Model-agnostic prediction intervals in Python and R: does nnetsauce’s `QuantileRegressor` hold up?

Point predictions tell you what a model thinks will happen. They don’t tell you how much to trust that number. nnetsauce’s QuantileRegressor takes any sklearn-compatible regressor and turns it into a full quantile machine by optimizing an offset around its point predictions to minimize the pinball (quantile) loss.