Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Register ResNet1d layers in a way compatible with DataParallel
The `self.add_module` is a technique to register parameters, but it does not work with DataParallel because it does not put the parameters into the __dict__ in the normal way. https://discuss.pytorch.org/t/discrepancy-between-manual-parameter-registration-vs-using-nn-modulelist-when-parallelizing/181055 pytorch/pytorch#3174
- Loading branch information