INSTRUCTOR._load_sbert_model() got an unexpected keyword argument 'token'

[phoenixvictory]

Hey All,

Following the installation instructions of Windows 10. The installation of all dependencies went smoothly. I have tried several different models but the problem I am seeing appears to be the somewhere in the instructor. I am using ingest.py with a device type cpu using the pre-existing source pdf. Unfortunately I am met with this error constantly, any ideas what this may be?

2024-01-29 16:30:49,880 - INFO - SentenceTransformer.py:110 - Load pretrained SentenceTransformer: hkunlp/instructor-large
Traceback (most recent call last):
File "C:\Users\Phx\Documents\GitHub\Localgpt\ingest.py", line 181, in
main()
File "C:\Users\Phx.conda\envs\localGPT\lib\site-packages\click\core.py", line 1157, in call
return self.main(*args, **kwargs)
File "C:\Users\Phx.conda\envs\localGPT\lib\site-packages\click\core.py", line 1078, in main
rv = self.invoke(ctx)
File "C:\Users\Phx.conda\envs\localGPT\lib\site-packages\click\core.py", line 1434, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "C:\Users\Phx.conda\envs\localGPT\lib\site-packages\click\core.py", line 783, in invoke
return __callback(*args, **kwargs)
File "C:\Users\Phx\Documents\GitHub\Localgpt\ingest.py", line 157, in main
embeddings = HuggingFaceInstructEmbeddings(
File "C:\Users\Phx.conda\envs\localGPT\lib\site-packages\langchain\embeddings\huggingface.py", line 148, in init
self.client = INSTRUCTOR(
File "C:\Users\Phx.conda\envs\localGPT\lib\site-packages\sentence_transformers\SentenceTransformer.py", line 194, in init
modules = self._load_sbert_model(
TypeError: INSTRUCTOR._load_sbert_model() got an unexpected keyword argument 'token'

[ujwpi]

Specify v2.2.2 for Sentence Transformers in your requirements.txt, the latest version for it (v2.3.0) was released on 29th Jan and there seems to be a breaking change.

[phoenixvictory]

That certainly did the trick! Thank you for taking the time to provide clarity on this one!

[g1ra]

I have a same error here , and did trick setting in requirements.txt :

sentence-transformers==2.2.2

[ujwpi]

PR raised for this issue, waiting for it to be reviewed.
@phoenixvictory can we keep this issue open till its been fixed.

[Phycotron]

I have the same error here :
TypeError: INSTRUCTOR._load_sbert_model() got an unexpected keyword argument 'token'

[janas-adam]

had similar issue in latest version 2.6.0, like mentioned above: with 2.2.2 works well

line 191, in __init__
    modules = self._load_sbert_model(
TypeError: INSTRUCTOR._load_sbert_model() got an unexpected keyword argument 'token'

[qsdhj]

I also get this error:
TypeError: INSTRUCTOR._load_sbert_model() got an unexpected keyword argument 'token'

My code is from the langchain doc:

from langchain.embeddings import HuggingFaceInstructEmbeddings

model_name = "hkunlp/instructor-large"
model_kwargs = {'device': 'cpu'}
encode_kwargs = {'normalize_embeddings': True}
hf = HuggingFaceInstructEmbeddings(
    model_name=model_name,
    model_kwargs=model_kwargs,
    encode_kwargs=encode_kwargs
)

sentence-transformers 2.7.0

[NebeyouMusie]

run pip install sentence-transformers==2.2.2 then restart your kernel. It worked for me.

[EZZAH]

1. Install pip install sentence-transformers==2.2.2
2. restart your kernel.
3. Wait for few second
4. Rerun

from langchain_community.embeddings import HuggingFaceInstructEmbeddings
from InstructorEmbedding import INSTRUCTOR
instructembeddings= HuggingFaceInstructEmbeddings(model_name = "hkunlp/instructor-xl",
model_kwargs = {'device': 'cpu'}, encode_kwargs = {'normalize_embeddings': True})