/.*/Attack(
  (search_method): GreedySearch
  (goal_function):  UntargetedClassification
  (transformation):  WordSwapEmbedding(
    (max_candidates):  15
    (embedding):  WordEmbedding
  )
  (constraints): 
    (0): MaxWordsPerturbed(
        (max_percent):  0.5
        (compare_against_original):  True
      )
    (1): ThoughtVector(
        (word_embedding):  WordEmbedding
        (metric):  max_euclidean
        (threshold):  -0.2
        (window_size):  inf
        (skip_text_shorter_than_window):  False
        (compare_against_original):  True
      )
    (2): GPT2(
        (max_log_prob_diff):  2.0
        (compare_against_original):  True
      )
    (3): RepeatModification
    (4): StopwordModification
  (is_black_box):  True
) 
/.*/
--------------------------------------------- Result 1 ---------------------------------------------
[92mPositive (100%)[0m --> [91mNegative (69%)[0m

it 's a [92mcharming[0m and [92moften[0m affecting journey . 

it 's a [91mloveable[0m and [91mordinarily[0m affecting journey . 


--------------------------------------------- Result 2 ---------------------------------------------
[91mNegative (83%)[0m --> [92mPositive (90%)[0m

unflinchingly bleak and [91mdesperate[0m 

unflinchingly bleak and [92mdesperation[0m 



+-------------------------------+--------+
| Attack Results                |        |
+-------------------------------+--------+
| Number of successful attacks: | 2      |
| Number of failed attacks:     | 0      |
| Number of skipped attacks:    | 0      |
| Original accuracy:            | 100.0% |
| Accuracy under attack:        | 0.0%   |
| Attack success rate:          | 100.0% |
| Average perturbed word %:     | 25.0%  |
| Average num. words per input: | 6.0    |
| Avg num queries:              | 48.5   |
+-------------------------------+--------+
