MetropolisWithinGibbs (bristlecone)

MetropolisWithinGibbs Module

Namespace: Bristlecone.Optimisation
Assembly: Bristlecone.dll
Parent Module: MonteCarlo

Functions and values

Function or value	Description
`core isAdaptive writeOut random domain f results batchLength batchNumber theta sigmas` Full Usage: `core isAdaptive writeOut random domain f results batchLength batchNumber theta sigmas` Parameters: isAdaptive : `bool` writeOut : `LogEvent -> unit` random : `Random` domain : `('a * 'b * Constraint)[]` f : `float[] -> float` results : `(float * float[]) list` batchLength : `int` batchNumber : `int` theta : `float[]` sigmas : `float[]` Returns: `int * (float * float[]) list * float[]`	Adaptive-metropolis-within-Gibbs algorithm, which can work in both adaptive and fixed modes isAdaptive : `bool` writeOut : `LogEvent -> unit` random : `Random` domain : `('a * 'b * Constraint)[]` f : `float[] -> float` results : `(float * float[]) list` batchLength : `int` batchNumber : `int` theta : `float[]` sigmas : `float[]` Returns: `int * (float * float[]) list * float[]`
`propose theta j lsj domain random` Full Usage: `propose theta j lsj domain random` Parameters: theta : `float[]` j : `int` lsj : `float` domain : `('a * 'b * Constraint)[]` random : `Random` Returns: `float[]`	Propose a jump, while leaving all but one parameter value fixed theta : `float[]` j : `int` lsj : `float` domain : `('a * 'b * Constraint)[]` random : `Random` Returns: `float[]`
`tune sigma acceptanceRate batchNumber` Full Usage: `tune sigma acceptanceRate batchNumber` Parameters: sigma : `float` acceptanceRate : `float` batchNumber : `int` Returns: `float`	Tune variance of a parameter based on its acceptance rate. The magnitude of tuning reduces as more batches have been run. sigma : `float` acceptanceRate : `float` batchNumber : `int` Returns: `float`