This repository has been archived by the owner on Mar 21, 2024. It is now read-only.
Automatically and linearly scale the learning rate of the SSL encoder to the number of GPUS #667
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
fepegar
reviewed
Feb 22, 2022
fepegar
suggested changes
Feb 22, 2022
fepegar
previously approved these changes
Feb 22, 2022
ant0nsc
previously approved these changes
Mar 16, 2022
fepegar
approved these changes
Mar 21, 2022
mebristo
approved these changes
Mar 22, 2022
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
The learning rate is now linearly scaled by the number of GPUs available, e.g., lr = 0.001, 8 GPUs are available => lr=0.008.
I tested it for SimCLR and BYOL (https://ml.azure.com/experiments/id/81fa8775-1a25-47ae-9fe7-c13a6b91a421?wsid=/subscriptions/db9fc1d1-b44e-45a8-902d-8c766c255568/resourceGroups/innereyerg/providers/Microsoft.MachineLearningServices/workspaces/innereye4ws&tid=72f988bf-86f1-41af-91ab-2d7cd011db47).
In both cases the results are as expected:
I did not add any test since we have these two tests already: test_simclr_num_gpus() and test_simclr_num_nodes()