microsoft/onnxruntime

Pull Requests Commits

fs-eire committed 1 year ago

d589677d

[TensorRT EP] Fix bug for DDS output handling for empty tensor (#19575)

chilo-ms committed 1 year ago

Verified d9730c7f

Implement CUDA IsInf-10,20 (#19772)

yuslepukhin committed 1 year ago

Verified 1e78bcea

Adding cuda kernel (optimized for sm80) for block-wise 4b quantized float 16 GEMM. (#18619)

chenfucn committed 1 year ago

Verified 06e684c9

Fix CUDA BatchNorm bugs and add support for NHWC (#19742)

mtavenrath committed 1 year ago

Verified bdf678df

enable embedding sparse optimization by default (#19714)

guyang3532 committed 1 year ago

Verified cd56ea4a

[quant] supports act_order inputs in Matmulnbits and new quantization algorithm "hqq" (#19106)

wejoncy committed 1 year ago

Verified 7e613ee8

Zhijxu/fix conv1d replacement (#19758)

zhijxu-MS committed 1 year ago

Verified 2a5c9b86

Expose SessionOtions.DisablePerSessionThreads (#19730)

yuslepukhin committed 1 year ago

Verified 0cdf36fa

[DML] MatrixMultiplyIntegerToFloat (#19608)

AnaghaRaoAMD committed 1 year ago

Verified 27b1dc91

fix split shape inference error for opset >= 13 (#19756)

inisis committed 1 year ago

Verified 2e13d5f0

Benchmark - Updating llama-2 requirement files (#19716)

ironmangithub committed 1 year ago

Verified 9acaf534

Update copying API header files (#19736)

mszhanyi committed 1 year ago

Verified 9460597b

[QNN Quant] Add preprocessing option to transpose graph inputs/outputs to channel-last (#19731)

adrianlizarraga committed 1 year ago

Verified 2d79052e

[WebNN EP] Add contraints for MatMul (#19713)

zesongw committed 1 year ago

Verified de3158e7

Enable CPUINFO for all Windows build (#19655)

Changming Sun committed 1 year ago

Verified a0521f89

[js/web] transfer input buffer back to caller thread (#19677)

fs-eire committed 1 year ago

Verified f06164ef

disable gemm f16 on CPU (#19744)

yufenglee committed 1 year ago

Verified 22176a5f

Update google benchmark to 1.8.3. (#19734)

edgchen1 committed 1 year ago

Verified 5672cdeb

Change webgpu CI pipeline to use a preinstalled chrome (#19729)

Changming Sun committed 1 year ago

Verified ed550b5f

Follow up fix for Gelu impl (#19693)

pengwa committed 1 year ago

Verified acbfc29f

Add ML Program support for more operators (#19527)

skottmckay committed 1 year ago

Verified 2a857d9a

CUDA Resize-18 implementation (#19595)

yuslepukhin committed 1 year ago

Verified 5ee62a6b

Introducing customizable input names for loss in generate_artifacts. (#19705)

AdamLouly committed 1 year ago

Verified d5606cd7

Parallel Transpose_BSNH_to_BNSH (#19406)

yihonglyu committed 1 year ago

Verified ec0e4d3b

[ORTMODULE] Support Register Custom Triton Kernel (#19690)

Lafi7e committed 1 year ago

Verified 937cdd65

[ROCm] Update dockerfile (#19661)

PeixuanZuo committed 1 year ago

Verified c311d1fa

[QNN Quant] Ensure 16bit tensor quant overrides set MS domain (#19684)

adrianlizarraga committed 1 year ago

Verified c1bf7fcd

Merge GatherToSplitFusion and #19218 to a General Fusion (#19600)

Lafi7e committed 1 year ago

Verified d2e6dd25

Update labeler.yml to change permissions (#19709)

sophies927 committed 1 year ago

Verified 7455dd1f

Older