At the end the logp is summed together, so for multiple observed variable each with its own penalty, you can add multiple pm.Potential.
1 Like
At the end the logp is summed together, so for multiple observed variable each with its own penalty, you can add multiple pm.Potential.