regression about coop HOT 4 OPEN

Not sure if turning CLIP into a regression model would work because CLIP was pretrained using a contrastive learning loss, which is significantly different from a regression objective

But a straightforward way to try is to train a single prompt, e.g., "V_1 ... V_M target value", using a regression loss like the mean square error loss (so the multiplication of the resulting text features with a single image's features would lead to a continuous value)

from coop.

@dribnet hi, just a follow-up because I'm also interested to know if CLIP works for regression tasks, any update to share?

from coop.

I ended up not taking this route but instead adapted the dataset to be classes (positive / negative).

I did find a more detailed read of issues trying to use percentage type labels as I had originally proposed in gwerns GPT-3 writeup in the calibration section.

from coop.

interesting! thanks

from coop.