frankenstallm / data /__init__.py
pathcosmos's picture
feat: Add data pipeline scripts + phase reports (Tier 3 - reproducibility)
b3d361d verified
"""
data package — dataset utilities for LLM training.
"""
from data.dataset import PackedDataset, TextDataset
from data.sft_dataset import SFTDataset
__all__ = ["TextDataset", "PackedDataset", "SFTDataset"]