-
Notifications
You must be signed in to change notification settings - Fork 21
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
Not getting expected results for zero-shot molecule editing #23
Comments
Hi @gihanpanapitiya, I noticed that the losses are different, as in the notebook:
This seems to be caused by the corrupted checkpoints. Can you help double-check them? |
How can I double check? These are are checkpoint related parameters I use,
|
Hi, @chao1224 , thank you for your great work! I've been working with your notebook for molecule editing, but I'm running into some issues. Despite multiple attempts, the output doesn't seem to make sense. While the clip loss and L2 loss change, none of the results are valid. I've directly downloaded the checkpoints from your Hugging Face project, and I'm experiencing the same problem even when using the optimal checkpoints in
|
Hi @HantingYang, Can you help check #16? You need to tune the hyper |
Hi, thanks for getting back to me so quickly! I've been playing around with different l2 lambda values, but I'm still getting the same weird output. Are there any other settings I should be tweaking? Really appreciate your help on this!
|
Sorry I make a stupid mistake. I forgot to download the pretrained MegaMolBART models. It's working fine now. Thanks for your help! |
For zero-shot molecule editing, I am not getting the expected results as shown in the repository's notebook.
Do you have a suggestion to improve this result.
This is what I get.
The text was updated successfully, but these errors were encountered: