nirs4all.controllers.models.stacking.crossbranch module

Cross-Branch Stacking Support (Phase 7).

This module provides support for stacking across multiple branches, allowing meta-models to use predictions from models in different preprocessing branches.

Key Features: 1. BranchScope.ALL_BRANCHES support - stack across all branches 2. Feature alignment validation - ensure samples match across branches 3. Branch compatibility detection - identify compatible vs incompatible branches 4. Cross-branch prediction aggregation - combine predictions from multiple branches

Compatibility Matrix: | Branch Type | Cross-Branch Stacking | Notes | |---------------------|———————-|----------------------------------| | Preprocessing | ✅ Supported | Same samples, different features | | Generator (_or_) | ✅ Supported | Same samples, model variants | | Outlier Excluder | ✅ Supported | Same samples, different training | | Sample Partitioner | ❌ Not Supported | Different samples per partition |

Example

>>> validator = CrossBranchValidator(prediction_store)
>>> result = validator.validate_cross_branch_stacking(
...     source_candidates=candidates,
...     context=context
... )
>>> if result.is_compatible:
...     aligned_features = validator.align_branch_features(...)
nirs4all.controllers.models.stacking.crossbranch.validate_all_branches_scope(prediction_store: Predictions, source_candidates: list[ModelCandidate], context: ExecutionContext) → CrossBranchValidationResult[source]

Convenience function for validating BranchScope.ALL_BRANCHES.

Parameters:

  • prediction_store – Predictions storage.

  • source_candidates – List of candidate source models.

  • context – Execution context.

Returns:

CrossBranchValidationResult with compatibility info.