Spaces:
Sleeping
Sleeping
Upload folder using huggingface_hub
Browse files- server/environment.py +6 -2
server/environment.py
CHANGED
|
@@ -59,6 +59,7 @@ class SkillForgeEnvironment(Environment):
|
|
| 59 |
def step(self, action: SkillForgeAction) -> SkillForgeObservation:
|
| 60 |
self._state.step_count += 1
|
| 61 |
task = TASKS[self.task_idx]
|
|
|
|
| 62 |
|
| 63 |
# --- create_skill: store template, stay on current task ---
|
| 64 |
if action.action_type == "create_skill":
|
|
@@ -102,9 +103,12 @@ class SkillForgeEnvironment(Environment):
|
|
| 102 |
|
| 103 |
result_correct, result_output = self._evaluate(exec_code, task["dataframe"], task["expected_output"])
|
| 104 |
|
|
|
|
|
|
|
|
|
|
| 105 |
if result_correct:
|
| 106 |
-
reward = 2.0
|
| 107 |
-
reward -= 0.001 * token_cost
|
| 108 |
if action.action_type == "use_skill":
|
| 109 |
reward += 0.5
|
| 110 |
self.tasks_solved += 1
|
|
|
|
| 59 |
def step(self, action: SkillForgeAction) -> SkillForgeObservation:
|
| 60 |
self._state.step_count += 1
|
| 61 |
task = TASKS[self.task_idx]
|
| 62 |
+
reward = 0.0
|
| 63 |
|
| 64 |
# --- create_skill: store template, stay on current task ---
|
| 65 |
if action.action_type == "create_skill":
|
|
|
|
| 103 |
|
| 104 |
result_correct, result_output = self._evaluate(exec_code, task["dataframe"], task["expected_output"])
|
| 105 |
|
| 106 |
+
if action.action_type == "create_skill":
|
| 107 |
+
result_correct = True
|
| 108 |
+
|
| 109 |
if result_correct:
|
| 110 |
+
reward += 2.0
|
| 111 |
+
reward -= 0.001 * token_cost
|
| 112 |
if action.action_type == "use_skill":
|
| 113 |
reward += 0.5
|
| 114 |
self.tasks_solved += 1
|