leakr0.1.0 package

Enhanced column name cleaning with better robustness

Enhanced report compilation with numeric severity scores

Enhanced date detection handling multiple formats and data types

Detect file format from extension and content

Registry-based Detector System

Determine risk level and CSS class from severity counts.

Initialise built-in detectors

Helper function to return an empty snapshot info dataframe

Export data with consistent messaging

Format detector names for display.

Generate diagnostic plots for a leakr_report

Generate evidence section with format-specific handling and DRY logic.

Report generator

Generate detailed issues section with output formatting and truncation...

Format recommendations for output.

Generate actionable recommendations based on report findings.

Get detector information

Null-coalescing operator for clean default value handling

Import CSV files with robust parsing

Import Excel files with enhanced sheet support

Import JSON files with better structure handling

Import Parquet files

Import RDS files with validation

Import TSV files with robust parsing

Audit dataset for data leakage

Create data snapshots with improved metadata handling

Export data in various formats

Convert caret training objects to standard format

Convert mlr3 Task objects to standard format

Convert tidymodels workflow to standard format

Import data from various sources for leakage analysis

List available snapshots with enhanced information

Load data snapshot with enhanced validation

Plot leakage detection results

Fast import with default preprocessing

Enhanced summarise with better formatting

leakr: Data Leakage Detection for Machine Learning in R

List Registered Detectors

Create a new temporal detector

Create a new train-test detector

Plot a detector_result object

Plot a udld_report object

Enhanced data preparation with robust preprocessing

Enhanced preprocessing with better performance and robustness

Print method for leakr_report

Register a new detector

Run a detector on data

Run multiple detectors on audit data

Stratified sampling helper

Robust data validation and preprocessing

Enhanced data validation with better error messages