sravaniamere commited on
Commit
2be7532
·
1 Parent(s): eecac6e

Fix stagnation penalty floor to 0.01, never 0.0

Browse files
Files changed (3) hide show
  1. sql_env/env.py +1 -1
  2. sql_env/models.py +1 -1
  3. sql_env/server.py +1 -1
sql_env/env.py CHANGED
@@ -74,7 +74,7 @@ class SQLCorrectionEnv:
74
  if abs(reward - self._last_reward) < 0.01 and self._step_count > 1:
75
  self._stagnation_count += 1
76
  if self._stagnation_count >= 2:
77
- reward = max(0.0, reward - 0.1)
78
  else:
79
  self._stagnation_count = 0
80
 
 
74
  if abs(reward - self._last_reward) < 0.01 and self._step_count > 1:
75
  self._stagnation_count += 1
76
  if self._stagnation_count >= 2:
77
+ reward = max(0.01, reward - 0.1)
78
  else:
79
  self._stagnation_count = 0
80
 
sql_env/models.py CHANGED
@@ -42,7 +42,7 @@ class SQLState(State):
42
 
43
 
44
  class SQLReward(BaseModel):
45
- # Allow full [0.0, 1.0] range so perfect matches can return exactly 1.0
46
  value: float = Field(gt=0.0, lt=1.0)
47
  reason: str
48
 
 
42
 
43
 
44
  class SQLReward(BaseModel):
45
+ # Strictly between 0 and 1 as required by the OpenEnv spec
46
  value: float = Field(gt=0.0, lt=1.0)
47
  reason: str
48
 
sql_env/server.py CHANGED
@@ -55,7 +55,7 @@ class SQLCorrectionEnvironment(Environment):
55
  if abs(reward - self._last_reward) < 0.01 and self._step_count > 1:
56
  self._stagnation_count += 1
57
  if self._stagnation_count >= 2:
58
- reward = max(0.0, reward - 0.1)
59
  else:
60
  self._stagnation_count = 0
61
 
 
55
  if abs(reward - self._last_reward) < 0.01 and self._step_count > 1:
56
  self._stagnation_count += 1
57
  if self._stagnation_count >= 2:
58
+ reward = max(0.01, reward - 0.1)
59
  else:
60
  self._stagnation_count = 0
61