No user-submitted related servers found.
Scored across 9 tools
Each tool maps to a distinct pipeline stage: training, per-model prediction, combined prediction, metrics, model selection, SHAP explanation, sample input, and audit export. The only mild overlap is predict_both versus predict_logistic_regression/predict_knn, but the descriptions make the distinction clear (single model vs. both models with agreement check).
Most names follow a clear verb_noun pattern (train_models, get_sample_input, select_best_model, export_parquet_audit, predict_logistic_regression). A couple deviate into noun-first or nested phrasing (confusion_matrix_metrics, shap_explain_selected), which is a minor inconsistency but still readable and predictable.
Nine tools is well-scoped for a fixed ML workflow covering train, evaluate, select, explain, and predict. Every tool corresponds to a concrete step; none appear redundant or gratuitous.
The surface covers the full lifecycle: training and artifact export, sample input retrieval, single and combined prediction, evaluation metrics, model selection, SHAP explanation, and audit export. Minor gaps exist, such as no tool to list or inspect saved models/artifacts, but core workflows are fully supported.