Hugging Face's logo

bipin
/
image-caption-generator

Image-to-Text
Transformers
PyTorch
Safetensors
vision-encoder-decoder
image-text-to-text
image-captioning
Model card Files
xet
 Community
5

snippet error

#1
by fcakyon - opened
Discussion
fcakyon

When I test on a image using the UI in the model card page, it gives this error:

Can't load tokenizer using from_pretrained, please update its configuration: <class 'transformers.models.vision_encoder_decoder.configuration_vision_encoder_decoder.VisionEncoderDecoderConfig'>
jonoirwin

same here
bipin
 Owner
•
edited Nov 30, 2022

Thanks for raising the issue @fcakyon . Currently the the model can't be loaded from the UI :( you will have to load it using the transformers library.
itsyogesh

Hey @bipin , really wanted to try this out but couldn't figure out the process or the details. Can you add some installation or running instructions in the documentation?
bipin
 Owner

Hey @itsyogesh , I've updated the model card to include all required steps for doing inference and also the links to the training procedure.
Thanks everyone for your patience. I'm closing this issue for now. If you face any issues with using the model in code, please feel free to re-open this or raise another issue :)
bipin changed discussion status to closed

Sign up or log in to comment