Classification Nodes

Classification nodes assign spectra or feature rows to classes.

Nodes

Node	Use When	Inputs	Outputs	Key Configuration
Train KNN Classifier (`classification.knn`)	You need a distance-based classifier after appropriate scaling or dimensionality reduction.	`X: Array2D`; `y: Categorical?`	`model`; `predictions`; `probabilities`; `metrics`; `distances`; `neighbor_indices`; `train_accuracy`; `cv_accuracy`; `plots`	`n_neighbors`; `weights`; `metric`; `scale`; `cv_folds`.
Train PLS-DA Classifier (`classification.plsda`)	You want a latent-variable classifier for spectral class separation.	`X: Array2D`; `y: Categorical?`	`model`; `predictions`; `probabilities`; `X_scores`; `X_loadings`; `metrics`	`n_components`; `scale`; `cv_folds`; `calibrate_probabilities`; `probability_method`. Requires SpectroChemPy.
Train SIMCA Classifier (`classification.simca`)	You want class-specific PCA models that can reject samples outside known classes.	`X: Array2D`; `y: Categorical?`	`model`; `predictions`; `class_assignment`; `distances`; `class_distance_matrix`; `confusion_matrix`; `plots`; metrics	`n_components`; `confidence_level`; `critical_limits_method`; `cv_folds`. Requires SpectroChemPy.
Apply Classification Model (`classification.predict`)	Apply a fitted classifier to new spectra.	`X_new: SpectralDataset`; `model: ClassificationModel`	`y_pred: Categorical`; `y_prob: Array2D`	no parameters. Keep preprocessing identical to training.

Key Outputs

Expect predictions, class probabilities or distances where appropriate, confusion matrices, and split-aware metrics when validation is part of the workflow.

PLS-DA probability outputs are derived from model scores/regression outputs unless explicitly calibrated; treat them as relative confidence unless calibrated probability estimation is part of the workflow.

SIMCA Note

SIMCA is not just nearest-class classification. It can reject samples that do not fit any class model, which is important for QC workflows.