nevernever69 commited on
Commit
ea9019d
·
verified ·
1 Parent(s): 5ce515a

Upload redveil/server/redveil_environment.py with huggingface_hub

Browse files
redveil/server/redveil_environment.py CHANGED
@@ -212,7 +212,7 @@ class RedVeilEnvironment(Environment):
212
  task_description=self._task.description,
213
  milestones_reached=[],
214
  done=False,
215
- reward=0.0,
216
  )
217
 
218
  def step(
@@ -657,15 +657,15 @@ class RedVeilEnvironment(Environment):
657
  def _compute_reward(self) -> float:
658
  milestones = self._get_reached_milestones()
659
  if not milestones or not self._task:
660
- return 0.0
661
 
662
- reward = 0.0
663
  milestone_rewards = {name: val for name, val in self._task.milestones}
664
  for m in milestones:
665
  if m in milestone_rewards:
666
  reward = max(reward, milestone_rewards[m])
667
 
668
- return round(reward, 2)
669
 
670
  @property
671
  def state(self) -> State:
 
212
  task_description=self._task.description,
213
  milestones_reached=[],
214
  done=False,
215
+ reward=0.01,
216
  )
217
 
218
  def step(
 
657
  def _compute_reward(self) -> float:
658
  milestones = self._get_reached_milestones()
659
  if not milestones or not self._task:
660
+ return 0.01
661
 
662
+ reward = 0.01
663
  milestone_rewards = {name: val for name, val in self._task.milestones}
664
  for m in milestones:
665
  if m in milestone_rewards:
666
  reward = max(reward, milestone_rewards[m])
667
 
668
+ return max(0.01, min(0.99, round(reward, 2)))
669
 
670
  @property
671
  def state(self) -> State: