LucaGroup's picture
Update README.md
7765557 verified
metadata
license: apache-2.0
tags:
  - LucaVirus
  - LucaOne
  - Biological Foundation Model
  - Unified Nucleic Acid and Protein Language Model
  - Biology
  - AI4Science
  - AI4Biology
  - AI4Virus
language:
  - en

LucaVirus-Mask

LucaVirus-Mask - The LUCA gene-protein language model(only mask pretraining task) for Virus.

Installation

You can install the package from source using pip:

pip install tokenizers==0.19.1
pip install transformers==4.41.2
pip install lucagplm

Usage

from lucagplm import LucaGPLMModel, LucaGPLMTokenizer

# Load model
model = LucaGPLMModel.from_pretrained("LucaGroup/LucaVirus-default-step3.8M")
tokenizer = LucaGPLMTokenizer.from_pretrained("LucaGroup/LucaVirus-default-step3.8M")

# Example usage(for DNA or RNA, bath seq_type="gene")
seq = "ATCG"
inputs = tokenizer(seq, seq_type="gene",return_tensors="pt")
outputs = model(**inputs)

print(outputs.last_hidden_state.shape)

seq = "NSQTA"(for protein, bath seq_type="prot")
inputs = tokenizer(seq, seq_type="prot",return_tensors="pt")
outputs = model(**inputs)

print(outputs.last_hidden_state.shape)

Github

For long sequence embedding, please refer to the git repository:
https://github.com/LucaOne/LucaVirus