You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When using pytorch lightning with ddp accelerator on 4 gpus, I find every checkpoint is recorded 4 times on web UI with different ids. One is on default_output_uri (s3://...) and three are on file:///... .
When using pytorch lightning with ddp accelerator on 4 gpus, I find every checkpoint is recorded 4 times on web UI with different ids. One is on
default_output_uri
(s3://...) and three are onfile:///...
.Note pytorch lightning automatically ensures that the model is saved only on the main process: https://pytorch-lightning.readthedocs.io/en/stable/weights_loading.html#manual-saving-with-accelerators
The text was updated successfully, but these errors were encountered: