refactor: 重构数据管道模块

This commit is contained in:
2026-03-20 21:36:37 +08:00
parent 7045de9c32
commit 3fb2326115
20 changed files with 739 additions and 345 deletions
+17
View File
@@ -0,0 +1,17 @@
from .tokenizer import BpeTokenizer
from .datapipeline import (
DataPipeline,
ProcessorFactory,
IOHandler,
TextNormalizer,
SequencePacker
)
__all__ = [
'BpeTokenizer',
'DataPipeline',
'ProcessorFactory',
'IOHandler',
'TextNormalizer',
'SequencePacker'
]