Encountered an error while running the example.

#1
by i2vec - opened

image.png

Is the processor_config.json missing from the weight ?

That's because when processing the image,the patch_size is None,I solve this by manual set the patch_size in anaconda3/envs/your_envs_name/lib/python3.10/site-packages/transformers/models/llava_next/processing_llava_next.py line 84.
I don't know what the patch_size should be,so I set the same patch_size as the one in config.json which is 14.
After doing this the model runs well temporarily.
Hope BAAI authority can fix this soon.

Sign up or log in to comment