Defining non-standard likelihood

Questions
8

It looks like you may be mixing the Potential and DensityDist approaches outlined in Junpeng’s original post. Changing this line:

like = pm.DensityDist('like',
                      Likelihood,
                      observed=dict(phi = phi,
                                    d = 1,
                                    value = data)
                     )

to this:

like = pm.Potential('like', Likelihood(phi, 1, data))

at least gets it sampling. Is that what you intended?

1 Like