In InterSystems IRIS 2024.3 and subsequent IRIS versions, the AutoML component is now delivered as a separate Python package that is installed after installation. Unfortunately, some recent versions of Python packages that AutoML relies on have introduced incompatibilities, and can cause failures when training models (TRAIN MODEL statement). If you see an error mentioning "TypeError" and the keyword argument "fit_params" or "sklearn_tags", read on for a quick fix.
Inevitably, you will eventually need to move your code up from one version of IRIS or Cache to a more recent version of IRIS. There are a few good steps you can take to set yourself up for success in that process.
Starting with InterSystems IRIS 2025.1, the way dependent cubes are handled in cube builds and cube synchronizes was changed.
This change may require modifying custom build/synchronize methods. If you are using the Cube Manager, these changes are already considered and handled, which means no action is needed.
Prior to this change, cubes were required to be built and synchronized in the proper order and account for any cube relationships/dependencies. With this change, dependent cubes are automatically updated as needed when using the %BuildCube or %SynchronizeCube APIs.
Year