fix: 修复特殊token 问题

This commit is contained in:
2026-04-02 16:16:02 +08:00
parent e01ec081b3
commit f44ad6912e
28 changed files with 334 additions and 163 deletions
+18 -13
View File
@@ -5,24 +5,29 @@ from pipeline.packing import SequencePacker
from pipeline.io import IOHandler, export_dataset, cache_jsonl
from pipeline.processors import ProcessorFactory, BaseProcessor
from pipeline.utils import setup_logging
from pipeline.strategies import PromptStrategy, ChatMLStrategy, AlpacaStrategy, StrategyFactory
from pipeline.strategies import (
PromptStrategy,
ChatMLStrategy,
AlpacaStrategy,
StrategyFactory,
)
# Configure project-level logging
setup_logging()
__all__ = [
# Core modules
'BpeTokenizer',
'TextNormalizer',
'SequencePacker',
'IOHandler',
'ProcessorFactory',
'BaseProcessor',
'export_dataset',
'cache_jsonl',
"BpeTokenizer",
"TextNormalizer",
"SequencePacker",
"IOHandler",
"ProcessorFactory",
"BaseProcessor",
"export_dataset",
"cache_jsonl",
# Strategy pattern
'PromptStrategy',
'ChatMLStrategy',
'AlpacaStrategy',
'StrategyFactory',
"PromptStrategy",
"ChatMLStrategy",
"AlpacaStrategy",
"StrategyFactory",
]