Open gklabs opened 2 years ago
Hi team, great work on the pipeline. As the heading suggests, this might be an enhancement on the existing codebase but could be quite helpful when training Large Datasets with the NERDA pipeline. Some early pointers to implementation- https://pytorch.org/docs/master/notes/serialization.html#recommend-saving-models https://discuss.pytorch.org/t/how-to-save-a-model-from-a-previous-epoch/20252/2
hey you, @gklabs :)
I am open to the idea. If you make a Pull Request on it, I will be happy to merge it in. Do you want to give it a go?
Hi team, great work on the pipeline. As the heading suggests, this might be an enhancement on the existing codebase but could be quite helpful when training Large Datasets with the NERDA pipeline. Some early pointers to implementation- https://pytorch.org/docs/master/notes/serialization.html#recommend-saving-models https://discuss.pytorch.org/t/how-to-save-a-model-from-a-previous-epoch/20252/2