Safetensors
English
llava_next
custom_code

Error while running the example on Multimodal Retrieval (HuggingFace Transformers)

#3
by KedarUMass - opened

I am getting the below error when I run the model using the example given on Hugging Face.

TypeError: LlavaNextForConditionalGeneration.pack_image_features() missing 1 required positional argument: 'vision_feature_select_strategy'

Can anyone help me with resolving this error?

Thanks

@KedarUMass Same error. Do you have any solution now?

Also same error

@Haon-Chen @sonia-rao Yes, the issue was incorrect versions of the required package.

You need the exact versions of the packages mentioned.

@KedarUMass I saw that package version troubleshooting section in the model card. Unfortunately, when I try to run the test inference script after installing the correct package versions, the error I get is: "ImportError: This modeling file requires the following packages that were not found in your environment: peft. Run pip install peft. . Installing peft however breaks all the previously installed package versions. Did you run into this issue? Not sure what version of peft is successful

Sign up or log in to comment