Commit ·
bcbd071
1
Parent(s): dd41132
fix: add score to reset response for validator compliance
Browse files- server/app.py +14 -1
server/app.py
CHANGED
|
@@ -69,7 +69,20 @@ def reset(request: ResetRequest = ResetRequest()):
|
|
| 69 |
obs = env.reset()
|
| 70 |
|
| 71 |
# return as dict so it plays nice with the OpenEnv spec
|
| 72 |
-
return {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
|
| 74 |
|
| 75 |
def _clamp_score(v: float) -> float:
|
|
|
|
| 69 |
obs = env.reset()
|
| 70 |
|
| 71 |
# return as dict so it plays nice with the OpenEnv spec
|
| 72 |
+
return {
|
| 73 |
+
"observation": obs.model_dump(),
|
| 74 |
+
"task_id": task,
|
| 75 |
+
"info": {
|
| 76 |
+
"score": 0.5,
|
| 77 |
+
"task_id": task,
|
| 78 |
+
"step_count": 0,
|
| 79 |
+
"all_task_scores": {
|
| 80 |
+
"survive_8_hours": 0.5,
|
| 81 |
+
"maximize_productivity": 0.5,
|
| 82 |
+
"smart_break_taker": 0.5,
|
| 83 |
+
}
|
| 84 |
+
}
|
| 85 |
+
}
|
| 86 |
|
| 87 |
|
| 88 |
def _clamp_score(v: float) -> float:
|