Couldn't instantiate the backend tokenizer about text-generation-inference

[OlivierDehaene]

What's the error? Can you use 0.5.0 instead of latest?
I'm unable to reproduce on my end.

[MohamedAliRashad]

@OlivierDehaene

ValueError: Couldn't instantiate the backend tokenizer from one of: 
(1) a `tokenizers` library serialization file, 
(2) a slow tokenizer instance to convert or 
(3) an equivalent slow tokenizer class to instantiate and convert. 
You need to have sentencepiece installed to convert a slow tokenizer to a fast one.

[OlivierDehaene]

docker run --gpus all --shm-size 1g -p 8080:80 -v /home/mohamedr/cache:/data ghcr.io/huggingface/text-generation-inference:0.5.0 --model-id Tribbiani/vicuna-7b --num-shard 1 --max-total-tokens 2048

[MohamedAliRashad]

@OlivierDehaene
Still the same error

[OlivierDehaene]

Can you clear your model cache (in /home/mohamedr/cache) and re-try?

[MohamedAliRashad]

@OlivierDehaene
Still the same problem

[MohamedAliRashad]

@OlivierDehaene
It now gives me this error:

ModuleNotFoundError: No module named 'einops'

[MohamedAliRashad]

It worked somehow in the new versions