Easy evaluation of the zero-shot capabilities of clip-rsicd
#2
by fhvilshoj - opened
Hi!
Thanks for your efforts in training this model. It's helping push forward the field of AI!
I wanted to show how easy it is to evaluate clip-rsicd against a bunch of others on HF datasets.
Have a look here: https://github.com/encord-team/text-to-image-eval
Would be curious to hear if this is useful to anyone 🤷♂️