f49e5edde58a0f25042d2de1cae044e34f3fc8e9,ml/rl/test/gym/open_ai_gym_environment.py,OpenAIGymEnvironment,policy,#OpenAIGymEnvironment#Any#Any#Any#,173

Before Change

                action_to_take = list(actions[action_idx].keys())
                action_idx = normed_action_keys.index(action_to_take[0])
            elif isinstance(predictor, DDPGPredictor):
                return self.env.action_space.sample()
        action[action_idx] = 1.0
        return action

    def insert_into_memory(

After Change

                action_idx = normed_action_keys.index(action_to_take[0])
            elif isinstance(predictor, DDPGPredictor):
                actions = predictor.predict(next_state_dict)
                return np.array([actions[0].data.numpy()])
        action[action_idx] = 1.0
        return action

    def insert_into_memory(

Instances