These files have categorical values numerically encoded and missing values imputed, which makes them usable for any scikit-learn algo. This preprocessing can take considerable time and is excluded from time measurements currently, but should be included at some point in the future.
These files have categorical values numerically encoded and missing values imputed, which makes them usable for any scikit-learn algo. This preprocessing can take considerable time and is excluded from time measurements currently, but should be included at some point in the future.