Reconstruct setting in Rerank,Vector,Graph #202
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Introduced a unified configuration system for LLM providers using pydantic-settings. - Added provider-specific settings for Anthropic, Azure, DeepSeek, Ollama, OpenAI, Qwen, Vllm, and Zai. - Improved environment variable handling and validation through Field and AliasChoices. - Removed legacy initialization methods in favor of a cleaner, more maintainable structure. - Updated LLMFactory to utilize the new provider registration mechanism.
…riable handling - Refactor LLM configuration imports to use BaseLLMConfig. - Replace direct attribute access with getattr for safer environment variable retrieval. - Remove deprecated LLMConfig and streamline related code for better maintainability.
- Introduced BaseRerankConfig for improved configuration management across rerank providers. - Updated rerank integration files to utilize the new base configuration structure. - Added support for additional configuration fields such as api_base_url and top_n. - Refactored rerank factory to accommodate new configuration handling and provider registration. - Removed deprecated RerankConfig and streamlined related code for better maintainability. - Updated API request handling in rerank classes to support custom HTTP clients.
…module - Removed `VectorStoreConfig` and `GraphStoreConfig` classes - Deleted associated validation logic and import statements - Streamlined codebase by eliminating unused components
- Introduced BaseSparseEmbedderConfig for unified sparse embedding configuration. - Updated MemoryConfig to utilize BaseSparseEmbedderConfig. - Refactored SparseEmbedderFactory to support new configuration handling. - Improved handling of sparse embedder settings across various components.
- Added a registry mechanism to UserProfileStoreBase for automatic provider registration. - Implemented class paths for OceanBase and SQLite user profile storage implementations. - Updated UserProfileStoreFactory to utilize the new registry for provider class retrieval. - Refactored imports to trigger auto-registration of user profile storage classes. - Improved handling of provider names in the factory for better compatibility.
…rations - Added logic to sync `embedding_model_dims` from the embedder to both `vector_store` and `graph_store` if not already set. - Updated `config_loader.py` and `configs.py` to ensure consistent embedding dimensions across components.
- Added `enable_native_hybrid` field to `OceanBaseConfig` for native hybrid search support. - Updated query handling in `OceanBaseVectorStore` to use a safe query format, preventing SQL injection risks.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
Reconstruct setting in Rerank,Vector,Graph