microsoft/onnxruntime

Pull Requests Commits

QNN EP: offload Q/DQ at graph boundary to CPU EP. TODO: make it a provider option

adrianlizarraga committed 1 year ago

d78c4de3

Add qnn android package (#22296)

sheetalarkadam committed 1 year ago

Verified dd2ea846

Remove deprecation warnings for TensorRT 10.5 builds (#22374)

kevinch-nv committed 1 year ago

Verified 709368ea

Fix SAM2 benchmark script on cuda graph (#22377)

tianleiwu committed 1 year ago

Verified 2584d02c

Add SetEpDynamicOptions and remove workload_type from run/session options (#22282)

l31g committed 1 year ago

Verified 1bc546af

Upgrade absl to the latest released version (#22365)

Changming Sun committed 1 year ago

Verified 2bef89c1

Re-enable CUDA 12 python package test pipeline (#22370)

Changming Sun committed 1 year ago

Verified dcf1e0c3

Add conv fp16 kernel in xnnpack EP (#22301)

mszhanyi committed 1 year ago

Verified 25b1c38e

Fix security warning: use sha256 instead of md5 (#22373)

tianleiwu committed 1 year ago

Verified 2b0ea6c8

Fix EmbedLayerNormalization fusion related to LayerNormalization optional input (#22371)

tianleiwu committed 1 year ago

Verified 5cc25293

[CUDA] Add CUDA_VERSION and CUDNN_VERSION etc. arguments to Dockerfile.cuda (#22351)

tianleiwu committed 1 year ago

Verified 071b6078

Fix the QNN nuget package issue (#22358)

HectorSVC committed 1 year ago

Verified 3b00024b

Update manylinux version (#22355)

Changming Sun committed 1 year ago

Verified 9ee96311

Revert "ThreadPool: Spend less time busy waiting. (#21545)" (#22350)

Pranav Sharma committed 1 year ago

Verified c415991c

Initial WebGPU EP checkin (#22318)

fs-eire committed 1 year ago

Verified c5d28cac

Remove duplicate fp32_bits definition. (#22340)

edgchen1 committed 1 year ago

Verified bc84958d

[CUDA] Update Docker to use Ubuntu 24.04, cuda 12.6, cudnn 9.4 and python 3.12 (#22339)

tianleiwu committed 1 year ago

Verified 8595e56d

Stop publishing python 3.8/3.9 packages (#22343)

Changming Sun committed 1 year ago

Verified d9834096

Fix Memory Issue GQA CPU Rotary (#22290)

aciddelgado committed 1 year ago

Verified cc0193cd

[MIGraphX EP Support]Add rocm to transformers/benchmark.py script (#22299)

stsokolo committed 1 year ago

Verified efb8703a

Fix conversion of TensorData, TensorsData to json (#22166)

xadupre committed 1 year ago

Verified 407c1ab2

Fix warning when building on Windows (#22327)

skottmckay committed 1 year ago

Verified 280c013d

Increanse TensorRT tolerance from default 1e-5 to 1e-3 after TRT 10.4 (#22321)

jingyanwangms committed 1 year ago

Verified 5036e63d

[WebNN EP] Support Tile operator (#22148)

mingmingtasd committed 1 year ago

Verified 004bd36f

[VitisAI]Add interface to get configurations from Sessionoption (#22096)

chenduan-amd committed 1 year ago

Verified 98a75900

[JS/WebGPU] Fixed bugs in inputs validation of Resize (#21955)

Honry committed 1 year ago

Verified 39c8b375

Support onnx data types (bfloat16, float8) in python I/O binding APIs (#22306)

tianleiwu committed 1 year ago

Verified b5ef8555

[C#] Address Packaging pipeline failure (#22307)

yuslepukhin committed 1 year ago

Verified 96a1ce1c

Address leftover comments for Lora support (#22322)

yuslepukhin committed 1 year ago

Verified 9f3676bc

[PyBind] Expose enable_mem_arena property for SessionOptions (#22323)

yuslepukhin committed 1 year ago

Verified 0645ad19

Older