The workflow starts by reading input data from a CSV file and cleaning it by dropping rows with null values in selected columns to ensure data quality. A pre-trained Scikit-learn Gradient Boosting classification model is then loaded using the Sklearn Model Load node. The cleaned dataset is passed to the Sklearn Predict node, where predictions are generated using the loaded model. Model performance is evaluated using the Sklearn Classification Evaluator, which computes classification metrics such as accuracy and other relevant measures. In parallel, prediction results and evaluation outputs are captured by the ML Data Metrics node for tracking, monitoring, and downstream analysis.