Probability distributions

The skpro.distributions module contains probability distributions which combine a pandas.DataFrame-like API with a scikit-base compatible object interface.

All distributions in skpro can be listed using the skpro.registry.all_objects utility, using object_types="distribution", optionally filtered by tags. Valid tags can be listed using skpro.registry.all_tags.

Base

BaseDistribution([index, columns])

Base probability distribution.

Parametric distributions

Continuous support - full reals

Laplace(mu, scale[, index, columns])	Laplace distribution.
Logistic(mu, scale[, index, columns])	Logistic distribution.
Normal(mu, sigma[, index, columns])	Normal distribution aka Gaussian distribution.
SkewNormal(mu, sigma[, alpha, index, columns])	Skew-Normal Probability Distribution.
TDistribution(mu, sigma[, df, index, columns])	Student's t-distribution.
TruncatedNormal(mu, sigma, l_trunc, r_trunc)	A truncated normal probability distribution.
Uniform(lower, upper[, index, columns])	Continuous uniform distribution.

Continuous support - non-negative reals

Alpha(a[, index, columns])	Alpha distribution.
Beta(alpha, beta[, index, columns])	Beta distribution.
BurrIII(c[, scale, index, columns])	Burr III probability distribution.
BurrXII(c, d[, scale, index, columns])	Burr XII probability distribution.
ChiSquared(dof[, index, columns])	Chi-Squared distribution.
Exponential(rate[, index, columns])	Exponential Distribution.
Erlang(rate[, k, index, columns])	Erlang Distribution.
FatigueLife(c[, scale, index, columns])	Fatigue-life (Birnbaum–Saunders) probability distribution.
FDist(dfn, dfd[, index, columns])	F (Fisher-Snedecor) probability distribution.
Fisk([alpha, beta, index, columns])	Fisk distribution, aka log-logistic distribution.
Gamma(alpha, beta[, index, columns])	Gamma Distribution.
GeneralizedPareto(c[, scale, loc, index, ...])	Generalized Pareto probability distribution.
GumbelL([mu, sigma, index, columns])	Gumbel Left probability distribution.
GumbelR([mu, sigma, index, columns])	Gumbel Right probability distribution.
Levy([mu, scale, index, columns])	Levy probability distribution.
LogGamma(c[, index, columns])	Log-Gamma Distribution.
HalfCauchy(beta[, index, columns])	Half-Cauchy distribution.
HalfLogistic(beta[, index, columns])	Half-Logistic distribution.
HalfNormal(sigma[, index, columns])	Half-Normal distribution.
InverseGamma(alpha, beta[, index, columns])	Inverse Gamma Distribution.
InverseGaussian(mu, scale[, index, columns])	Inverse Gaussian distribution, aka Wald distribution.
LogLaplace(scale[, index, columns])	Log-Laplace distribution.
Pareto(scale, alpha[, index, columns])	Pareto distribution.
Rayleigh([scale, index, columns])	Rayleigh distribution.
TruncatedPareto(b[, scale, lower, upper, ...])	Truncated Pareto probability distribution.
Weibull(scale, k[, index, columns])	Weibull distribution.

Integer support

Binomial(n, p[, index, columns])	Binomial distribution.
Geometric(p[, index, columns])	Geometric Distribution.
NegativeBinomial(mu, alpha[, index, columns])	Negative binomial distribution.
Poisson(mu[, index, columns])	Poisson distribution.
ZeroInflated(p, distribution[, index, columns])	A Zero-Inflated distribution, adding a Bernoulli gate to any distribution.
Skellam(mu1, mu2[, index, columns])	Skellam probability distribution.

Non-parametric and empirical distributions

Delta(c[, index, columns])	Delta distribution aka constant distribution aka certain distribution.
Empirical(spl[, weights, time_indep, ...])	Empirical distribution, or weighted sum of delta distributions.
Histogram(bins, bin_mass[, index, columns])	Histogram Probability Distribution.
QPD_Empirical(quantiles[, time_indep, ...])	Empirical quantile parametrized distribution.
QPD_Johnson(alpha, qv_low, qv_median, qv_high)	Johnson Quantile-Parameterized Distribution.
QPD_U(alpha, qv_low, qv_median, qv_high[, ...])	Johnson Quantile-Parameterized Distributions with unbounded mode.
QPD_S(alpha, qv_low, qv_median, qv_high, lower)	Johnson Quantile-Parameterized Distributions with semi-bounded mode.
QPD_B(alpha, qv_low, qv_median, qv_high, ...)	Johnson Quantile-Parameterized Distributions with bounded mode.

Composite distributions

Transformation composition

MeanScale(d[, mu, sigma, index, columns])	Composition for offset and scaling, and mean/scale family of distributions.
TransformedDistribution(distribution, transform)	Distribution transformed by an entry-wise function.

Truncated and inflated distributions

Hurdle(p, distribution[, index, columns])	A Hurdle distribution.
LeftTruncated(distribution, lower[, index, ...])	A left truncated distribution _not_ including the lower bound.
TruncatedDistribution(distribution, *[, ...])	A truncated distribution.
ZeroInflated(p, distribution[, index, columns])	A Zero-Inflated distribution, adding a Bernoulli gate to any distribution.

Mixture composition

KernelMixture(support[, h, kernel, weights, ...])	Kernel mixture distribution, a.k.a.
Mixture(distributions[, weights, ...])	Mixture of distributions.
NormalMixture(pi, mu, sigma[, index, columns])	Mixture of Normal distributions with per-sample mixing weights.

Sampling and multivariate composition

IID(distribution[, index, columns])

An iid sample of a given distribution.