Naoto MINAMI

Thanks for reviewing @madrob! I added two unit tests to demonstrate the failure. I think it's not look good way to use DocTransformer that uses scoring informations in SolrCloud. I...

+1 I agree with both to update documentation and to throw an exception. I will add these updates. However, I want to support interleaving in SolrCloud ideally. My idea is,...

@ArthurZucker Thank you for reviewing! I fixed all issues related to empty `precompiled_charsmap` referring to following code. https://github.com/huggingface/transformers/blob/dc67da01829090ec92dfc24653242cf3f56d1a01/src/transformers/convert_slow_tokenizer.py#L625-L628