An error occurs when using “langchain_community.embeddings HuggingFaceEmbeddings”

[yingxin-chen]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain_community.embeddings import HuggingFaceEmbeddings

# 创建向量数据库
embedding_model_dict = {
        "bge-large-zh": "E:\Project\VScode_Project\Demo_Project\RAG-Demo\model\\bge-large-zh"
    }
EMBEDDING_MODEL = "bge-large-zh"
embeddings = HuggingFaceEmbeddings(model_name=embedding_model_dict[EMBEDDING_MODEL])

Description

I use HuggingFaceEmbeddings to build the local embedding model, but the following error occurs:

---

ImportError Traceback (most recent call last)
E:\Anaconda3\Lib\site-packages\transformers\utils\import_utils.py in _get_module(self, module_name)
1183 # docstyle-ignore
-> 1184 ESSENTIA_IMPORT_ERROR = """
1185 {0} requires essentia library. But that was not found in your environment. You can install them with pip:

e:\Anaconda3\lib\importlib_init_.py in import_module(name, package)
126 level += 1
--> 127 return _bootstrap._gcd_import(name[level:], package, level)
128

e:\Anaconda3\lib\importlib_bootstrap.py in _gcd_import(name, package, level)

e:\Anaconda3\lib\importlib_bootstrap.py in find_and_load(name, import)

e:\Anaconda3\lib\importlib_bootstrap.py in find_and_load_unlocked(name, import)

e:\Anaconda3\lib\importlib_bootstrap.py in _load_unlocked(spec)

e:\Anaconda3\lib\importlib_bootstrap_external.py in exec_module(self, module)

e:\Anaconda3\lib\importlib_bootstrap.py in _call_with_frames_removed(f, *args, **kwds)

E:\Anaconda3\Lib\site-packages\transformers\integrations\peft.py in
...
1188 """

RuntimeError: Failed to import transformers.models.bert.modeling_bert because of the following error (look up to see its traceback):
Failed to import transformers.integrations.peft because of the following error (look up to see its traceback):
cannot import name 'DEFAULT_CIPHERS' from 'urllib3.util.ssl_' (E:\Anaconda3\Lib\site-packages\urllib3\util\ssl_.py)

System Info

python==3.9.13
langchain==0.1.17
transformers==4.37.2
urllib3==2.2.2

[dosubot[bot]]

Hello @yingxin-chen! I'm here to assist you with your issue. I can help you solve bugs, answer questions, and become a contributor.

To resolve the ImportError when using HuggingFaceEmbeddings from langchain_community.embeddings with the error message "cannot import name 'DEFAULT_CIPHERS' from 'urllib3.util.ssl_'", you need to update the urllib3 package to a version that is compatible with your environment. This error typically occurs due to an incompatibility between the versions of urllib3 and other packages.

You can update urllib3 by running the following command:

pip install --upgrade urllib3

If the issue persists, you may need to ensure that all related packages are updated to compatible versions. You can do this by updating all packages:

pip install --upgrade requests urllib3

This should resolve the ImportError related to DEFAULT_CIPHERS in urllib3.util.ssl_ ^[1].

To continue talking to Dosu, mention @dosu.

[yingxin-chen]

Thanks, but the problem remains, I have two identical environments, one of which works fine. Is there any other solution?

[dosubot[bot]]

If updating urllib3 does not resolve the ImportError and you have two identical environments where one works fine, another solution is to ensure that the sentence_transformers package is installed. The HuggingFaceEmbeddings class requires this package, and if it is not installed, it will raise an ImportError.

You can install the sentence_transformers package using pip:

pip install sentence-transformers

Additionally, if you are using HuggingFaceHubEmbeddings, ensure that the huggingface_hub package is installed and that you have set the HUGGINGFACEHUB_API_TOKEN environment variable or passed it as a named parameter to the constructor. Install the huggingface_hub package using pip:

pip install huggingface_hub

Ensure that you are importing HuggingFaceEmbeddings correctly, as the import path might have been deprecated. The langchain library uses a dynamic import mechanism to handle deprecated imports.

This should resolve the issue if the ImportError is due to the missing sentence_transformers or huggingface_hub package, or incorrect import paths ^[1][2][3].

To continue talking to Dosu, mention @dosu.