Improved KG integration, semantic calculations, dynamic temp adjustment, llm config and lots of other things to make it a bit more solid. I am going to start refactoring and making into a proper codebase so there may be few updates till I do so. Here is my proposed structure. Feel free to offer opinions and ideas:
semantic_cascade/
├── __init__.py
├── core/
│ ├── __init__.py
│ ├── config.py
│ ├── models.py
│ └── exceptions.py
├── processors/
│ ├── __init__.py
│ ├── semantic_processor.py
│ ├── knowledge_processor.py
│ └── layer_processor.py
├── services/
│ ├── __init__.py
│ ├── llm_service.py
│ ├── vectorization_service.py
│ └── graph_service.py
├── repositories/
│ ├── __init__.py
│ ├── knowledge_repository.py
│ └── conversation_repository.py
├── utils/
│ ├── __init__.py
│ ├── text_utils.py
│ └── graph_utils.py
└── interfaces/
├── __init__.py
└── processor_interface.py