onnx/onnx

question good first issue topic: spec clarification contributions welcome

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

(in _op_common_pool.py) CommonPool._run returns incorrect output shape when dilations is set 8 days ago

bug topic: test good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

cumprod operation 15 days ago

topic: operator good first issue topic: enhancement contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Implement DepthToSpace as a function 16 days ago

topic: operator good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Reference implementation of ReduceLogSumExp crashes with integer input 19 days ago

bug topic: operator good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

[Docs] Reference the ir-py project for Python APIs about 1 month ago

topic: documentation good first issue

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Does antialias in Resize-18 work with empty scale? about 1 month ago

topic: operator topic: documentation good first issue topic: spec clarification contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Specify that present_key and past_key relationship for Attention 2 months ago

good first issue topic: spec clarification contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

ONNX 1.18.0 wheel building fails with Python 3.12: cmake uses system Python instead of runtime Python 2 months ago

topic: build good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

MeanVarianceNormalization without division by zero 4 months ago

bug topic: operator good first issue contributions welcome no-stale

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

How operator TopK should handle nan value? 4 months ago

question good first issue topic: spec clarification contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Unspecified Div Op behavior for negative quotients 4 months ago

good first issue topic: spec clarification contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Why does SplitToSequence not allow zeros in `split` input? 5 months ago

question module: spec good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

[Feature request] Allow Type Inference Without Shape Inference in ONNX 5 months ago

good first issue topic: enhancement module: shape inference contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

onnx-weekly has inconsistent version metadata 5 months ago

bug topic: build good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Clean up reference evaluator tests 5 months ago

good first issue module: reference implementation

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

ONNX printer does not print initializers when they are types like float16 5 months ago

bug good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Parser treats a float attribute as int and produces invalid model 5 months ago

bug good first issue module: parser contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Compiler-Options-Hardening-Guide-for-C-and-C++ 6 months ago

question good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

nanobind config can not be found for openSUSE builds 7 months ago

bug good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Combine windows pipelines into one file 7 months ago

AI Summary: This GitHub issue requests the consolidation of two separate Windows release pipelines, `release_win_x86_64.yml` and `release_win_aarch64.yml`, into a single GitHub Actions workflow file. The goal is to streamline the CI/CD process for Windows releases, likely by utilizing a matrix strategy to build for both architectures within one unified workflow.

Complexity: 2/5

good first issue

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

[Feature request] Check whether we should introduce https://scientific-python.org/specs/spec-0004/ 9 months ago

good first issue

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

`onnx.helper.check_model` segfaults due to bug in OneHotEncoder's shape inference 10 months ago

bug good first issue contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Reference implementation of reduction operations ignores `keepdims` for `noop_with_empty_axes=1` 11 months ago

bug good first issue module: reference implementation contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

[ONNX] Elu operator should support input tensor with any shape rather than only 1D about 1 year ago

AI Summary: The task is to update the ONNX Elu operator specification to support input tensors of any shape, not just 1D, as currently documented. This involves modifying the ONNX specification and potentially the associated codebase to accommodate this change.

Complexity: 4/5

topic: operator good first issue module: schema contributions welcome

onnx/onnx

21.1K

Open standard for machine learning interoperability

Python

#ai#artificial-intelligence#deep-learning#deep-neural-networks#dnn#keras#machine-learning#ml#neural-network#onnx#pytorch#scikit-learn#tensorflow

Open Issues Need Help