Hey guys please could someone help me figure out the bert-restore-punctuation model

I’m trying to use accelerated interference on the bert-restore-punctuation. Unfortunately it is not working and i’m getting errors saying “Can’t load tokenizer using from_pretrained, please update its configuration: No such file or directory (os error 2)”

I’ve managed to get it working in google with transformers but I really want to use the api. Is there anyone that could lend me a helping hand. It’s the final piece to my puzzle.