overlay/scripts/benchmark_datasets.py

#!/usr/bin/env python3
from __future__ import annotations

from pathlib import Path


CANONICAL_SUBSETS = {
    "MBPP": Path("data/benchmarks/mbpp.cycle1.jsonl"),
    "GSM8K": Path("data/benchmarks/gsm8k.cycle1.jsonl"),
}


def resolve_benchmark_dataset(benchmark_name: str, explicit_path: Path | None) -> Path:
    if explicit_path is not None:
        return explicit_path
    if benchmark_name not in CANONICAL_SUBSETS:
        raise ValueError(f"Unsupported benchmark dataset: {benchmark_name}")
    return Path.cwd() / CANONICAL_SUBSETS[benchmark_name]