Auto-diff when using NUTS sampler with blackbox likelihoods

aseyboldt · September 11, 2020, 8:18am

All of pymc3 is using automatic differentiation.
We are using theano, so you can use arbitrary theano code in your model. If you wanted to use autograd, you would have to link theano and autograd together. Quite possible, but some work and knowledge of theano and autograd would be required for that.

Topic		Replies	Views
Is there a way to call NUTS directly to sample a customized likelihood together with its gradient? Questions	2	779	March 13, 2020
Using pymc3 NUTS sampler for an "external model" having its derivatives with respect to its parameters Questions	4	2390	March 30, 2018
Using pm.DensityDist and customized likelihood with a black-box function Questions	10	6853	August 29, 2018
NUTS - FromFunctionOp has no attribute 'grad' Questions theano	2	1005	August 3, 2018
NUTS can sample even failing to differentiate? Questions	6	813	September 23, 2018

Auto-diff when using NUTS sampler with blackbox likelihoods

Related topics