wrapping all environments in timelimit to stop runs longer than 100s, since some others were erroring

src/evaluation/evaluation.py

@@ -6,8 +6,9 @@ from typing import Callable, Dict, Optional
 import os
 import pickle
 from tqdm import tqdm
+from src.util.wrappers import IntersimpleTimeLimit
 from src.options.envs import OptionsEnv
-from src.util.wrappers import OptionsTimeLimit
+from src.safe_options.options import SafeOptionsEnv
 
 class IntersimpleEvaluation:
     """
@@ -36,7 +37,10 @@ class IntersimpleEvaluation:
         self.env = eval_env
         self.n_episodes = eval_env.nv
         self.use_pbar = use_pbar
-        self.is_options_env = isinstance(self.env, (OptionsEnv, OptionsTimeLimit))
+        if isinstance(self.env, IntersimpleTimeLimit):
+            self.is_options_env = isinstance(self.env.env, (OptionsEnv, SafeOptionsEnv))
+        else:
+            self.is_options_env = isinstance(self.env, (OptionsEnv, SafeOptionsEnv))
 
         # metrics present on every step of every episode
         self.metric_keys_all = ['x_all', 'y_all', 'v_all', 'a_all', 'col_all']

src/evaluation/metrics.py

@@ -24,7 +24,7 @@ def rwse(expert:List[np.ndarray], policy:List[np.ndarray], dt:float=0.1) -> Dict
     assert len(expert) == len(policy)
 
     # calculate rwse
-    times = [1,2,5,10,15,20]
+    times = [1,2,5,10,15,20,25,30]
     time_indices = [int(t/dt) for t in times]
     rwse_dict_keys = [f'rwse_{t}s' for t in times]+['rwse_end']
     se_dict = {key:[] for key in rwse_dict_keys}

src/evaluation/utils.py

@@ -2,6 +2,7 @@ import pickle
 import os
 import numpy as np
 from typing import List,Dict
+
 def save_metrics(metrics:dict, filestr:str):
     """
     Save metric dict to filestr