Edward Ma

NLPAug uses the second approach which is replacing tokens one by one. The first approach provides a much faster response time but it does make sense (at least I believe)....

Able to reproduce it. Will spend time on it. Workaround is installing it via pip `pip install nlpaug`

which nlpaug version and transformers version are you using? My transformers version is 4.16.2

suggest to use nlpaug 1.1.9. Since version 1.1.9, it will use native transformers package API (not using transformers's pipline). It will use multi-thread (i am still troubshotting) and causing this...

let use "bert-base-multilingual-uncased" as a workaround. will look into "hfl/chinese-roberta-wwm-ext" model.

It seems like your file is a compressed file. Try to uncompress it first (expecting the file extension is vec). After that try to see whether you can execute the...

Currently, it does not support this feature yet