mbrookhart opened a new pull request #5845: URL: https://github.com/apache/incubator-tvm/pull/5845
@masahi @jwfromm discovered an issue with #5755 , infer_value chokes on ops with multiple outputs. I already knew this for Tuple, this adds a work around to infer batch_norm with the following TupleGetItem node instead of by itself. In adding a regression test, I discovered that the Onnx importer for batch norm wasn't tested, so I added a unit test for it. Are you guys aware of any other multi-output ops that might suffer the same problem? ---------------------------------------------------------------- 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]
