Model → Predict

Model Predict uses a trained (or loaded) machine learning model to make predictions on new, unseen data. It takes input features and returns the model output — class labels for classifiers, cluster assignments for clustering, or numerical values for regressors.

This is the final step in the ML pipeline where the model is applied to real data.

The input features for prediction must have the same structure (same number and order of features) as the data used during training. Use the same preprocessing steps (Feature Extractor, Feature Scaler, Encoder) that were applied to the training data.