| """Task 3 — Schedule Repair (Hard). |
| |
| The agent observes an infeasible scheduling instance and must return a |
| corrected schedule (JSON) that: |
| (a) is valid JSON with the required schema — 0.4 pts |
| (b) satisfies all scheduling constraints — 0.4 pts |
| (c) achieves a makespan within 30% of the known optimal— 0.2 pts |
| |
| Partial progress: parseable JSON earns 0.2 base reward per step. |
| Max steps per episode: 8. |
| Expected agent accuracy: ~30%. |
| """ |
|
|
| from __future__ import annotations |
|
|
| from typing import Any |
|
|
| from environment import INSTANCE_BANK, SchedulingOptEnv |
| from models import Action |
|
|
| TASK_ID = "schedule_repair" |
| MAX_STEPS = 8 |
| DIFFICULTY = "hard" |
|
|
|
|
| def run_episode(env: SchedulingOptEnv, agent_fn: Any) -> dict[str, Any]: |
| """Run a single schedule-repair episode. |
| |
| Args: |
| env: An initialized SchedulingOptEnv instance. |
| agent_fn: Callable receiving an Observation, returning a JSON schedule string. |
| |
| Returns: |
| Episode summary dict. |
| """ |
| obs = env.reset(task_id=TASK_ID) |
| total_reward = 0.0 |
| steps = 0 |
| info: dict[str, Any] = {} |
|
|
| for _ in range(MAX_STEPS): |
| response = agent_fn(obs) |
| action = Action(response=response, task_id=TASK_ID) |
| obs, reward, done, info = env.step(action) |
| total_reward += reward |
| steps += 1 |
| if done: |
| break |
|
|
| return { |
| "task": TASK_ID, |
| "difficulty": DIFFICULTY, |
| "steps": steps, |
| "total_reward": round(total_reward, 4), |
| "info": info, |
| } |
|
|
|
|
| def get_repairable_instances() -> list[dict[str, Any]]: |
| """Return instances that are infeasible and have known optimal schedules.""" |
| return [ |
| { |
| "instance": entry["instance"], |
| "optimal_schedule": entry["optimal_schedule"], |
| "optimal_makespan": entry["optimal_makespan"], |
| "violation_type": entry["violation_type"], |
| "description": entry["description"], |
| } |
| for entry in INSTANCE_BANK |
| if not entry["is_feasible"] and entry.get("optimal_schedule") |
| ] |
|
|