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

support for dadapaption V3 #530

Merged
merged 13 commits into from
May 25, 2023
Merged

support for dadapaption V3 #530

merged 13 commits into from
May 25, 2023

Conversation

sdbds
Copy link
Contributor

@sdbds sdbds commented May 22, 2023

Dadapaption Change Log V3
1、Major improvements to DAdaptAdam, improving the performance particularly on Transformer models. This variant may behave differently in practice. The old version is availiable under experimental/d_adapt_adam_preprint if you wish to continue to use it.The IP variant is now the main variant of the method.
2、Added Lion. This is highly experimental. Feedback on it's performance is welcome.

What i change:
1、keep old DadaptAdam for use DAdaptation or DAdaptAdamPreprint
2、use new DadaptAdam variant.
3、add DAdaptLion
4、add DAdaptAdan IP variant for use DAdaptAdanIP

update way:

.\venv\Scripts\activate
pip install dadaptation -U

@kohya-ss
Copy link
Owner

Thank you for this! I will review and merge as soon as I have time.

@kohya-ss kohya-ss merged commit 5cdf4e3 into kohya-ss:dev May 25, 2023
@FurkanGozukara
Copy link

what learning rate and weight decay suggested?

@sdbds
Copy link
Contributor Author

sdbds commented Jun 15, 2023

what learning rate and weight decay suggested?

see here

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

Successfully merging this pull request may close these issues.

3 participants