INNER CODE UNIT · Python

__init__

tensorchord/modelz-llm · src/modelz_llm/emb.py:12

    def __init__(self, model_name: str, device: str) -> None:
        self.model_name = model_name
        self.tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
        self.model = transformers.AutoModel.from_pretrained(model_name)
        if device == "auto":
            self.device = (
                torch.cuda.current_device() if torch.cuda.is_available() else "cpu"
            )
        else:
            self.device = device

        self.model = self.model.to(self.device)
        self.model.eval()

    # copied from https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2#usage-huggingface-transformers
    def get_embedding_with_token_count(self, sentences: Union[str, List[str]]):
        # Mean Pooling - Take attention mask into account for correct averaging
        def mean_pooling(model_output, attention_mask):

View source record →