Moonn1205/AntigenLM-prediction-sequence

AntigenLM - HA/NA Sequence Forecasting

Generative head of AntigenLM that autoregressively forecasts the next time-window influenza HA/NA nucleotide block from three historical blocks.

Paper: AntigenLM: Structure-Aware DNA Language Modeling for Influenza (arXiv:2602.09067). Code: https://github.com/Moonn1205/AntigenLM <!-- update with the real repo URL -->

Usage

pip install huggingface_hub transformers==4.29.2 torch==1.13.1

from huggingface_hub import snapshot_download
snapshot_download(
    repo_id="Moonn1205/AntigenLM-prediction-sequence",
    local_dir="prediction_sequence",
    local_dir_use_symlinks=False,
)

Then follow the instructions in the code repository to run inference.

Files

• —pytorch_model.bin - model weights
• —config.json - GPT-2 backbone config + (classifier only) label2id / id2label
• —vocab.json, tokenizer_config.json, special_tokens_map.json, added_tokens.json - tokenizer