CODEWITHSUNDEEP
pythonbayesianpymcmcmc
aloha
aloha

Reputation: 4774

Number of parameters in MCMC

I want to sample from my posterior distribution using the pymc package.

I am wondering if there is a limit on the number of dimensions such algorithm can handle. My log likelihood is the sum of 3 Gaussians and 1 mixture of Gaussians. I have approx 750 parameters in my model. Can pymc handle such a big number of parameters?

Upvotes: 1

Views: 488

Answers (1)

inversion
inversion

Reputation: 1304

I recently ran (successfully) a model with 2,958 parameters. It was on a 8 Gb Windows machine. You should be fine with 750.

Upvotes: 2

Related Questions