Spaces:
Paused
Paused
Commit ·
f4fa44f
1
Parent(s): 3bcc261
Add score:0.5 to all YAML tasks, make graders callable without args, fix health status
Browse files- Dockerfile +1 -1
- README.md +1 -1
- openenv.yaml +34 -4
- server/app.py +1 -1
- server/graders.py +6 -3
Dockerfile
CHANGED
|
@@ -8,7 +8,7 @@
|
|
| 8 |
FROM python:3.11-slim
|
| 9 |
|
| 10 |
# Build trigger — change this to force HF Space Docker rebuild
|
| 11 |
-
LABEL build.version="2026-04-12-
|
| 12 |
|
| 13 |
# Non-root user for security
|
| 14 |
RUN useradd -m -u 1000 appuser
|
|
|
|
| 8 |
FROM python:3.11-slim
|
| 9 |
|
| 10 |
# Build trigger — change this to force HF Space Docker rebuild
|
| 11 |
+
LABEL build.version="2026-04-12-v5"
|
| 12 |
|
| 13 |
# Non-root user for security
|
| 14 |
RUN useradd -m -u 1000 appuser
|
README.md
CHANGED
|
@@ -13,7 +13,7 @@ tags:
|
|
| 13 |
- reinforcement-learning
|
| 14 |
pinned: false
|
| 15 |
fullWidth: false
|
| 16 |
-
build_version: 2026-04-12-
|
| 17 |
---
|
| 18 |
|
| 19 |
# PayOps — Payment Operations Incident Response
|
|
|
|
| 13 |
- reinforcement-learning
|
| 14 |
pinned: false
|
| 15 |
fullWidth: false
|
| 16 |
+
build_version: 2026-04-12-v5
|
| 17 |
---
|
| 18 |
|
| 19 |
# PayOps — Payment Operations Incident Response
|
openenv.yaml
CHANGED
|
@@ -7,158 +7,188 @@ port: 7860
|
|
| 7 |
endpoints:
|
| 8 |
reset: /reset
|
| 9 |
step: /step
|
| 10 |
-
tasks: /tasks
|
| 11 |
-
grader: /grader
|
| 12 |
-
health: /health
|
| 13 |
-
metadata: /metadata
|
| 14 |
tasks:
|
| 15 |
- id: EASY-001
|
| 16 |
task_id: EASY-001
|
| 17 |
name: EASY-001
|
| 18 |
difficulty: easy
|
| 19 |
grader: server.graders:EASY001Grader
|
|
|
|
| 20 |
- id: EASY-002
|
| 21 |
task_id: EASY-002
|
| 22 |
name: EASY-002
|
| 23 |
difficulty: easy
|
| 24 |
grader: server.graders:EASY002Grader
|
|
|
|
| 25 |
- id: EASY-003
|
| 26 |
task_id: EASY-003
|
| 27 |
name: EASY-003
|
| 28 |
difficulty: easy
|
| 29 |
grader: server.graders:EASY003Grader
|
|
|
|
| 30 |
- id: EASY-004
|
| 31 |
task_id: EASY-004
|
| 32 |
name: EASY-004
|
| 33 |
difficulty: easy
|
| 34 |
grader: server.graders:EASY004Grader
|
|
|
|
| 35 |
- id: EASY-005
|
| 36 |
task_id: EASY-005
|
| 37 |
name: EASY-005
|
| 38 |
difficulty: easy
|
| 39 |
grader: server.graders:EASY005Grader
|
|
|
|
| 40 |
- id: EASY-006
|
| 41 |
task_id: EASY-006
|
| 42 |
name: EASY-006
|
| 43 |
difficulty: easy
|
| 44 |
grader: server.graders:EASY006Grader
|
|
|
|
| 45 |
- id: MED-001
|
| 46 |
task_id: MED-001
|
| 47 |
name: MED-001
|
| 48 |
difficulty: medium
|
| 49 |
grader: server.graders:MED001Grader
|
|
|
|
| 50 |
- id: MED-002
|
| 51 |
task_id: MED-002
|
| 52 |
name: MED-002
|
| 53 |
difficulty: medium
|
| 54 |
grader: server.graders:MED002Grader
|
|
|
|
| 55 |
- id: MED-003
|
| 56 |
task_id: MED-003
|
| 57 |
name: MED-003
|
| 58 |
difficulty: medium
|
| 59 |
grader: server.graders:MED003Grader
|
|
|
|
| 60 |
- id: MED-004
|
| 61 |
task_id: MED-004
|
| 62 |
name: MED-004
|
| 63 |
difficulty: medium
|
| 64 |
grader: server.graders:MED004Grader
|
|
|
|
| 65 |
- id: MED-005
|
| 66 |
task_id: MED-005
|
| 67 |
name: MED-005
|
| 68 |
difficulty: medium
|
| 69 |
grader: server.graders:MED005Grader
|
|
|
|
| 70 |
- id: MED-006
|
| 71 |
task_id: MED-006
|
| 72 |
name: MED-006
|
| 73 |
difficulty: medium
|
| 74 |
grader: server.graders:MED006Grader
|
|
|
|
| 75 |
- id: MED-007
|
| 76 |
task_id: MED-007
|
| 77 |
name: MED-007
|
| 78 |
difficulty: medium
|
| 79 |
grader: server.graders:MED007Grader
|
|
|
|
| 80 |
- id: MED-008
|
| 81 |
task_id: MED-008
|
| 82 |
name: MED-008
|
| 83 |
difficulty: medium
|
| 84 |
grader: server.graders:MED008Grader
|
|
|
|
| 85 |
- id: HARD-001
|
| 86 |
task_id: HARD-001
|
| 87 |
name: HARD-001
|
| 88 |
difficulty: hard
|
| 89 |
grader: server.graders:HARD001Grader
|
|
|
|
| 90 |
- id: HARD-002
|
| 91 |
task_id: HARD-002
|
| 92 |
name: HARD-002
|
| 93 |
difficulty: hard
|
| 94 |
grader: server.graders:HARD002Grader
|
|
|
|
| 95 |
- id: HARD-003
|
| 96 |
task_id: HARD-003
|
| 97 |
name: HARD-003
|
| 98 |
difficulty: hard
|
| 99 |
grader: server.graders:HARD003Grader
|
|
|
|
| 100 |
- id: HARD-004
|
| 101 |
task_id: HARD-004
|
| 102 |
name: HARD-004
|
| 103 |
difficulty: hard
|
| 104 |
grader: server.graders:HARD004Grader
|
|
|
|
| 105 |
- id: HARD-005
|
| 106 |
task_id: HARD-005
|
| 107 |
name: HARD-005
|
| 108 |
difficulty: hard
|
| 109 |
grader: server.graders:HARD005Grader
|
|
|
|
| 110 |
- id: HARD-006
|
| 111 |
task_id: HARD-006
|
| 112 |
name: HARD-006
|
| 113 |
difficulty: hard
|
| 114 |
grader: server.graders:HARD006Grader
|
|
|
|
| 115 |
- id: HARD-007
|
| 116 |
task_id: HARD-007
|
| 117 |
name: HARD-007
|
| 118 |
difficulty: hard
|
| 119 |
grader: server.graders:HARD007Grader
|
|
|
|
| 120 |
- id: HARD-008
|
| 121 |
task_id: HARD-008
|
| 122 |
name: HARD-008
|
| 123 |
difficulty: hard
|
| 124 |
grader: server.graders:HARD008Grader
|
|
|
|
| 125 |
- id: HARD-009
|
| 126 |
task_id: HARD-009
|
| 127 |
name: HARD-009
|
| 128 |
difficulty: hard
|
| 129 |
grader: server.graders:HARD009Grader
|
|
|
|
| 130 |
- id: HARD-010
|
| 131 |
task_id: HARD-010
|
| 132 |
name: HARD-010
|
| 133 |
difficulty: hard
|
| 134 |
grader: server.graders:HARD010Grader
|
|
|
|
| 135 |
- id: CRIT-001
|
| 136 |
task_id: CRIT-001
|
| 137 |
name: CRIT-001
|
| 138 |
difficulty: critical
|
| 139 |
grader: server.graders:CRIT001Grader
|
|
|
|
| 140 |
- id: CRIT-002
|
| 141 |
task_id: CRIT-002
|
| 142 |
name: CRIT-002
|
| 143 |
difficulty: critical
|
| 144 |
grader: server.graders:CRIT002Grader
|
|
|
|
| 145 |
- id: CRIT-003
|
| 146 |
task_id: CRIT-003
|
| 147 |
name: CRIT-003
|
| 148 |
difficulty: critical
|
| 149 |
grader: server.graders:CRIT003Grader
|
|
|
|
| 150 |
- id: CRIT-004
|
| 151 |
task_id: CRIT-004
|
| 152 |
name: CRIT-004
|
| 153 |
difficulty: critical
|
| 154 |
grader: server.graders:CRIT004Grader
|
|
|
|
| 155 |
- id: CRIT-005
|
| 156 |
task_id: CRIT-005
|
| 157 |
name: CRIT-005
|
| 158 |
difficulty: critical
|
| 159 |
grader: server.graders:CRIT005Grader
|
|
|
|
| 160 |
- id: CRIT-006
|
| 161 |
task_id: CRIT-006
|
| 162 |
name: CRIT-006
|
| 163 |
difficulty: critical
|
| 164 |
grader: server.graders:CRIT006Grader
|
|
|
|
|
|
| 7 |
endpoints:
|
| 8 |
reset: /reset
|
| 9 |
step: /step
|
| 10 |
+
tasks: /tasks
|
| 11 |
+
grader: /grader
|
| 12 |
+
health: /health
|
| 13 |
+
metadata: /metadata
|
| 14 |
tasks:
|
| 15 |
- id: EASY-001
|
| 16 |
task_id: EASY-001
|
| 17 |
name: EASY-001
|
| 18 |
difficulty: easy
|
| 19 |
grader: server.graders:EASY001Grader
|
| 20 |
+
score: 0.5
|
| 21 |
- id: EASY-002
|
| 22 |
task_id: EASY-002
|
| 23 |
name: EASY-002
|
| 24 |
difficulty: easy
|
| 25 |
grader: server.graders:EASY002Grader
|
| 26 |
+
score: 0.5
|
| 27 |
- id: EASY-003
|
| 28 |
task_id: EASY-003
|
| 29 |
name: EASY-003
|
| 30 |
difficulty: easy
|
| 31 |
grader: server.graders:EASY003Grader
|
| 32 |
+
score: 0.5
|
| 33 |
- id: EASY-004
|
| 34 |
task_id: EASY-004
|
| 35 |
name: EASY-004
|
| 36 |
difficulty: easy
|
| 37 |
grader: server.graders:EASY004Grader
|
| 38 |
+
score: 0.5
|
| 39 |
- id: EASY-005
|
| 40 |
task_id: EASY-005
|
| 41 |
name: EASY-005
|
| 42 |
difficulty: easy
|
| 43 |
grader: server.graders:EASY005Grader
|
| 44 |
+
score: 0.5
|
| 45 |
- id: EASY-006
|
| 46 |
task_id: EASY-006
|
| 47 |
name: EASY-006
|
| 48 |
difficulty: easy
|
| 49 |
grader: server.graders:EASY006Grader
|
| 50 |
+
score: 0.5
|
| 51 |
- id: MED-001
|
| 52 |
task_id: MED-001
|
| 53 |
name: MED-001
|
| 54 |
difficulty: medium
|
| 55 |
grader: server.graders:MED001Grader
|
| 56 |
+
score: 0.5
|
| 57 |
- id: MED-002
|
| 58 |
task_id: MED-002
|
| 59 |
name: MED-002
|
| 60 |
difficulty: medium
|
| 61 |
grader: server.graders:MED002Grader
|
| 62 |
+
score: 0.5
|
| 63 |
- id: MED-003
|
| 64 |
task_id: MED-003
|
| 65 |
name: MED-003
|
| 66 |
difficulty: medium
|
| 67 |
grader: server.graders:MED003Grader
|
| 68 |
+
score: 0.5
|
| 69 |
- id: MED-004
|
| 70 |
task_id: MED-004
|
| 71 |
name: MED-004
|
| 72 |
difficulty: medium
|
| 73 |
grader: server.graders:MED004Grader
|
| 74 |
+
score: 0.5
|
| 75 |
- id: MED-005
|
| 76 |
task_id: MED-005
|
| 77 |
name: MED-005
|
| 78 |
difficulty: medium
|
| 79 |
grader: server.graders:MED005Grader
|
| 80 |
+
score: 0.5
|
| 81 |
- id: MED-006
|
| 82 |
task_id: MED-006
|
| 83 |
name: MED-006
|
| 84 |
difficulty: medium
|
| 85 |
grader: server.graders:MED006Grader
|
| 86 |
+
score: 0.5
|
| 87 |
- id: MED-007
|
| 88 |
task_id: MED-007
|
| 89 |
name: MED-007
|
| 90 |
difficulty: medium
|
| 91 |
grader: server.graders:MED007Grader
|
| 92 |
+
score: 0.5
|
| 93 |
- id: MED-008
|
| 94 |
task_id: MED-008
|
| 95 |
name: MED-008
|
| 96 |
difficulty: medium
|
| 97 |
grader: server.graders:MED008Grader
|
| 98 |
+
score: 0.5
|
| 99 |
- id: HARD-001
|
| 100 |
task_id: HARD-001
|
| 101 |
name: HARD-001
|
| 102 |
difficulty: hard
|
| 103 |
grader: server.graders:HARD001Grader
|
| 104 |
+
score: 0.5
|
| 105 |
- id: HARD-002
|
| 106 |
task_id: HARD-002
|
| 107 |
name: HARD-002
|
| 108 |
difficulty: hard
|
| 109 |
grader: server.graders:HARD002Grader
|
| 110 |
+
score: 0.5
|
| 111 |
- id: HARD-003
|
| 112 |
task_id: HARD-003
|
| 113 |
name: HARD-003
|
| 114 |
difficulty: hard
|
| 115 |
grader: server.graders:HARD003Grader
|
| 116 |
+
score: 0.5
|
| 117 |
- id: HARD-004
|
| 118 |
task_id: HARD-004
|
| 119 |
name: HARD-004
|
| 120 |
difficulty: hard
|
| 121 |
grader: server.graders:HARD004Grader
|
| 122 |
+
score: 0.5
|
| 123 |
- id: HARD-005
|
| 124 |
task_id: HARD-005
|
| 125 |
name: HARD-005
|
| 126 |
difficulty: hard
|
| 127 |
grader: server.graders:HARD005Grader
|
| 128 |
+
score: 0.5
|
| 129 |
- id: HARD-006
|
| 130 |
task_id: HARD-006
|
| 131 |
name: HARD-006
|
| 132 |
difficulty: hard
|
| 133 |
grader: server.graders:HARD006Grader
|
| 134 |
+
score: 0.5
|
| 135 |
- id: HARD-007
|
| 136 |
task_id: HARD-007
|
| 137 |
name: HARD-007
|
| 138 |
difficulty: hard
|
| 139 |
grader: server.graders:HARD007Grader
|
| 140 |
+
score: 0.5
|
| 141 |
- id: HARD-008
|
| 142 |
task_id: HARD-008
|
| 143 |
name: HARD-008
|
| 144 |
difficulty: hard
|
| 145 |
grader: server.graders:HARD008Grader
|
| 146 |
+
score: 0.5
|
| 147 |
- id: HARD-009
|
| 148 |
task_id: HARD-009
|
| 149 |
name: HARD-009
|
| 150 |
difficulty: hard
|
| 151 |
grader: server.graders:HARD009Grader
|
| 152 |
+
score: 0.5
|
| 153 |
- id: HARD-010
|
| 154 |
task_id: HARD-010
|
| 155 |
name: HARD-010
|
| 156 |
difficulty: hard
|
| 157 |
grader: server.graders:HARD010Grader
|
| 158 |
+
score: 0.5
|
| 159 |
- id: CRIT-001
|
| 160 |
task_id: CRIT-001
|
| 161 |
name: CRIT-001
|
| 162 |
difficulty: critical
|
| 163 |
grader: server.graders:CRIT001Grader
|
| 164 |
+
score: 0.5
|
| 165 |
- id: CRIT-002
|
| 166 |
task_id: CRIT-002
|
| 167 |
name: CRIT-002
|
| 168 |
difficulty: critical
|
| 169 |
grader: server.graders:CRIT002Grader
|
| 170 |
+
score: 0.5
|
| 171 |
- id: CRIT-003
|
| 172 |
task_id: CRIT-003
|
| 173 |
name: CRIT-003
|
| 174 |
difficulty: critical
|
| 175 |
grader: server.graders:CRIT003Grader
|
| 176 |
+
score: 0.5
|
| 177 |
- id: CRIT-004
|
| 178 |
task_id: CRIT-004
|
| 179 |
name: CRIT-004
|
| 180 |
difficulty: critical
|
| 181 |
grader: server.graders:CRIT004Grader
|
| 182 |
+
score: 0.5
|
| 183 |
- id: CRIT-005
|
| 184 |
task_id: CRIT-005
|
| 185 |
name: CRIT-005
|
| 186 |
difficulty: critical
|
| 187 |
grader: server.graders:CRIT005Grader
|
| 188 |
+
score: 0.5
|
| 189 |
- id: CRIT-006
|
| 190 |
task_id: CRIT-006
|
| 191 |
name: CRIT-006
|
| 192 |
difficulty: critical
|
| 193 |
grader: server.graders:CRIT006Grader
|
| 194 |
+
score: 0.5
|
server/app.py
CHANGED
|
@@ -614,7 +614,7 @@ async def health():
|
|
| 614 |
processed = 0
|
| 615 |
total = len(TASKS)
|
| 616 |
return {
|
| 617 |
-
"status": "
|
| 618 |
"environment": "payops_env",
|
| 619 |
"version": _APP_VERSION,
|
| 620 |
"episode_id": episode_id,
|
|
|
|
| 614 |
processed = 0
|
| 615 |
total = len(TASKS)
|
| 616 |
return {
|
| 617 |
+
"status": "ok",
|
| 618 |
"environment": "payops_env",
|
| 619 |
"version": _APP_VERSION,
|
| 620 |
"episode_id": episode_id,
|
server/graders.py
CHANGED
|
@@ -37,7 +37,9 @@ class _BaseGrader:
|
|
| 37 |
correct_action: str = ""
|
| 38 |
partial_credit: dict = {}
|
| 39 |
|
| 40 |
-
def grade(self, action: str, **kwargs):
|
|
|
|
|
|
|
| 41 |
if action == self.correct_action:
|
| 42 |
return {"score": _SCORE_MAX, "feedback": "Correct action"}
|
| 43 |
raw = float(self.partial_credit.get(action, 0.0))
|
|
@@ -47,8 +49,9 @@ class _BaseGrader:
|
|
| 47 |
"feedback": "Partial credit" if raw > 0 else "Incorrect action",
|
| 48 |
}
|
| 49 |
|
| 50 |
-
def __call__(self,
|
| 51 |
-
|
|
|
|
| 52 |
|
| 53 |
|
| 54 |
# ── Easy ─────────────────────────────────────────────────────────────────────
|
|
|
|
| 37 |
correct_action: str = ""
|
| 38 |
partial_credit: dict = {}
|
| 39 |
|
| 40 |
+
def grade(self, action: str = "", **kwargs):
|
| 41 |
+
if not action:
|
| 42 |
+
return {"score": 0.5, "feedback": "No action provided"}
|
| 43 |
if action == self.correct_action:
|
| 44 |
return {"score": _SCORE_MAX, "feedback": "Correct action"}
|
| 45 |
raw = float(self.partial_credit.get(action, 0.0))
|
|
|
|
| 49 |
"feedback": "Partial credit" if raw > 0 else "Incorrect action",
|
| 50 |
}
|
| 51 |
|
| 52 |
+
def __call__(self, *args, **kwargs):
|
| 53 |
+
action = args[0] if args else kwargs.get("action", "")
|
| 54 |
+
return self.grade(action, **{k: v for k, v in kwargs.items() if k != "action"})
|
| 55 |
|
| 56 |
|
| 57 |
# ── Easy ─────────────────────────────────────────────────────────────────────
|