{"data":{"kind":"file","path":"README.md","version_id":"e35jmzu27i2rxyxnq6ql7e6x","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1574,"modified_at":"2025-09-18T13:50:42.143000","content_hash":"0a0405800da177715df8bc369f43f5ff23339529442f158bce7e9db0c8baf883"},"entries":[],"content":"# traffic-environment\n\n### Overview\n- **Environment ID**: `traffic-environment`\n- **Short description**: A multi-turn traffic optimization environment for the Prime Intellect platform.\n- **Tags**: multi-turn, optimization, traffic-simulation\n\n### Datasets\n- **Primary dataset(s)**: monash_tsf, specifically the 'traffic_hourly' configuration (used for evaluation)\n- **Source links**: https://huggingface.co/datasets/monash_tsf\n- **Split sizes**: 10 examples from the 'train' split are used for evaluation.\n\n### Task\n- **Type**: multi-turn\n- **Parser**: XMLParser\n- **Rubric overview**: The rubric rewards traffic efficiency, penalizes congestion, and rewards minimizing wait times. It also includes a format reward.\n\n### Quickstart\nRun an evaluation with default settings:\n\n```bash\nuv run vf-eval traffic-environment\n```\n\nConfigure model and sampling:\n\n```bash\nuv run vf-eval traffic-environment \\\n  -m gpt-4.1-mini \\\n  -n 20 -r 3 -t 1024 -T 0.7 \\\n  -a '{\"key\": \"value\"}'  # env-specific args as JSON\n```\n\nNotes:\n- Use `-a` / `--env-args` to pass environment-specific configuration as a JSON object.\n\n### Environment Arguments\nThis environment does not currently support any environment-specific arguments.\n\n### Metrics\n| Metric | Meaning |\n| ------ | ------- |\n| `reward` | Main scalar reward (weighted sum of criteria) |\n| `efficiency_reward` | Reward based on traffic efficiency. |\n| `congestion_penalty` | Penalty for high congestion levels. |\n| `wait_time_reward` | Reward for minimizing wait times. |\n| `format_reward` | Reward for correctly formatting the output. |\n\n","encoding":"utf-8","truncated":false,"total_bytes":1574},"status":null}