Merge pull request #556 from QData/fix-clare-bug

Fix CLARE Extra Character Bug
QData · Nov 1, 2021 · 74ba1ff · 74ba1ff
2 parents c014fc1 + bb53d7c
commit 74ba1ff
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 0 deletions.
diff --git a/textattack/transformations/word_insertions/word_insertion_masked_lm.py b/textattack/transformations/word_insertions/word_insertion_masked_lm.py
@@ -155,6 +155,7 @@ def _get_transformations(self, current_text, indices_to_modify):
             index_to_modify = indices_to_modify[i]
             word_at_index = current_text.words[index_to_modify]
             for word in new_words[i]:
+                word = word.strip("Ġ")
                 if word != word_at_index:
                     transformed_texts.append(
                         current_text.insert_text_before_word_index(

diff --git a/textattack/transformations/word_merges/word_merge_masked_lm.py b/textattack/transformations/word_merges/word_merge_masked_lm.py
@@ -162,6 +162,7 @@ def _get_transformations(self, current_text, indices_to_modify):
             index_to_modify = merge_indices[i]
             word_at_index = current_text.words[index_to_modify]
             for word in merged_words[i]:
+                word = word.strip("Ġ")
                 if word != word_at_index:
                     temp_text = current_text.delete_word_at_index(index_to_modify + 1)
                     transformed_texts.append(

diff --git a/textattack/transformations/word_swaps/word_swap_masked_lm.py b/textattack/transformations/word_swaps/word_swap_masked_lm.py
@@ -272,6 +272,7 @@ def _get_transformations(self, current_text, indices_to_modify):
                 )
 
                 for r in replacement_words:
+                    r = r.strip("Ġ")
                     if r != word_at_index:
                         transformed_texts.append(
                             current_text.replace_word_at_index(i, r)
@@ -288,6 +289,7 @@ def _get_transformations(self, current_text, indices_to_modify):
                 index_to_modify = indices_to_modify[i]
                 word_at_index = current_text.words[index_to_modify]
                 for word in replacement_words[i]:
+                    word = word.strip("Ġ")
                     if word != word_at_index and len(utils.words_from_text(word)) == 1:
                         transformed_texts.append(
                             current_text.replace_word_at_index(index_to_modify, word)