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.

Sign up for GitHub

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

Jump to bottom

About training!!! #9

Closed
Maikuraky opened this issue Mar 24, 2021 · 2 comments
Closed

About training!!! #9

Maikuraky opened this issue Mar 24, 2021 · 2 comments

Comments

@Maikuraky
Copy link

Hi, Ivan
Can you relax the training code which contains data processing such as subsample and 300 residues limit?
This confuse me a long time.
Thanks!
@gjoni
Copy link
Owner

gjoni commented Mar 25, 2021

Hi Maikuraky,
There were two main reasons for doing msa subsampling and slicing: msa subsampling is a simple data diversification techniques and using it gives a better model compared to not using, while slicing is used for technical reasons and helps to fit into the gpu memory (otherwise we get OOMs caused by long sequences).

Ivan

@gjoni gjoni closed this as completed Mar 30, 2021
@gjoni
Copy link
Owner

gjoni commented Mar 30, 2021

closed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@gjoni @Maikuraky