in augly/text/augmenters/words_augmenter.py [0:0]
def delete(self, data: str) -> str:
"""Augmenter that merges selected words with the following word"""
results = []
tokens = tokenize(data)
aug_word_cnt = self._generate_aug_cnt(
len(tokens), self.aug_min, self.aug_max, self.aug_p
)
# Skip last word in the sentence as merges occur with the following word
filtered_word_idxes = self.pre_skip_aug(tokens[:-1])
aug_word_idxes = set(
get_aug_idxes(
self,
tokens,
filtered_word_idxes,
aug_word_cnt,
Method.WORD,
self.min_char,
)
)
if not aug_word_idxes:
return data
t_i = 0
while t_i < len(tokens):
if t_i in aug_word_idxes and len(tokens[t_i + 1]) >= self.min_char:
results.append(tokens[t_i] + tokens[t_i + 1])
t_i += 1
else:
results.append(tokens[t_i])
t_i += 1
return detokenize(results)