Validation of results with quantitative evaluation metrics - Text to Image LORA

[HripsimeS]

Hello,

I used text to image with LORA to do fine tuning on my customized dataset and then tested model with prompts.

https://github.com/huggingface/diffusers/tree/main/examples/text_to_image

It seems it works fine, but I was wondering if there are some quantitative evaluation metrics to validate results.

Thanks in advance!

[SahilCarterr]

Hey @HripsimeS you can use following metrics for evalution

• ImageReward
• Aesthetic
• PickScore
• CLIP
• HPSv2
• HPSv2.1
• MPS

[moha23]

@HripsimeS Hey, I'm also trying to follow this script but having trouble with setting up the environment. Could you please let me know your pytorch and peft versions? Also do you install diffusers from this repo as recommended? Thank you!

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.