xming521 e90f05c12a refactor(data): improve data cleaning robustness ...

Allow `vllm_infer` to return `None` for failed JSON parsing results,
clarifying the sequence index for such failures. This enables
`LLMCleaningStrategy` to robustly align parsed LLM scores with original data,
correctly handling examples where JSON parsing failed by assigning a default
score of 0.

Adds an option to disable data cleaning during training setup.

2025-08-17 14:30:17 +08:00

..

__init__.py

重构项目

2025-04-21 20:30:55 +08:00

export_model.py

重构项目

2025-04-21 20:30:55 +08:00

train_sft.py

refactor(data): improve data cleaning robustness

2025-08-17 14:30:17 +08:00