Leakage-Safe Modeling and Auditing for Genomic and Clinical Data
Convert LeakSplits to an rsample resample set
Audit leakage per learner
Audit leakage and confounding
Render an HTML audit report
Calibration diagnostics for binomial predictions
Confounder sensitivity summaries
Circular block permutation indices
Ensure consistent categorical levels for guarded preprocessing
Fit leakage-safe preprocessing pipeline
Restricted permutation label factory
Quantile break cache for permutation stratification
Stationary bootstrap indices
Fit and evaluate with leakage guards over predefined splits
Leakage-safe data imputation via guarded preprocessing
S4 Classes for bioLeak Pipeline
Create leakage-resistant splits
Plot calibration curve for binomial predictions
Plot confounder sensitivity
Plot fold balance of class counts per fold
Plot overlap diagnostics between train/test groups
Plot permutation distribution for a LeakAudit object
Plot ACF of test predictions for time-series leakage checks
Apply a fitted GuardFit transformer to new data
Display summary for LeakSplits objects
Simulate leakage scenarios and audit results
Summarize a leakage audit
Summarize a LeakFit object
Summarize a nested tuning result
Leakage-aware nested tuning with tidymodels
Prevents and detects information leakage in biomedical machine learning. Provides leakage-resistant split policies (subject-grouped, batch-blocked, study leave-out, time-ordered), guarded preprocessing (train-only imputation, normalization, filtering, feature selection), cross-validated fitting with common learners, permutation-gap auditing, batch and fold association tests, and duplicate detection.