Spaces:
Runtime error
Runtime error
File size: 1,611 Bytes
6dfa658 | 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 | """
Build a 150-question evaluation set from HuggingFace Turkish legal QA data.
Output format (JSONL):
{
"question": "...",
"gold_answer": "...",
"source_id": "HF_test_123"
}
"""
from __future__ import annotations
import json
from pathlib import Path
from datasets import load_dataset
def norm(text: str) -> str:
return " ".join(str(text).split()).strip()
def build_eval_set(output_path: Path, target_size: int = 150) -> int:
ds = load_dataset("Renicames/turkish-law-chatbot")
rows = []
# Held-out split for evaluation (do not index these rows in ``corpus_index.jsonl``).
split_name = "test"
if split_name not in ds:
raise RuntimeError("Expected split 'test' in Renicames/turkish-law-chatbot")
split = ds[split_name]
for idx, row in enumerate(split):
q = norm(row.get("Soru", ""))
a = norm(row.get("Cevap", ""))
if not q or not a:
continue
rows.append(
{
"question": q,
"gold_answer": a,
"source_id": f"HF_{split_name}_{idx}",
}
)
if len(rows) >= target_size:
break
output_path.parent.mkdir(parents=True, exist_ok=True)
with output_path.open("w", encoding="utf-8") as f:
for item in rows:
f.write(json.dumps(item, ensure_ascii=False) + "\n")
return len(rows)
if __name__ == "__main__":
root = Path(__file__).resolve().parent.parent
out = root / "data" / "eval_qa_150.jsonl"
n = build_eval_set(out, target_size=150)
print(f"[EvalSet] Wrote {n} items -> {out}")
|