CLIP repo for Bioinformatics data. The goal is to project all types of data into a shared embedding space similar to IntegrA (integrate any omics) but using the contrastive clip loss.
Embedding Space for Text and Image looks something like this:
Foundation Models: Time Series: vll https://github.com/amazon-science/chronos-forecasting