Get TypeError: Pooling.__init__() got an unexpected keyword argument 'pooling_mode_weightedmean_tokens' when init model

#12
by Boman - opened

Hello everyone!

When I try to use the instructor-xl model in Python, specifically during the model initialization step:

model = SentenceTransformer("hkunlp/instructor-xl")

I encountered a type error after the model finished downloading:

TypeError: Pooling.__init__() got an unexpected keyword argument 'pooling_mode_weightedmean_tokens'

I suspect this issue may be related to the environment. Here is my current environment:
torch=2.0.0
python=3.10
cuda=11.8

What is the quickest way for me to potentially solve this problem? Thank u again.

NLP Group of The University of Hong Kong org

Hi, Thanks a lot for your interest in the INSTRUCTOR model!

You may first install the package InstructorEmbedding:

pip install InstructorEmbedding

and then load the pre-trained model:

from InstructorEmbedding import INSTRUCTOR
model = INSTRUCTOR('hkunlp/instructor-xl')

Feel free to add any further questions or comments!

Hi, Thanks a lot for your interest in the INSTRUCTOR model!

You may first install the package InstructorEmbedding:

pip install InstructorEmbedding

and then load the pre-trained model:

from InstructorEmbedding import INSTRUCTOR
model = INSTRUCTOR('hkunlp/instructor-xl')

Feel free to add any further questions or comments!

Thank you! this works.
I previously overlooked the guidance provided in the model card and directly used the recommended commands.

Hello TypeError: INSTRUCTOR_Pooling.init() got an unexpected keyword argument 'include_prompt'
for

Sign up or log in to comment