Datasets:
File size: 3,711 Bytes
b48c9f8 | 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 | """Turn bench result files into the Hugging Face dataset folder.
python3 build_dataset.py OUT_DIR results/*.jsonl
Writes, under OUT_DIR:
data/cases.jsonl one row per case-run (the table most people want)
data/calls.jsonl one row per model request
data/runs.jsonl one row per bench invocation (model, settings, versions)
raw/<label>.jsonl the bench's own files, unchanged except for scrubbed local paths
summary/ summarize.py's tables over everything
code/ the bench, the summarizer, the Kaggle runner and this script
Nested fields (input, expected, output, mismatches, path, ...) are stored as JSON strings in
data/, so every row of a table has the same columns and types. Hugging Face's viewer and
`datasets.load_dataset` both need that; the raw/ files keep the original shapes.
"""
import json
import os
import re
import shutil
import subprocess
import sys
HERE = os.path.dirname(os.path.abspath(__file__))
KAGGLE_RUNNER = os.path.join(HERE, "kaggle_runner.py")
# Anything that looks like a path on the machine that ran the bench. The Kaggle paths are
# kept: /kaggle/working is the same for everyone and says where a run happened.
LOCAL_PATH = re.compile(r"/(?:Users|home)/[^\"\s]+")
NESTED = {
"case": ("input", "expected", "output", "mismatches", "path", "gates", "escalations",
"rescued_expected"),
"call": ("output",),
"run": ("arms", "packs", "extra_body", "setup", "pack_digests", "stepmold", "bench"),
}
def scrub(text):
return LOCAL_PATH.sub("<local checkout>", text)
def flatten(record):
row = dict(record)
for key in NESTED.get(record.get("kind"), ()):
if key in row and row[key] is not None and not isinstance(row[key], str):
row[key] = json.dumps(row[key], sort_keys=True)
return row
def main(out, files):
for sub in ("data", "raw", "summary", "code"):
os.makedirs(os.path.join(out, sub), exist_ok=True)
tables = {"case": [], "call": [], "run": []}
for path in files:
label = os.path.splitext(os.path.basename(path))[0]
with open(path) as handle, open(os.path.join(out, "raw", label + ".jsonl"), "w") as raw:
for line in handle:
line = scrub(line)
raw.write(line)
record = json.loads(line)
if record.get("kind") in tables:
tables[record["kind"]].append(flatten(record))
names = {"case": "cases", "call": "calls", "run": "runs"}
for kind, rows in tables.items():
with open(os.path.join(out, "data", names[kind] + ".jsonl"), "w") as handle:
for row in rows:
handle.write(json.dumps(row, sort_keys=True) + "\n")
print("%-6s %6d rows" % (names[kind], len(rows)))
subprocess.run([sys.executable, os.path.join(HERE, "summarize.py")] + list(files), check=True,
stdout=subprocess.DEVNULL)
for name in ("summary.md", "summary.csv"):
shutil.move(os.path.join(os.path.dirname(files[0]), name), os.path.join(out, "summary", name))
for name in ("bench.py", "summarize.py", "build_dataset.py", "README.md"):
shutil.copy(os.path.join(HERE, name), os.path.join(out, "code", name))
if os.path.exists(KAGGLE_RUNNER):
shutil.copy(KAGGLE_RUNNER, os.path.join(out, "code", "kaggle_runner.py"))
leaks = [p for p in (os.path.join(dp, f) for dp, _, fs in os.walk(out) for f in fs)
if LOCAL_PATH.search(open(p, errors="ignore").read()) and "/code/" not in p]
print("local paths left outside code/:", leaks or "none")
if __name__ == "__main__":
if len(sys.argv) < 3:
sys.exit(__doc__)
main(sys.argv[1], sys.argv[2:])
|