mikav / scripts /prepare_dataset.py
mikav-dev's picture
Initial Mikav project scaffold
0912b10 verified
Raw
History Blame Contribute Delete
284 Bytes
"""Prepare tokenized datasets for pretraining."""
from pathlib import Path
def main() -> None:
output_dir = Path("data/tokenized")
output_dir.mkdir(parents=True, exist_ok=True)
print("Add dataset loading and tokenization here.")
if __name__ == "__main__":
main()