ProDeF - Probability Density Functions

A simple Rust crate for handling probability distributions.

Core Traits

All probability density functions (PDFs) implement the [Density], providing common operations like:

• Evaluation: Compute probability density (non-normalized) at a given point
• Sampling: Generate random samples from the distribution
• Domain queries: Checking valid input ranges

The Domain trait represents the valid input space for a PDF. A domain may be bounded, unbounded, or have a special structure.

Available domain types:

• MDomain - Bounded multivariate domains (hypercubes)
• UDomain - Unbounded multivariate domains
• SDomain - Special univariate domains (for the use of uni dimensional PDFs)

Distribution Types

Multivariate Distributions

• MultivariateDensity Combines multiple independent univariate distributions into a multivariate density. Use when dimensions are statistically independent.

• MultiNormalDensity A full multivariate normal (Gaussian) distribution with arbitrary covariance. Use when modeling correlated multi-dimensional data.

• ParticleDensity A non-parametric density represented by weighted particles/samples. Use for complex distributions that can't be expressed analytically or for particle filter applications.

Univariate Distributions

For one-dimensional cases, use these directly:

• ConstantDensity - A degenerate distribution at a fixed value
• CosineDensity - Cosine distribution
• LogUniformDensity - Uniform in log-space (for positive values)
• NormalDensity - Nnormal distribution
• UniformDensity - Uniform distribution over an interval