Weaviate integration

[Vargha-Kh]

Pull Request Description

Summary of Changes

• Added Weaviate VectorDb Integration: Introduced a new Weaviate class under phi/vectordb/weaviate/weaviate.py, enabling seamless integration with Weaviate vector database.
• Implemented Required Abstract Methods: Ensured compliance with the base VectorDb interface by implementing methods like doc_exists, name_exists, exists, insert, and search.
• Enhanced Support for Vector, Keyword, and Hybrid Search: Provides different search strategies (vector, keyword, and hybrid) leveraging user-configurable embeddings and Weaviate’s query features.
• Lazy-Loaded Weaviate Client: Uses a property (@property) for client to defer creation of the Weaviate client until first use, matching the pattern used in other vector DB integrations (e.g., Pinecone, Milvus).
• Schema Creation and Management: Allows users to create, drop, and check for existing Weaviate classes (indexes) via the create() and drop() methods.

Related Issues

• I personally needed supporting Weaviate as a vector database in phiData.

Motivation and Context

• Centralized Vector Database Options: Extends phiData’s supported vector databases, offering a flexible choice for managing high-dimensional embeddings.
• Ease of Use: Simplifies setup for developers who want to combine Weaviate with phiData’s agent system without manually creating or managing Weaviate integrations.
• Search Variety: Supports vector-based semantic queries, direct keyword matching, and a rudimentary hybrid approach that merges results from both.

Environment or Dependencies

• Weaviate Python Client: Requires weaviate-client>=4.4.0,<5.0 (or an appropriate version of Weaviate client v4).

Impact on AI/ML Components

• Model Embedding Usage: This PR does not change existing AI models but provides an additional storage/search mechanism for AI-generated embeddings.

Type of Change

• New feature (non-breaking change which adds functionality)

Checklist

• My code follows Phidata's style guidelines and best practices.
• I have performed a self-review of my code.
• I have added docstrings and comments for complex logic.
• My changes generate no new warnings or errors.
• I have added cookbook examples for my new addition (if needed).
• I have verified my changes in a clean environment.

Additional Notes

• Deployment: Local deployments using a standard Weaviate container from tutorial (https://weaviate.io/developers/weaviate/installation/docker-compose) at http://localhost:8080 should work out of the box.
• Customization: Users can extend or modify the schema or distance configurations in create() to match their Weaviate cluster setup.
• Documentation: Future enhancements might add advanced hybrid search capabilities (e.g., text + vector weighting) and more robust filtering.

[manthanguptaa]

Hey @Vargha-Kh! We recently rebranded to Agno, which means that this PR needs to be ported to Agno. Let me know if you have questions or need help. Sorry for the inconvenience caused