Closed luukhd2 closed 4 months ago
I found this script on huggingface, https://huggingface.co/ronig/protein_biencoder Can you confirm this script is still accurate? If so, again to confirm, a lower distance equates to a higher binding score correct?
Thanks in advance for the help!
import torch from transformers import AutoModel, AutoTokenizer tokenizer = AutoTokenizer.from_pretrained("ronig/protein_biencoder") model = AutoModel.from_pretrained("ronig/protein_biencoder", trust_remote_code=True) model.eval() peptide_sequence = "AAA" protein_sequence = "MMM" encoded_peptide = tokenizer.encode_plus(peptide_sequence, return_tensors='pt') encoded_protein = tokenizer.encode_plus(protein_sequence, return_tensors='pt') with torch.no_grad(): peptide_output = model.forward1(encoded_peptide) protein_output = model.forward2(encoded_protein) print("distance: ", torch.norm(peptide_output - protein_output, p=2))
Hi @luukhd2 The script is accurate and you are correct , lower distances means higher binding scores.
Inference script
I found this script on huggingface, https://huggingface.co/ronig/protein_biencoder Can you confirm this script is still accurate? If so, again to confirm, a lower distance equates to a higher binding score correct?
Thanks in advance for the help!