File size: 9,205 Bytes
778d47d | 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 | """
v3 combined SFT data: BIRD-train paper rollouts (with fb_*) + SynSQL synthetic.
Reads:
- eval_results/paper_SFT_VF_passAt8_bird_TRAIN.jsonl (after pipeline regen 89417)
- data/external/synsql/synsql_candidates_30k.jsonl (after SynSQL gen 89486)
Writes:
- data/sft_selector_v3_combined/{train,test}
Format: pointwise YES/NO with rich-schema + fb_* (when available, else "None").
Includes "Planner SQL executed: YES/NO" line to expose planner_exec_ok signal.
For SynSQL, no exec/fb data — fields are 'None' / 'synthetic'. Schema is generated
via render_rich_schema if BIRD-style schema dict available, else a minimal description.
"""
import argparse, json, os, re, sys, random
from concurrent.futures import ThreadPoolExecutor, as_completed
os.environ.setdefault("PYTHONNOUSERSITE", "1")
os.environ.setdefault("DB_EXEC_API_DISABLE", "1")
ROOT = "/weka/s225250685/mats-tist"
os.chdir(ROOT); sys.path.insert(0, ROOT)
from validator_data.validator import _execute_sql
from datasets import Dataset, DatasetDict
from scripts.rich_schema import render_rich_schema
POINTWISE_PROMPT = (
"You are a SQL correctness judge.\n"
"Database Schema (with column meanings, value descriptions, and example values):\n"
"{schema}\n\n"
"Question: {question}\n"
"External knowledge: {evidence}\n\n"
"Candidate SQL:\n{sql}\n\n"
"Execution result of the candidate:\n{exec_result}\n\n"
"Planner SQL executed without error: {exec_ok}\n\n"
"Validator critique of the planner draft (for context):\n"
" - select: {fb_select}\n"
" - condition: {fb_condition}\n"
" - join: {fb_join}\n"
" - order: {fb_order}\n\n"
"Does this SQL correctly answer the question, given the schema, the column "
"descriptions, the external knowledge, the execution result, and the validator's critique? "
"Answer YES or NO."
)
MAX_SCHEMA_CHARS = 3000
def safe_truncate(s, n):
s = str(s) if s is not None else ""
return s if len(s) <= n else s[:n] + "..."
def exec_str(db_path, sql, timeout=8):
if not sql or not sql.strip(): return "Error: empty SQL"
try:
r, err = _execute_sql("./" + db_path if not db_path.startswith("./") else db_path, sql, timeout=timeout)
except Exception as e:
return f"Error: {str(e)[:160]}"
if err: return f"Error: {str(r)[:160]}"
rows = str(r)[:260]
return f"OK. Rows preview: {rows}" if rows.strip() and rows.strip() != "[]" else "OK. (no rows returned)"
def render_bird(sample, t, schema_text):
sql_fixed = (t.get("fixed_sql") or "").strip()
sql = sql_fixed or (t.get("planner_sql") or "").strip()
if not sql: return None
is_correct = bool(t.get("is_fixed_correct") if sql_fixed else t.get("is_planner_correct"))
ex = exec_str(sample["db_path"], sql)
label = "YES" if is_correct else "NO"
exec_ok = "YES" if t.get("planner_exec_ok") else "NO"
prompt = POINTWISE_PROMPT.format(
schema=schema_text,
question=sample.get("question", ""),
evidence=sample.get("evidence", "") or "None",
sql=safe_truncate(sql, 800),
exec_result=safe_truncate(ex, 300),
exec_ok=exec_ok,
fb_select=safe_truncate(t.get("fb_select") or "None", 200),
fb_condition=safe_truncate(t.get("fb_condition") or "None", 200),
fb_join=safe_truncate(t.get("fb_join") or "None", 200),
fb_order=safe_truncate(t.get("fb_order") or "None", 200),
)
return {
"prompt": prompt, "completion": label,
"messages": [{"role": "user", "content": prompt},
{"role": "assistant", "content": label}],
"question": sample.get("question", ""),
"db_id": sample.get("db_id", ""),
"is_yes": int(label == "YES"),
"source": "bird_train",
}
def render_synsql(rec, cand):
"""Render SynSQL synthetic — no exec data, no fb_*."""
sql = cand.get("sql", "").strip()
if not sql: return None
label = "YES" if cand.get("is_correct") else "NO"
# Minimal schema (just db_id name) since we don't have full schema for SynSQL
schema_text = f"(SynSQL database: {rec.get('db_id', 'unknown')}; schema dump unavailable for this synthetic source.)"
prompt = POINTWISE_PROMPT.format(
schema=schema_text,
question=rec.get("question", ""),
evidence=rec.get("evidence", "") or "None",
sql=safe_truncate(sql, 800),
exec_result="(synthetic: no execution available)",
exec_ok="(unknown)",
fb_select="None", fb_condition="None", fb_join="None", fb_order="None",
)
return {
"prompt": prompt, "completion": label,
"messages": [{"role": "user", "content": prompt},
{"role": "assistant", "content": label}],
"question": rec.get("question", ""),
"db_id": rec.get("db_id", ""),
"is_yes": int(label == "YES"),
"source": "synsql",
}
def main():
ap = argparse.ArgumentParser()
ap.add_argument("--bird", default="eval_results/paper_SFT_VF_passAt8_bird_TRAIN.jsonl")
ap.add_argument("--synsql", default="data/external/synsql/synsql_candidates_30k.jsonl")
ap.add_argument("--out", default="data/sft_selector_v3_combined")
ap.add_argument("--use_synsql", action="store_true", default=True)
ap.add_argument("--no_synsql", dest="use_synsql", action="store_false")
args = ap.parse_args()
rng = random.Random(42)
records = []
n_yes = n_no = 0
# --- BIRD-train ---
if os.path.exists(args.bird):
print(f"Reading BIRD-train rollouts: {args.bird}", flush=True)
bird_jobs = []
schema_cache = {}
n_q = 0
with open(args.bird) as f:
for line in f:
line = line.strip()
if not line: continue
s = json.loads(line)
n_q += 1
seen = set()
for t in s.get("trajectories", []):
sql_fixed = (t.get("fixed_sql") or "").strip()
sql = sql_fixed or (t.get("planner_sql") or "").strip()
if not sql: continue
norm = re.sub(r"\s+", " ", sql.lower())
if norm in seen: continue
seen.add(norm)
bird_jobs.append((s, t))
print(f" BIRD jobs: {len(bird_jobs)} (from {n_q} questions)", flush=True)
for s, _ in bird_jobs:
if s["db_id"] not in schema_cache:
schema_cache[s["db_id"]] = safe_truncate(render_rich_schema(s, split="train"), MAX_SCHEMA_CHARS)
with ThreadPoolExecutor(max_workers=32) as exe:
futs = [exe.submit(render_bird, s, t, schema_cache[s["db_id"]]) for s, t in bird_jobs]
for fut in as_completed(futs):
try: r = fut.result()
except Exception: r = None
if r is None: continue
records.append(r)
if r["is_yes"]: n_yes += 1
else: n_no += 1
print(f" BIRD records: {sum(1 for r in records if r['source']=='bird_train')} (Y={n_yes}, N={n_no})", flush=True)
else:
print(f"BIRD file MISSING: {args.bird}", flush=True)
# --- SynSQL ---
if args.use_synsql and os.path.exists(args.synsql):
print(f"Reading SynSQL: {args.synsql}", flush=True)
n_syn = 0
with open(args.synsql) as f:
for line in f:
line = line.strip()
if not line: continue
rec = json.loads(line)
seen = set()
for c in rec.get("candidates", []):
sql_norm = re.sub(r"\s+", " ", (c.get("sql") or "").strip().lower())
if not sql_norm or sql_norm in seen: continue
seen.add(sql_norm)
r = render_synsql(rec, c)
if r:
records.append(r)
if r["is_yes"]: n_yes += 1
else: n_no += 1
n_syn += 1
print(f" SynSQL records: {n_syn}", flush=True)
else:
print("SynSQL skipped", flush=True)
print(f"\nTotal: {len(records)} records (Y={n_yes}, N={n_no})", flush=True)
# Balance NO ≤ 1.2 * YES
yes_r = [r for r in records if r["is_yes"]]
no_r = [r for r in records if not r["is_yes"]]
rng.shuffle(no_r)
keep_no = no_r[: min(len(no_r), int(1.2 * len(yes_r)))]
final = yes_r + keep_no
rng.shuffle(final)
print(f"After balance: {len(final)} (Y={len(yes_r)}, N={len(keep_no)})", flush=True)
by_q = {}
for r in final:
by_q.setdefault((r["question"], r["db_id"]), []).append(r)
qs = list(by_q.keys()); rng.shuffle(qs)
n_test_q = max(80, len(qs) // 50)
test_qs = set(qs[:n_test_q])
train, test = [], []
for k, recs in by_q.items():
(test if k in test_qs else train).extend(recs)
rng.shuffle(train); rng.shuffle(test)
print(f"train: {len(train)} test: {len(test)}")
DatasetDict({"train": Dataset.from_list(train), "test": Dataset.from_list(test)}).save_to_disk(args.out)
print(f"SAVED: {args.out}")
if __name__ == "__main__":
main()
|