Files
AstrAI/astrai/dataset/__init__.py
T
ViperEkura 553a42702d refactor: replace diamond inheritance with mixin composition
- StreamStore/RecordStore → Streamable/Recordable (stateless mixins)
- Store is sole base class, no MRO ambiguity
- H5Store/MmapStore/JsonlStore mix in both traits explicitly
- segments_are_records declared per-subclass (H5/Jsonl=True, bin=False)
- Add tests for dpo_tokenize, lazy jsonl, dual-mode H5, stream-only bin
- Remove unused _to_tensor helper
2026-07-18 23:20:41 +08:00

44 lines
736 B
Python

from astrai.dataset.dataset import (
BaseDataset,
DatasetFactory,
dpo_collate_fn,
grpo_collate_fn,
)
from astrai.dataset.sampler import RDSampler
from astrai.dataset.storage import (
H5Store,
JsonlStore,
MmapStore,
Recordable,
Store,
StoreFactory,
Streamable,
detect_format,
)
from astrai.serialization import (
load_bin,
load_h5,
save_bin,
save_h5,
)
__all__ = [
"BaseDataset",
"DatasetFactory",
"dpo_collate_fn",
"grpo_collate_fn",
"Store",
"Streamable",
"Recordable",
"StoreFactory",
"H5Store",
"MmapStore",
"JsonlStore",
"detect_format",
"save_h5",
"load_h5",
"save_bin",
"load_bin",
"RDSampler",
]