anugrah55 commited on
Commit
656cb14
·
verified ·
1 Parent(s): 81fb5f6

Fix score decimal formatting to strictly 2 places

Browse files
Files changed (1) hide show
  1. inference.py +2 -2
inference.py CHANGED
@@ -15,7 +15,7 @@ HF_TOKEN = os.getenv("HF_TOKEN")
15
 
16
  BENCHMARK = "data_clean_env"
17
  MAX_STEPS = 10
18
- TEMPERATURE = 0.7
19
 
20
  SYSTEM_PROMPT = textwrap.dedent(
21
  """
@@ -54,7 +54,7 @@ def log_step(step: int, action: str, reward: float, done: bool, error: Optional[
54
 
55
  def log_end(success: bool, steps: int, score: float, rewards: List[float]) -> None:
56
  rewards_str = ",".join(f"{r:.2f}" for r in rewards)
57
- print(f"[END] success={str(success).lower()} steps={steps} score={score:.3f} rewards={rewards_str}", flush=True)
58
 
59
  def get_model_action(client: OpenAI, obs_dict: dict) -> dict:
60
  user_prompt = f"Observation:\n{json.dumps(obs_dict, indent=2)}\nWhat is your next action?"
 
15
 
16
  BENCHMARK = "data_clean_env"
17
  MAX_STEPS = 10
18
+ TEMPERATURE = 0.0
19
 
20
  SYSTEM_PROMPT = textwrap.dedent(
21
  """
 
54
 
55
  def log_end(success: bool, steps: int, score: float, rewards: List[float]) -> None:
56
  rewards_str = ",".join(f"{r:.2f}" for r in rewards)
57
+ print(f"[END] success={str(success).lower()} steps={steps} score={score:.2f} rewards={rewards_str}", flush=True)
58
 
59
  def get_model_action(client: OpenAI, obs_dict: dict) -> dict:
60
  user_prompt = f"Observation:\n{json.dumps(obs_dict, indent=2)}\nWhat is your next action?"