Spaces:
Running
Running
File size: 10,540 Bytes
50f71a7 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 | """Pre-submission validator for the Fake Gang Detection OpenEnv environment.
Checks all submission requirements and prints pass/fail for each.
Exits 0 if all checks pass, 1 if any fail.
Usage:
python validate.py # server must be running on :8000
python validate.py --url http://host:8001
python validate.py --local # skip HTTP checks, test locally
"""
from __future__ import annotations
import argparse
import importlib
import json
import sys
import time
import urllib.error
import urllib.request
from pathlib import Path
from typing import Any, Dict, List, Optional, Tuple
_ROOT = Path(__file__).parent
sys.path.insert(0, str(_ROOT))
sys.path.insert(0, str(_ROOT / "server"))
# ---------------------------------------------------------------------------
# Helpers
# ---------------------------------------------------------------------------
_PASS = "[PASS]"
_FAIL = "[FAIL]"
_results: List[Tuple[bool, str]] = []
def check(name: str, ok: bool, detail: str = "") -> bool:
tag = _PASS if ok else _FAIL
line = f" {tag} {name}"
if detail and not ok:
line += f"\n → {detail}"
print(line)
_results.append((ok, name))
return ok
def _get(url: str) -> Tuple[Optional[dict], Optional[str]]:
try:
with urllib.request.urlopen(url, timeout=10) as r:
return json.loads(r.read()), None
except Exception as exc:
return None, str(exc)
def _post(url: str, body: Any = None) -> Tuple[Optional[dict], Optional[str]]:
try:
data = json.dumps(body or {}).encode()
req = urllib.request.Request(
url, data=data, headers={"Content-Type": "application/json"}, method="POST"
)
with urllib.request.urlopen(req, timeout=30) as r:
return json.loads(r.read()), None
except urllib.error.HTTPError as exc:
try:
body_bytes = exc.read()
return None, f"HTTP {exc.code}: {body_bytes.decode()}"
except Exception:
return None, f"HTTP {exc.code}"
except Exception as exc:
return None, str(exc)
# ---------------------------------------------------------------------------
# HTTP checks
# ---------------------------------------------------------------------------
def run_http_checks(base_url: str) -> None:
print(f"\nHTTP checks against {base_url}\n")
# /health
data, err = _get(f"{base_url}/health")
check("/health reachable", data is not None and data.get("status") == "healthy", err or "")
# /tasks — must have action_schema and 3 tasks
data, err = _get(f"{base_url}/tasks")
if check("/tasks reachable", data is not None, err or ""):
has_schema = isinstance(data.get("action_schema"), dict)
has_3_tasks = isinstance(data.get("tasks"), list) and len(data["tasks"]) == 3
has_score_range = "score_range" in data
check("/tasks has action_schema", has_schema, str(data))
check("/tasks has 3 tasks", has_3_tasks, str(data))
check("/tasks has score_range", has_score_range, str(data))
# /reset for each task
for task in ["easy", "medium", "hard"]:
data, err = _post(f"{base_url}/reset", {"task": task, "seed": 0})
check(f"/reset task={task}", data is not None and "observation" in data, err or "")
# /step — INSPECT, FLAG, SUBMIT cycle
_post(f"{base_url}/reset", {"task": "easy", "seed": 0})
obs_resp, err = _post(f"{base_url}/step", {"action_type": "inspect",
"account_id": "acc_0000"})
check("/step INSPECT", obs_resp is not None, err or "")
# Get a visible account ID from the observation to flag
acc_to_flag = None
if obs_resp:
vis_ids = obs_resp.get("observation", {}).get("visible_account_ids", [])
if vis_ids:
acc_to_flag = vis_ids[0]
if acc_to_flag:
flag_resp, err = _post(f"{base_url}/step", {"action_type": "flag",
"account_id": acc_to_flag})
check("/step FLAG", flag_resp is not None, err or "")
sub_resp, err = _post(f"{base_url}/step", {"action_type": "submit"})
check("/step SUBMIT", sub_resp is not None and sub_resp.get("done") is True, err or "")
# /grader — must return float in [0, 1]
data, err = _get(f"{base_url}/grader")
if check("/grader reachable", data is not None, err or ""):
score = data.get("score")
check("/grader returns [0,1] float",
isinstance(score, (int, float)) and 0.0 <= score <= 1.0,
f"score={score}")
# /baseline — must return 3 task scores in [0, 1]
data, err = _post(f"{base_url}/baseline")
if check("/baseline reachable", data is not None, err or ""):
scores = data.get("scores", {})
all_valid = (
set(scores.keys()) == {"easy", "medium", "hard"}
and all(isinstance(v, (int, float)) and 0.0 <= v <= 1.0
for v in scores.values())
)
check("/baseline returns 3 valid scores", all_valid,
f"got: {scores}")
# ---------------------------------------------------------------------------
# Local checks (no server needed)
# ---------------------------------------------------------------------------
def run_local_checks() -> None:
print("\nLocal checks\n")
# scoring.py importable and correct
try:
from scoring import ( # type: ignore[import]
compute_fake_risk, compute_hub_legitimacy, grader_score
)
gang_risk = compute_fake_risk(0.75, 0.65, 0.85, 0.10)
hub = compute_hub_legitimacy(2_000_000, 200, 2000, 0.05)
celeb_risk = compute_fake_risk(0.02, 0.02, 0.10, hub)
# Perfect score: 10 TP, 0 FP, 0 FN, 0 steps used → efficiency=1.0 → score=1.0
perfect = grader_score(10, 0, 0, 0, 30)
ok = (gang_risk >= 0.60 and celeb_risk < 0.20 and perfect == 1.0)
check("scoring.py math correct", ok,
f"gang_risk={gang_risk} celeb_risk={celeb_risk} perfect={perfect}")
except Exception as exc:
check("scoring.py importable", False, str(exc))
# models.py has AccountStatus + new fields
try:
from models import AccountStatus, AccountProfile, FakeGangObservation # type: ignore[import]
p = AccountProfile(
account_id="acc_0001", follower_count=100, following_count=50,
post_count=10, avg_post_hour=14.0, photo_reuse_score=0.8,
bio_template_score=0.7, account_age_days=60,
)
check("models.py AccountProfile has fake_risk_score",
hasattr(p, "fake_risk_score"), "")
check("models.py FakeGangObservation has suspect_ids",
hasattr(FakeGangObservation(), "suspect_ids"), "")
check("models.py AccountStatus enum exists",
AccountStatus.SUSPECT == "suspect", "")
except Exception as exc:
check("models.py new fields", False, str(exc))
# environment.py runs episode + status cascade
try:
from environment import FakeGangEnvironment # type: ignore[import]
from models import FakeGangAction, ActionType # type: ignore[import]
env = FakeGangEnvironment()
obs = env.reset(task="easy", seed=0)
ep_path = _ROOT / "episodes" / "easy_000.json"
if ep_path.exists():
gang_id = json.loads(ep_path.read_text())["gang_member_ids"][0]
obs = env.step(FakeGangAction(action_type=ActionType.INSPECT, account_id=gang_id))
obs = env.step(FakeGangAction(action_type=ActionType.FLAG, account_id=gang_id))
cascade_ok = len(obs.suspect_ids) > 0
check("environment.py SUSPECT cascade works", cascade_ok,
f"suspect_ids={obs.suspect_ids[:3]}")
p_flagged = next((p for p in obs.visible_accounts if p.account_id == gang_id), None)
check("environment.py fake_risk_score computed",
p_flagged is not None and p_flagged.fake_risk_score > 0, "")
else:
check("episode file exists for cascade test", False,
f"run python server/generator.py first")
except Exception as exc:
check("environment.py status cascade", False, str(exc))
# inference.py importable + runs one episode locally
try:
from inference import run_rule_based_episode # type: ignore[import]
from environment import FakeGangEnvironment # type: ignore[import]
env2 = FakeGangEnvironment()
score = run_rule_based_episode(env2, task="easy", seed=1)
check("inference.py runs locally",
isinstance(score, float) and 0.0 <= score <= 1.0,
f"score={score}")
except Exception as exc:
check("inference.py importable", False, str(exc))
# Episodes have new features
ep_path = _ROOT / "episodes" / "easy_000.json"
if ep_path.exists():
ep = json.loads(ep_path.read_text())
accounts = ep["network"]["accounts"]
first = accounts[0]["features"]
has_features = "comment_repeat_score" in first and "shared_ip_count" in first
check("episodes have new features (comment_repeat_score, shared_ip_count)",
has_features, f"keys: {list(first.keys())}")
has_celebs = "celeb_ids" in ep
check("episodes have celeb_ids field", has_celebs, "")
else:
check("episodes directory has files", False, "run python server/generator.py")
# ---------------------------------------------------------------------------
# Main
# ---------------------------------------------------------------------------
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("--url", default="http://localhost:8000")
parser.add_argument("--local", action="store_true",
help="Run local checks only (no server needed)")
args = parser.parse_args()
run_local_checks()
if not args.local:
run_http_checks(args.url)
total = len(_results)
passed = sum(1 for ok, _ in _results if ok)
failed = total - passed
print(f"\n{'='*50}")
print(f"Results: {passed}/{total} passed", end="")
if failed:
print(f" ({failed} FAILED)")
failed_names = [name for ok, name in _results if not ok]
for name in failed_names:
print(f" - {name}")
print()
sys.exit(1)
else:
print(" — all OK")
print()
sys.exit(0)
|