ML Community Day is November 9! Join us for updates from TensorFlow, JAX, and more Learn more

tfp.substrates.jax.math.log_bessel_ive

Computes log(tfp.math.bessel_ive(v, z)).

This function is a more numerically stable version of log(tfp.math.bessel_ive(v, z)), along with more numerically stable gradients.

v Floating-point Tensor broadcastable with z for which log(Ive(v, z)) should be computed. v is expected to be non-negative.
z Floating-point Tensor broadcastable with v for which log(Ive(v, z)) should be computed. If z is negative, v is expected to be an integer.
name A name for the operation (optional). Default value: None (i.e., 'log_bessel_ive').

log_bessel_ive Log of Exponentially modified Bessel Function of the first kind.