Tests Catalog

Advanced feature engineering: temporal and nested-CV leakage, train–serve skew, entity embeddings, automated interactions, monotonic constraints, drift monitoring, null importance, and causal vs predictive features. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.

Intermediate feature engineering: encoding high-cardinality categories, binning, interactions, point-in-time aggregates, cross-validated encoders, multicollinearity, and feature selection trade-offs. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.

Foundations of feature engineering: raw inputs vs model features, numeric and categorical handling, basic encoding, scaling, missing values, and avoiding obvious leakage. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.

Advanced ML evaluation: nested CV pitfalls, calibration error metrics, fairness-aware reporting, bootstrap and paired tests, off-policy and slice analysis, and probabilistic forecasting scores. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.

Intermediate ML evaluation: stratified splits, precision–recall tradeoffs, macro vs micro metrics, calibration basics, learning curves, threshold tuning, and regression diagnostics. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.

Foundations of ML model evaluation: train/validation/test splits, confusion matrix, precision and recall, overfitting, and choosing sensible baselines. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.

Advanced supervised ML: calibration, class hierarchies, multi-label vs multiclass, sample weights, label smoothing, distillation, ordinal and cost-sensitive learning, kernel methods, and production pitfalls. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.

Intermediate supervised ML: regularization, class imbalance, decision boundaries, logistic and softmax outputs, feature scaling, encoding categoricals, tree-based learners, and training pitfalls. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.

Foundations of supervised machine learning: labeled data, classification vs regression, training and prediction, common losses, overfitting, and basic model families. Mix of single-answer and multiple-answer items; multi-select questions ask you to select every correct statement.