jwfromm opened a new pull request #8321: URL: https://github.com/apache/tvm/pull/8321
TVM doesn't support `group_conv1d` currently and adding it formally is a non-trivial amount of work. Meanwhile, we're encountering an increasing number of models that use it. As an interim solution, we can import `group_conv1d` using `group_conv2d` by simply adding a new spatial axis with size 1. This PR adds that converter into the onnx frontend along with corresponding tests. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
