Reduced regression features

[GernotMaier]

No description provided.

[Copilot]

Pull request overview

This pull request introduces a configurable “reduced” feature set for stereo regression training, allowing training to use a stable, explicitly enumerated subset of flattened columns (while keeping the default “extended” behavior unchanged).

Changes:

• Add regression_feature_columns(..., profile=...) with extended vs reduced selection and strict required-column validation for the reduced profile.
• Wire --feature_profile {extended,reduced} into stereo training configuration and update regression training to use the selected profile.
• Add tests and README documentation for the reduced regression feature profile; adjust default stereo hyperparameters.

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
tests/test_regression_contracts.py	Adds a contract test ensuring the reduced profile selects the exact requested regression columns.
tests/test_features.py	Adds tests pinning the reduced regression column list and verifying missing-column errors.
tests/test_config.py	Verifies --feature_profile reduced is parsed for stereo training.
src/eventdisplay_ml/models.py	Switches regression training feature selection to regression_feature_columns based on feature_profile.
src/eventdisplay_ml/features.py	Introduces regression_feature_columns with extended/reduced profiles and required-column checks.
src/eventdisplay_ml/configs/default_hyperparameters_stereo.json	Tweaks default stereo regression hyperparameters (max_depth, min_child_weight).
src/eventdisplay_ml/config.py	Adds the stereo-only CLI flag --feature_profile and logs the chosen profile.
README.md	Documents the reduced regression feature profile and provides an example command.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.