-
Notifications
You must be signed in to change notification settings - Fork 345
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Include batch_size as an argument, instead of sample_rate #473
Conversation
This pull request was exported from Phabricator. Differential Revision: D38799564 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
👍
Summary: Pull Request resolved: #473 Other examples in Opacus follow this logic for arguments related to batch size: - batch_size and batch_size_test CIFAR-10 still uses the `sample_rate * len(train_data)`. This diff updates that, to aslo make it clearer (someone was confused [here](https://discuss.pytorch.org/t/the-opacus-example-train-batch-size-vs-sampling-rate/158771)) Reviewed By: ffuuugor Differential Revision: D38799564 fbshipit-source-id: 51035caada49c90ee95aec51e75e66837eec4338
This pull request was exported from Phabricator. Differential Revision: D38799564 |
metavar="SR", | ||
help="sample rate used for batch construction (default: 0.005)", | ||
metavar="N", | ||
help="train bacth size", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
may be change this to approximate batch size in the help string, since the batch size will not necessarily be 2K.
This pull request was exported from Phabricator. Differential Revision: D38799564 |
This pull request was exported from Phabricator. Differential Revision: D38799564 |
1 similar comment
This pull request was exported from Phabricator. Differential Revision: D38799564 |
Summary:
Other examples in Opacus follow this logic for arguments related to batch size:
CIFAR-10 still uses the
sample_rate * len(train_data)
.This diff updates that, to aslo make it clearer (someone was confused here)
Differential Revision: D38799564