bf.dist.zero_inflated_distribution: Generic Zero Inflated distribution.

In BayesForge: Bayesian Inference using 'numpyro' and 'XLA'

Generic Zero Inflated distribution.

Description

The Zero-Inflated Poisson distribution is a discrete count-distribution designed for data with *more zeros* than would be expected under a standard Poisson. Essentially, it assumes two underlying processes: * With probability

\pi

you are in a "structural zero" state (i.e., you automatically get a zero count). * With probability

1 - \pi

you draw from a standard Poisson distribution with parameter

\lambda

.

This results in a mixture distribution that places more mass at zero than a Poisson alone would. It's widely used in, for instance, ecology (species counts with many zeros), insurance/claims problems, and any count-data setting with excess zeros.

Usage

bf.dist.zero_inflated_distribution(
  base_dist,
  gate = py_none(),
  gate_logits = py_none(),
  validate_args = py_none(),
  name = "x",
  obs = py_none(),
  mask = py_none(),
  sample = FALSE,
  seed = py_none(),
  shape = c(),
  event = 0,
  create_obj = FALSE,
  to_jax = TRUE
)

Arguments

base_dist	Distribution: The base distribution to be zero-inflated (e.g., Poisson, NegativeBinomial).
gate	numeric(1): Probability of extra zeros (between 0 and 1).
gate_logits	numeric(1): Log-odds of extra zeros.
validate_args	Logical: Whether to validate parameter values. Defaults to 'reticulate::py_none()'.
name	A character string representing the name of the random variable within a model. This is used to uniquely identify the variable. Defaults to 'x'.
obs	A numeric vector or array of observed values. If provided, the random variable is conditioned on these values. If 'NULL', the variable is treated as a latent (unobserved) variable. Defaults to 'NULL'.
mask	logical(1): Optional boolean array to mask observations.
sample	A logical value that controls the function's behavior. If 'TRUE', the function will directly draw samples from the distribution. If 'FALSE', it will create a random variable within a model. Defaults to 'FALSE'.
seed	An integer used to set the random seed for reproducibility when 'sample = TRUE'. This argument has no effect when 'sample = FALSE', as randomness is handled by the model's inference engine. Defaults to 0.
shape	numeric(1): A multi-purpose argument for shaping. When 'sample=False' (model building), this is used with ‘.expand(shape)' to set the distribution’s batch shape. When 'sample=True' (direct sampling), this is used as 'sample_shape' to draw a raw JAX array of the given shape. Provide as a numeric vector (e.g., 'c(10)').
event	int(1): The number of batch dimensions to reinterpret as event dimensions (used in model building).
create_obj	Logical: If True, returns the raw BI distribution object instead of creating a sample site. This is essential for building complex distributions like 'MixtureSameFamily'.
to_jax	Boolean. Indicates whether to return a JAX array or not.

Value

- When sample=FALSE, a BI Zero Inflated distribution object (for model building).

- When sample=TRUE, a JAX array of samples drawn from the Zero Inflated distribution (for direct sampling).

- When create_obj=TRUE, the raw BI distribution object (for advanced use cases).

See Also

https://num.pyro.ai/en/stable/distributions.html#zeroinflateddistribution

Examples

library(BayesForge)
m=importBF(platform='cpu')
bf.dist.zero_inflated_distribution(
base_dist = bf.dist.poisson(5, create_obj = TRUE),
gate=0.3, sample = TRUE)

BayesForge documentation built on June 9, 2026, 1:09 a.m.