microsoft/onnxruntime

Pull Requests Commits

Uppdate ORT_VERSION path to separate current version from future version.

jchen351 committed 2 years ago

b03a1f50

Uppdate ORT_VERSION path

jchen351 committed 2 years ago

7faa6802

ORT_SYS_MANIFEST_PATH

jchen351 committed 2 years ago

53e39fea

remove just vendor step when we need a justfile

jchen351 committed 2 years ago

e01f9793

remove just vendor step when we need a justfile

jchen351 committed 2 years ago

d5f30226

update to install toolchain to nightly

jchen351 committed 2 years ago

db80598c

Update ORT_RUST_STRATEGY=download

jchen351 committed 2 years ago

1d794345

Merge branch 'main' into Cjian/rust_ci

jchen351 committed 2 years ago

4b9203de

[CoreML] Add Softmax and Split op support (#18358)

YUNQIUGUO committed 2 years ago

Verified 62f00ad8

[ROCm] Update ck for GemmFloat8 (#18487)

cloudhan committed 2 years ago

Verified 6f3c1f9d

[QNN EP] Support LpNormalization (#18561)

adrianlizarraga committed 2 years ago

Verified 1c79897c

Memory optimization refactor and refinement (#17481)

pengwa committed 2 years ago

Verified 43a5147e

[js/webgpu] Add BatchNormalization Op (#18468)

qjia7 committed 2 years ago

Verified 64dacc28

[js/webgpu] Refactor matmul conv to support uniforms for matmul (#18452)

axinging committed 2 years ago

Verified fa106942

Update transpose optimization to be more QDQ aware (#18444)

skottmckay committed 2 years ago

Verified 42c6799c

[[JS/Web]Added uniform to Expand op. (#18558)

satyajandhyala committed 2 years ago

Verified 841f7ed3

[JS/Web] Resize & BiasSplitGelu fp16 support (#18536)

dakenf committed 2 years ago

Verified 1c555c5f

Fix opset import in GemmFloat8 python unit tests (#18489)

xadupre committed 2 years ago

Verified 3f0ebd67

Fix opset version of the optimizer in function generate_artifacts (#18300)

xadupre committed 2 years ago

Verified 32fabb55

[WebNN EP] Mark and fallback unsupported op for WebNN CPU backend (#18472)

Honry committed 2 years ago

Verified 89723c86

[ORTModule] Adjust Attention Patterns for Efficient Attention ATen Fallback (#18471)

Lafi7e committed 2 years ago

Verified 3bc9efc7

[QDQ Optimizer] Fix logic that drops Q/DQ ops from QDQ split node groups (#18394)

adrianlizarraga committed 2 years ago

Verified 7c573054

SDXL Latent Consistency Model (LCM) optimization (#18526)

tianleiwu committed 2 years ago

Verified 62da3b1c

[js/web] use Chrome in CI for npm tests (#18522)

fs-eire committed 2 years ago

Verified d455b0f8

[js/webgpu] enable f16 for concat (#18528)

qjia7 committed 2 years ago

Verified ac8598a8

Make TensorShapeVector to use InlinedVector<Int64_t> to reduce on template instantiations (#18519)

yuslepukhin committed 2 years ago

Verified 81a763a9

Training packaging pipeline for cuda12 (#18524)

ajindal1 committed 2 years ago

Verified 680a526e

Hide NPU Adapter selection behind macro (#18515)

smk2007 committed 2 years ago

Verified 2a016225

Add missing flags DISABLE_FLOAT8_TYPES in GemmFloat8 custom operator for CUDA < 11.8 (#18162)

xadupre committed 2 years ago

Verified 29a409ac

fix past-kv in general LLM exporter (#18529)

wejoncy committed 2 years ago

Verified a608c002

Older