fix

test.py

@@ -187,14 +187,12 @@ class MineRLSQILBaselineAgent(MineRLAgentBase):
         self.env = env
 
     def load_agent(self):
-        """
         boundaries = cached_reward_boundary(
             cache_dir=os.environ.get('BOUNDARY_CACHE'),
             env_id=MINERL_GYM_ENV,
             n_groups=OPTION_N_GROUPS,
             random_state=KMEANS_SEED)
         reward_channel_scale = 1. / boundaries[-1]
-        """
         self.agent = get_agent(
             n_actions=self.env.action_space.n, arch=ARCH, n_input_channels=self.env.observation_space.shape[0],
             final_epsilon=FINAL_EPSILON,
@@ -206,6 +204,7 @@ class MineRLSQILBaselineAgent(MineRLAgentBase):
             target_update_interval=TARGET_UPDATE_INTERVAL, clip_delta=CLIP_DELTA,
             batch_accumulator=BATCH_ACCUMULATOR, expert_dataset=None,
             exp_reward_scale=EXP_REWARD_SCALE, experience_lambda=EXPERIENCE_LAMBDA,
+            reward_boundaries=boundaries, reward_channel_scale=reward_channel_scale,
         )
 
         self.agent.load(os.path.abspath(os.path.join(__file__, os.pardir, 'train', 'latest', 'best')))