Improvements​

• Various improvements to the optimal actions optimisation algorithm.
• Upgraded Node.js client to align more closely to the Python client.
• Improved observability and monitoring functionalities.
• UI builder and explainability features complete, awaiting release.

Features​

• Added bounds support across models to prevent predictions outside of expected ranges.
• Internal data simulation for much faster testing and development.
• Internal logging and debugging system for faster issue resolution.
• Added constraints to find_best_actions to prevent invalid actions & outcomes.
• Added batch support for find_best_actions to make it easier to find best actions over multiple scenarios.
• Add support for weighted multi-objective optimisation in find_best_actions.

Improvements​

• Internal refactoring of server codebase for faster development cycles.
• Upgraded Node.js client to work with latest version of the server.
• Improved error handling for invalid user inputs.

March has been a busy month for the CausaDB team. We've been working hard to bring you the latest updates and improvements to the platform. Here's a summary of what's new in CausaDB this month.

Features​

• Major core functionality updates with improved speed and performance.
• Support for fixed argument in find_best_actions.
• Support for both continuous and binary data types.
• Support for pandas data loading.
• Improved containerisation for easier self-hosting.
• Added smart building heating example to documentation.
• Added updates blog (what you're reading right now!) to documentation site.

Improvements​

• Various improvements to user experience including renaming of functions and arguments.
• Improved documentation and examples.
• Improved error handling for invalid user inputs.
• Atomic artefact management for robust porting of models between versions.
• Various polishing and improvements to documentation site.