[question] how to convert embeddings into text?

[eisneim]

first of all thank you for sharing such a cool project! and other projects of yours are also amazing! i have test the model it works great but how do i convert inference results into English text?

any examples or explaining would be appreciated!

[josephrocca]

Glad to hear it's useful! Unfortunately CLIP doesn't have this capability on its own. It can do image-to-embedding, and text-to-embedding, but not the reverse. IIRC there are (slow) ways to "iteratively search" for some English text that matches a particular CLIP embedding, but that's beyond the scope of this repo.

[eisneim]

thanks for your replay! guess i'll have to investigate how image captioning works, thanks anyway.