This repository has been archived by the owner on Mar 19, 2024. It is now read-only.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR is a follow up on a few issues that were discussed in #264 regarding TensorBoard:
SummaryWriter
although they will not be used (only the primary process logs for the time being) leading to empty filesNow, things are not so simple regarding point 2: to distinguish two jobs running on the same configuration, I concatenated the JOB_ID of SLURM, which is unique (allows to distinguish jobs), stable (upon pre-emption it stays the same) and useful to find back which job produced what log.
But I do not know if there is such a trick for other job schedulers: the advantage of using the checkpoint directory was that it was unique to a JOB and stable, allowing TB to know in which file to log even after pre-emption. So I am a bit unsure about the consequences of the second point.