Find answers from the community

Updated 3 months ago

Hello guys, When I execute this piece

Hello guys, When I execute this piece of code

Plain Text

from llama_index.core.postprocessor import SentenceTransformerRerank
rerank = SentenceTransformerRerank(
            model="cross-encoder/ms-marco-MiniLM-L-2-v2", top_n=3
        )

Then this error will occur.
ImportError: ('Cannot import sentence-transformers or torch package,', 'please pip install torch sentence-transformers')

Even though I have already installed these packages properly, the error still occurs.

10 comments

WWhiteFang_Jr

Can you verify where you have installed the package and where your application is running this piece of code are the same

WWhiteFang_Jr

It could be that they are in diff env

VVishwas

Yes, I checked,
Work the functionality in a script PY file, and it runs without errors. However, when I configure this functionality in my actual project, it shows an error. I verified the dependencies for both, and they are the same.

LLogan M

Are you running in a notebook? You might have to restart

LLogan M

The code that raises this error is just doing from sentence_transformers import CrossEncoder

VVishwas

No, I am using vs code!
Yes, I think you are right. How can I resolve this?

VVishwas

And when i execute this code

Plain Text

from sentence_transformers import CrossEncoder

then it will show another error

Plain Text

ImportError: libcufft.so.11: failed to map segment from shared object

VVishwas

I am not sure what I am doing wrong. Can you please help me resolve this error?

LLogan M

oh that just means you have some issue with cuda installation I think

LLogan M

google should help there

Add a reply