google

/

maxim-s3-deblurring-reds

like 6

Follow

Google 7.07k

Some thoughts

• Are there any evaluation metrics you can include?
• Add the tag image-to-image so it shows up here?
• add language: en to metadata

Some nice to haves

• Some discussion of limitations/biases would be nice if you have that information
• discussion of the dataset/data preprocessing for training
• Discussion of training process
• Note somewhere about authors of the model card (you)
• papers with code metadata link if available
• arxiv metadata link if available

Acted on the thoughts.

Some discussion of limitations/biases would be nice if you have that information

No, the authors didn't provide that information.

discussion of the dataset/data preprocessing for training

Added under the "Training procedure and results" section.

Discussion of training process

Same as above.

Note somewhere about authors of the model card (you)

It's there under disclaimer:

Disclaimer: The team releasing MAXIM did not write a model card for this model so this model card has been written by the Hugging Face team.

papers with code metadata link if available

Could you provide an example?

arxiv metadata link if available

Could you provide an example?

LGTM

Will send metadata examples tomorrow

Cool. Thanks for the review, Nate. Appreciate it.

Metadata stuff is pretty minor. Feel free to merge this and I can PR/you can after I send info to you. Whatever is easier

You can send over the info, I will incorporate it. Merging.