src/core/embeddings.py

from sentence_transformers import SentenceTransformer
import torch
import numpy as np
import os
import logging

class EmbeddingModel:
    def __init__(self, model_name="BAAI/bge-m3"):
        try:
            # 使用 Hugging Face 模型 ID
            self.model = SentenceTransformer(model_name)
            self.device = "cuda" if torch.cuda.is_available() else "cpu"
            self.model.to(self.device)
            logging.info(f"成功加载嵌入模型 {model_name} 到 {self.device} 设备")
        except Exception as e:
            logging.error(f"加载模型失败: {str(e)}")
            raise

    def encode(self, texts, batch_size=32):
        """
        将文本转换为向量表示
        """
        embeddings = self.model.encode(
            texts,
            batch_size=batch_size,
            show_progress_bar=True,
            normalize_embeddings=True
        )
        return embeddings

    def encode_queries(self, queries):
        """
        为查询文本添加特殊前缀并编码
        BGE模型推荐在查询前添加"Represent this sentence for searching relevant passages: "
        """
        prefix = "Represent this sentence for searching relevant passages: "
        if isinstance(queries, str):
            queries = [queries]
        
        prefixed_queries = [prefix + query for query in queries]
        return self.encode(prefixed_queries)