idtlab

Use when the user asks "is my data AI ready", "is my dataset ready", "what is the quality of my data", whether data is good enough to train or publish, to check a dataset for bias, fairness, privacy, PII risk, HIPAA compliance, protected health information, PHI, class imbalance, duplicates, outliers, completeness, feature relevance, k-anonymity, feature correlation, collinearity, redundant features, skewness, kurtosis, distribution shape, or mentions AIDRIN. Supports CSV, Excel (.xls/.xlsb/.xlsx/.xlsm), JSON, NumPy (.npz), HDF5 (.h5), and Parquet files.