Quick start

mltpy fits a monotone transformation h(y|x) such that h(Y|X) follows a known reference distribution (standard normal by default). Once fitted, the model answers distributional queries — CDF, density, quantile, hazard, survivor, cumulative hazard, odds, the transformation itself, and the numerically-stable log-scale variants of each — through a single predict call.

The snippet below fits the unconditional model to synthetic log-normal data and prints the estimated median.

import numpy as np
import mltpy

rng = np.random.default_rng(0)
y = rng.lognormal(mean=3.5, sigma=0.8, size=200).clip(0, 200)

model = mltpy.MLT(order=6, support=(0, 200))
model.fit(y)

grid   = np.linspace(10, 180, 100)
cdf    = model.predict(grid, what="distribution")
median = model.predict(np.array([0.5]), what="quantile")[0]
print(f"Estimated median: {median:.1f}")

Next, work through the vignettes for three canonical use cases from the Hothorn papers: flexible Box-Cox regression, survival analysis under right-censoring, and conditional regression with covariates.

predict accepts fourteen what= options: "trafo", "distribution", "logdistribution", "survivor", "logsurvivor", "density", "logdensity", "hazard", "loghazard", "cumhazard", "logcumhazard", "odds", "logodds", and "quantile". For the full table of formulas and the CensoredData container, see the API reference.