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

confusion on epochs #44

Open
Siirui opened this issue Aug 1, 2023 · 3 comments
Open

confusion on epochs #44

Siirui opened this issue Aug 1, 2023 · 3 comments

Comments

@Siirui
Copy link

Siirui commented Aug 1, 2023

Hi!
I am wondering the number of epochs in experiment. The epoch is set to 200 for EC stated in the paper, but in the config the epoch is set to 50. Whether should I modify the epochs to 200 for reproducing the experiment?

Thanks for your help!

@Oxer11
Copy link
Collaborator

Oxer11 commented Aug 1, 2023

Hi, thanks for the question. In the original experiment, we train all models for 200 epochs. When I organize the code, I find training for 50 epochs can yield similar performance as training for 200 epochs. So it should be fine to train for just 50 epochs.

@Siirui
Copy link
Author

Siirui commented Aug 2, 2023

Thank you for your reply! I have another question, when I try to reproduce the downstream tasks like EC, you mentioned in other issues that the batch size should be set to 8 using single GPU. But I found the finetuning process with low GPU-Util. If there is enough memory in GPU, can I enlarge the batch size without influencing the performance of Gearnet?
Thanks for your help!

@Oxer11
Copy link
Collaborator

Oxer11 commented Aug 2, 2023

Actually, I think the downstream performance is very sensitive to batch size. For the current config, I find batch_size=8 is the best setting. Typically, if you want to change the batch_size, you also need to change the learning rate and number of epochs to get similar performance. You can try other batch_size to see whether similar performance can be achieved.

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

2 participants