Calculating marginal likelihood of GP Model

Questions
6

So silly to miss that on my part. Sorry for asking so many questions!

Just to confirm the output of logp by MAP is logp of the model. So the optimizer is maximizing log the probablity of the model.

logp = -220.68, ||grad|| = 149.66: 100%|██████████| 11/11 [00:00<00:00, 97.39it/s]

mp

{'noise_obs_log__': array(-0.7523434),
 'lengthscale_log__': array(-0.14632332),
 'noise_obs': array(0.47126091),
 'lengthscale': array(0.86387835)}

So, how do we get the log MLE ?

yobs.logp(mp)

Thanks a lot!