hugolatendresse opened a new pull request, #17752: URL: https://github.com/apache/tvm/pull/17752
Batch_norm is a different operator in training and inference. The previous interface defaulted to the training mode and required changing an ingested pytorch program itself to use the inference mode. This is sub-ideal, especially since torch.export explicitely communicates whether batch_norm should be in training or inference mode in a given torch program. This PR automates the selection of training/inference mode in the exported program translator. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
