File size: 1,543 Bytes
f764cdf 4b10a0d f764cdf 4b10a0d f764cdf 4b10a0d f764cdf 4b10a0d f764cdf | 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 | """Build repository training samples from a local checkout."""
from __future__ import annotations
import argparse
from pathlib import Path
from .builder import build_samples
from .model import BuildConfig
from .repo import repo_from_path, scan
from .serialization import write_parquet
from .tokenizer import DEFAULT_TOKENIZER_ID, load_tokenizer
from .repo_serialization import write_json
def main() -> None:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("repository", type=Path)
parser.add_argument("output", type=Path)
parser.add_argument("--info-output", type=Path,
help="also write dependency metadata JSON for this repository")
parser.add_argument("--max-tokens", type=int, default=8_192)
parser.add_argument("--tokenizer", default=DEFAULT_TOKENIZER_ID,
help="Transformers tokenizer name or local path")
args = parser.parse_args()
repo = repo_from_path(args.repository)
info = scan(repo, include_weak=False)
config = BuildConfig(max_tokens=args.max_tokens)
tokenizer = load_tokenizer(args.tokenizer)
samples = build_samples(repo, config=config, tokenizer=tokenizer, info=info)
args.output.parent.mkdir(parents=True, exist_ok=True)
write_parquet(samples, args.output)
if args.info_output:
args.info_output.parent.mkdir(parents=True, exist_ok=True)
write_json(info, args.info_output)
print(f"wrote {len(samples)} samples to {args.output}")
if __name__ == "__main__":
main()
|