I would like to construct a reversible jump MCMC to account for variation in dimensions. An example use case would be a clustering algorithm where the number of clusters is unknown. Is this feasible in PyMC3? So far I have only found references to this in a closed github issue and a very old questio…

Is reversible jump (RJMCMC) feasible?

prcummins December 13, 2018, 9:07am 4

Hello,
I am keen to find an implementation of reversible jump in Python. Can anyone tell me, has any progress on this been made since since year-old thread?
Thanks

1 Like

Topic		Replies	Views
Can PyMC be used for reversible-jump MCMC?	1	101	July 15, 2024
Mixture model with variable number of component distributions Questions	2	1589	November 29, 2018
Model yeilding no samples Questions	1	405	July 31, 2018
Indexing using a free random variable Questions	26	6452	October 16, 2019
Adaptive MCMC in PyMC? Development	9	2476	May 15, 2018

Is reversible jump (RJMCMC) feasible?

Related topics