhanyi committed 1 year ago

Verified 0d1da41c

Fix typos so to call correct vnni functions under vnni condition (#21625)

liqunfu committed 1 year ago

Verified f6f9657f

[EP Perf] Update cmake (#21624)

yf711 committed 1 year ago

Verified 1f907a23

Clean up some mobile package related files and their usages. (#21606)

edgchen1 committed 1 year ago

Verified a5ce65d8

Updates to apple packaging (#21611)

skottmckay committed 1 year ago

Verified bcc01ac1

bumps up version in main from 1.19 -> 1.20 (#21588)

prathikr committed 1 year ago

Verified 134f4774

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

Older