add attributes to population member

textattack/search_methods/genetic_algorithm.py

@@ -46,10 +46,16 @@ class GeneticAlgorithm(PopulationBasedSearch, ABC):
     @abstractmethod
     def _modify_population_member(self, pop_member, new_text, new_result, word_idx):
         """Modify `pop_member` by returning a new copy with `new_text`,
-        `new_result`, and `word_attr_list` altered appropriately for given
+        `new_result`, and, `attributes` altered appropriately for given
         `word_idx`"""
         raise NotImplementedError()
 
+    @abstractmethod
+    def _get_word_select_prob_weights(self, pop_member):
+        """Get the attribute of `pop_member` that is used for determining
+        probability of each word being selected for perturbation."""
+        raise NotImplementedError
+
     def _perturb(self, pop_member, original_result, index=None):
         """Perturb `pop_member` and return it. Replaces a word at a random
         (unless `index` is specified) in `pop_member`.
@@ -62,9 +68,11 @@ class GeneticAlgorithm(PopulationBasedSearch, ABC):
             Perturbed `PopulationMember`
         """
         num_words = pop_member.attacked_text.num_words
-        # `word_attr_list` contains some attribute of each word that helps us choose which word to transform.
-        word_attr_list = np.copy(getattr(pop_member, self._attr_name))
-        non_zero_indices = np.count_nonzero(word_attr_list)
+        # `word_select_prob_weights` is a list of values used for sampling one word to transform
+        word_select_prob_weights = np.copy(
+            self._get_word_select_prob_weights(pop_member)
+        )
+        non_zero_indices = np.count_nonzero(word_select_prob_weights)
         if non_zero_indices == 0:
             return pop_member
         iterations = 0
@@ -72,7 +80,9 @@ class GeneticAlgorithm(PopulationBasedSearch, ABC):
             if index:
                 idx = index
             else:
-                w_select_probs = word_attr_list / np.sum(word_attr_list)
+                w_select_probs = word_select_prob_weights / np.sum(
+                    word_select_prob_weights
+                )
                 idx = np.random.choice(num_words, 1, p=w_select_probs)[0]
 
             transformed_texts = self.get_transformations(
@@ -103,7 +113,7 @@ class GeneticAlgorithm(PopulationBasedSearch, ABC):
                 )
                 return pop_member
 
-            word_attr_list[idx] = 0
+            word_select_prob_weights[idx] = 0
             iterations += 1
         return pop_member
 
@@ -117,7 +127,7 @@ class GeneticAlgorithm(PopulationBasedSearch, ABC):
             pop_member1 (PopulationMember): The first population member.
             pop_member2 (PopulationMember): The second population member.
         Returns:
-            Tuple of `AttackedText` and `np.array` for new text and its corresponding `word_attr_list`.
+            Tuple of `AttackedText` and a dictionary of attributes.
         """
         raise NotImplementedError()
 
@@ -167,9 +177,7 @@ class GeneticAlgorithm(PopulationBasedSearch, ABC):
         num_tries = 0
         passed_constraints = False
         while num_tries < self.max_crossover_retries + 1:
-            new_text, word_attr_list = self._crossover_operation(
-                pop_member1, pop_member2
-            )
+            new_text, attributes = self._crossover_operation(pop_member1, pop_member2)
 
             replaced_indices = new_text.attack_attrs["newly_modified_indices"]
             new_text.attack_attrs["modified_indices"] = (
@@ -203,7 +211,7 @@ class GeneticAlgorithm(PopulationBasedSearch, ABC):
         else:
             new_results, self._search_over = self.get_goal_results([new_text])
             return PopulationMember(
-                new_text, result=new_results[0], **{self._attr_name: word_attr_list},
+                new_text, result=new_results[0], attributes=attributes
             )
 
     @abstractmethod