comaniac opened a new pull request #7675: URL: https://github.com/apache/tvm/pull/7675
This PR removes unnecessary reshape ops in the PyTorch frontend when converting to batch_matmul. This should help the performance of NLP models such as BERT. cc @siju-samuel @masahi ---------------------------------------------------------------- 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]
