Goal fucntion for Classification¶
Determine for if an attack has been successful in Classification¶
- class textattack.goal_functions.classification.classification_goal_function.ClassificationGoalFunction(model_wrapper, maximizable=False, use_cache=True, query_budget=inf, model_batch_size=32, model_cache_size=1048576)¶
A goal function defined on a model that outputs a probability for some number of classes.
Determine if maintaining the same predicted label¶
- class textattack.goal_functions.classification.input_reduction.InputReduction(*args, target_num_words=1, **kwargs)¶
Attempts to reduce the input down to as few words as possible while maintaining the same predicted label.
From Feng, Wallace, Grissom, Iyyer, Rodriguez, Boyd-Graber. (2018). Pathologies of Neural Models Make Interpretations Difficult. ArXiv, abs/1804.07781.
Determine if an attack has been successful in targeted Classification¶
- class textattack.goal_functions.classification.targeted_classification.TargetedClassification(*args, target_class=0, **kwargs)¶
A targeted attack on classification models which attempts to maximize the score of the target label.
Complete when the arget label is the predicted label.
Determine successful in untargeted Classification¶
- class textattack.goal_functions.classification.untargeted_classification.UntargetedClassification(*args, target_max_score=None, **kwargs)¶
An untargeted attack on classification models which attempts to minimize the score of the correct label until it is no longer the predicted label.
target_max_score (float) – If set, goal is to reduce model output to below this score. Otherwise, goal is to change the overall predicted class.