- Move CheckpointService import inside run_raptor_with_checkpoint function - Prevents module-level import that could cause initialization issues - Improves modularity and reduces coupling Note: task_executor.py has pre-existing NLTK dependencies from resume module that may require NLTK data in test environments. This is unrelated to checkpoint feature. |
||
|---|---|---|
| .. | ||
| app | ||
| flow | ||
| llm | ||
| nlp | ||
| prompts | ||
| res | ||
| svr | ||
| utils | ||
| __init__.py | ||
| benchmark.py | ||
| raptor.py | ||
| settings.py | ||