edd80ff9c99783868de91ef14aeceb0be1dbb090,ml/rl/test/gym/open_ai_gym_environment.py,OpenAIGymEnvironment,policy,#OpenAIGymEnvironment#Any#Any#Any#,159

Before Change

                    if best_action is None or best_score < action_score:
                        best_action = action_key
                        best_score = action_score
                action_idx = normed_action_keys.index(best_action)
        action[action_idx] = 1.0
        return action

    def insert_into_memory(

After Change

                    states.append(next_state_dict[0])
                    actions.append({action_key: 1})
                action_idx = predictor.policy(states, actions)[1]
                action_to_take = list(actions[action_idx].keys())
                action_idx = normed_action_keys.index(action_to_take[0])
        action[action_idx] = 1.0
        return action

    def insert_into_memory(

Instances