Skip to content
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

Regarding epochs and steps #1387

Open
vgaraujov opened this issue Aug 29, 2023 · 0 comments
Open

Regarding epochs and steps #1387

vgaraujov opened this issue Aug 29, 2023 · 0 comments

Comments

@vgaraujov
Copy link

Hello,

I already managed to configure everything, and the pre-training process starts without problems.
I noticed that by modifying TRAIN_STEP and eval_period I can control the artificial epochs and steps. However, I found that when I use a smaller or very large data set, the numbers of steps and epochs are the same.

Does it mean that in case the dataset is small, the examples are iterated multiple times to reach the total number of train steps, and if the dataset size exceeds the number of steps, some examples are not used?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant