CATHe2 Dataset and models

The dataset named "data.zip" consists of all the relevant data to train the CATHe2 models, from the primary sequences of the training, validation and test sets as well as many csv and fasta files created by the github project code. The main chunk of this dataset is the embeddings of the training set, for every pLM tested (pLM = Protein Language Model = LLM specialized in proteins).The saved_models.zip contains the trained models used to make CATH annotation prediction with CATHe2 as well as a model trained with the embeddings of each pLM tested.