import numpy as np
from sentence_transformers import SentenceTransformer

# 本地加载 nomic-embed-text 模型（首次会自动下载，之后直接加载）
model = SentenceTransformer("nomic-ai/nomic-embed-text-v1", trust_remote_code=True)

def embed(text):
    return model.encode([text])[0]

def cosine(a, b):
    return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))

docs = [
    "E 配速跑是有氧基础训练，强度低、时间长",
    "间歇跑提升最大摄氧量，强度高、休息短",
    "马拉松补给策略：每 5km 补水一次",
    "跑前热身 10 分钟可降低受伤风险",
    "心率区间二对应有氧耐力训练",
]

print("正在为文档生成嵌入向量...")
doc_vectors = [embed(d) for d in docs]

def search(query, top_k=2):
    q_vec = embed(query)
    scores = [(cosine(q_vec, dv), doc) for dv, doc in zip(doc_vectors, docs)]
    scores.sort(reverse=True, key=lambda x: x[0])
    return scores[:top_k]

results = search("心肺训练怎么做")
print("\n查询：心肺训练怎么做")
for score, doc in results:
    print(f"{score:.3f}  {doc}")