nirs4all.controllers.models.residual_model module

Controller for the residual model-combination operator (ResidualModel / {"residual": {...}}).

Fits a base model (leakage-safe OOF), trains a learner on the OOF residuals, and stores the combined prediction base + lambda*gate*learner. Both sub-models are routed through their own framework controllers, so the learner may be any nirs4all model (sklearn, PyTorch nicon, TabPFN, …).

The two sub-models are fit by delegating to the standard model controllers via the router; the residual target is delivered to the learner through a numeric-rooted processed-target series (add_processed_targets + context.with_y), so no learner-side change is needed.

class nirs4all.controllers.models.residual_model.ResidualModelController

Bases: OperatorController

Fit base + residual-learner combination. Priority 5 (above framework model controllers).

execute(step_info, dataset, context, runtime_context, source: int = -1, mode: str = 'train', loaded_binaries=None, prediction_store=None)

Run the operator with the given parameters and context.

Parameters:

• step_info – Parsed step containing operator, keyword, and metadata

• dataset – Dataset to operate on

• context – Pipeline execution context

• runtime_context – Runtime infrastructure context

• source – Data source index

• mode – Execution mode (“train” or “predict”)

• loaded_binaries – Pre-loaded binary objects for prediction mode

• prediction_store – External prediction store for model predictions

Returns:

Tuple of (updated_context, StepOutput)

classmethod matches(step: Any, operator: Any, keyword: str) → bool

Check if the operator matches the step and keyword.

priority: int = 5

classmethod supports_prediction_mode() → bool

Check if the controller should execute during prediction mode.

Returns:

True if the controller should execute in prediction mode, False if it should be skipped (e.g., chart controllers)

classmethod use_multi_source() → bool

Check if the operator supports multi-source datasets.