Add calibrator option to specify dtypes#19

Draft

mgleonard425 wants to merge 59 commits intomainfrom

mike-int64-tranges

Commits on Aug 28, 2023

Cherry-pick 1st Round (#17308)
Lafi7e
authored

Commits on Sep 7, 2023

Cherry-pick 2nd Round (#17386)
Lafi7e
authored

Commits on Sep 8, 2023

[rel-1.16.0] Use name of temporary provisioning profile. (#17456)
edgchen1
authored
[rel-1.16.0] Disable QNN QDQ test for release branch (#17463)
HectorSVC
authored
Remove 52 from CMAKE_CUDA_ARCHITECTURES to reduce Nuget package size (#17461)
snnn
authored

Commits on Sep 12, 2023

[rel-1.16.0] Cherry-pick 17507 (#17520)
chilo-ms
authored

Commits on Sep 14, 2023

[rel-1.16.0] Cherry-pick 16940 and 17523 (#17506)
Lafi7e
authored

Commits on Sep 18, 2023

Cherry-picks pipeline changes to 1.16.0 release branch (#17577)
snnn
authored

Commits on Oct 2, 2023

Cherry-picks for 1.16.1 release (#17741)
snnn
authored

Commits on Oct 3, 2023

Cancel EP check in python for 1.16.1 (#17768)

RandySheriffH
and
RandyShuai
authored

Commits on Oct 4, 2023

[hotfix] fix session option access in Node.js binding (#17762)
fs-eire
authored

Commits on Oct 5, 2023

Fix onnx quantizer activation and weight type attribute
yufenglee
authored

Commits on Oct 9, 2023

Upgrade transformers to fix CI (#17830)
snnn
authored

Commits on Oct 26, 2023

Increase version number for preparing the 1.16.2 release (#18070)
snnn
authored

Commits on Oct 30, 2023

Commits on Oct 31, 2023

Commits on Nov 1, 2023

Commits on Nov 2, 2023

Commits on Nov 3, 2023

Commits on Nov 6, 2023

Update eigen's URL (#18301)
snnn
authored

Commits on Nov 7, 2023

Update eigen version (#18308)
snnn
authored

Commits on Nov 8, 2023

Commits on Nov 10, 2023

Always quantize global average pool

mapetre
authored and
ndrego
committed
added qlinearconvtranpose

xdrBogdan22
authored and
ndrego
committed
updated code for convtranspose2d to work for the quadric version of onnxruntime

xdrBogdan22
authored and
ndrego
committed
fixed registry

xdrBogdan22
authored and
ndrego
committed
Adds shape inference for conv2d_transpose

mapetre
authored and
ndrego
committed
Add QLinearConvTranspose CPU implementation

mapetre
authored and
ndrego
committed
Added int32_t templated version of Col2im

mapetre
authored and
ndrego
committed
ci: Create wheel and release upon each push to main (#1)

syassami
authored and
ndrego
committed
Shape inference for QLinearAdd and QLinearConcat

mapetre
authored and
ndrego
committed
Shape inference for QLinearMul

mapetre
authored and
ndrego
committed
Shape inference for QLinearLeakyReLU

mapetre
authored and
ndrego
committed
Adds shape inference for remaining QLinear operators

mapetre
authored and
ndrego
committed
Add unittest for QLinear shape inference

mapetre
authored and
ndrego
committed
Updated shape inference to default to ONNX implementation

mapetre
authored and
ndrego
committed
Applied python linter

mapetre
authored and
ndrego
committed
Refactored QLinear shape inference test

mapetre
authored and
ndrego
committed
Addressed PR comments

mapetre
authored and
ndrego
committed
Add type inference support for custom operators

mapetre
authored and
ndrego
committed
Removed formatting changes

mapetre
authored and
ndrego
committed
Add shape inference for QLinearConvTranspose

mapetre
authored and
ndrego
committed
Add unnittest for QLinearConvTranspose shape inference

mapetre
authored and
ndrego
committed
ci: Enable workflow_call

syassami
authored and
ndrego
committed
ci: Add mac python3.10 release

syassami
authored and
ndrego
committed
ci: Create release on github.ref == refs/heads/main

syassami
authored and
ndrego
committed
wheel.yaml: No warning as error during compile
ndrego
committed
QuadricCustomOp handling (#12)
ndrego
committed
ci: Use self hosted arm64 macos runner (#16)

syassami
authored and
ndrego
committed
QuadricCustomOp: Handle multiple outputs when shape inferencing (#17)
ndrego
committed

Commits on Nov 14, 2023

quadric_custom_op: Handle duplicated input names (#18)
ndrego
authored

Commits on Jan 26, 2024

Add calibrator option to add extra dtypes
mgleonard425
committed