dp2022/WordPieceTokenizer.py
2022-01-11 23:01:48 +00:00

21 lines
739 B
Python

from tokenizers import Tokenizer
from tokenizers.models import WordPiece
tokenizer = Tokenizer.from_file("wordpiece-tokenizer-eujournal-sk.json")
tokenizedLine = ""
fileName = "eceuropa.sk"
files = ["train", "valid", "test"]
def listToString(s):
str1 = " "
return (str1.join(s))
for file in files:
with open('raw/'+fileName+'.'+file+'.raw') as read_file:
for line in read_file:
tokenizedLine = tokenizer.encode(line.rstrip())
with open('tokenized/wordpiece-tok_'+fileName+'.'+file+'.en', 'a') as input_file:
stringified = listToString(tokenizedLine.tokens)
print(stringified)
input_file.write(stringified)
input_file.write("\n")