mirror of
https://github.com/mii443/tokenizers.git
synced 2025-08-23 16:49:27 +00:00
Python - More updates to the new API
This commit is contained in:
@ -76,7 +76,7 @@ elif args.type == "bert":
|
||||
strip_accents=True,
|
||||
lowercase=True,
|
||||
)
|
||||
# tok_r.pre_tokenizer = pre_tokenizers.Whitespace.new()
|
||||
# tok_r.pre_tokenizer = pre_tokenizers.Whitespace()
|
||||
tok_r.pre_tokenizer = pre_tokenizers.BertPreTokenizer()
|
||||
tok_r.decoder = decoders.WordPiece()
|
||||
tok_r.post_processor = BertProcessing(
|
||||
|
Reference in New Issue
Block a user