Hugging Face 上的 Sentence Transformers
Hugging Face sentence-transformers is a Python framework for state-of-the-art sentence, text and image embeddings. You can use these embedding models from the
HuggingFaceEmbeddingsclass.
注意
在本地运行 sentence-transformers 可能会受到操作系统和其他全局因素的影响。建议仅由经验丰富的用户使用。
设置
你需要将 langchain_huggingface 包作为依赖项安装：
%pip install -qU langchain-huggingface
使用
from langchain_huggingface import HuggingFaceEmbeddings
embeddings = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
text = "This is a test document."
query_result = embeddings.embed_query(text)
# show only the first 100 characters of the stringified vector
print(str(query_result)[:100] + "...")
API 参考：HuggingFaceEmbeddings
[-0.038338568061590195, 0.12346471101045609, -0.028642969205975533, 0.05365273356437683, 0.008845377...
doc_result = embeddings.embed_documents([text, "This is not a test document."])
print(str(doc_result)[:100] + "...")
[[-0.038338497281074524, 0.12346471846103668, -0.028642890974879265, 0.05365274101495743, 0.00884535...
故障排除
如果遇到 accelerate 包未找到或导入失败的问题，安装或升级它可能会有所帮助：
%pip install -qU accelerate