GP-NAN

FloatingPointError                        Traceback (most recent call last)
<ipython-input-5-37d1187a7349> in <module>()
     11     likelihood = pm.ZeroInflatedPoisson("likelihood", psi=p, theta=tt.exp(f), observed=y)
     12
---> 13     tr = pm.sample(2000)
     14

~/anaconda3/lib/python3.6/site-packages/pymc3/sampling.py in sample(draws, step, init, n_init, start, trace, chain, njobs, tune, nuts_kwargs, step_kwargs, progressbar, model, random_seed, live_plot, discard_tuned_samples, live_plot_kwargs, **kwargs)
    241         start_, step = init_nuts(init=init, njobs=njobs, n_init=n_init,
    242                                  model=model, random_seed=random_seed,
--> 243                                  progressbar=progressbar, **args)
    244         if start is None:
    245             start = start_

~/anaconda3/lib/python3.6/site-packages/pymc3/sampling.py in init_nuts(init, njobs, n_init, model, random_seed, progressbar, **kwargs)
    698             callbacks=cb,
    699             progressbar=progressbar,
--> 700             obj_optimizer=pm.adagrad_window
    701         )  # type: pm.MeanField
    702         start = approx.sample(draws=njobs)

~/anaconda3/lib/python3.6/site-packages/pymc3/variational/inference.py in fit(n, local_rv, method, model, random_seed, start, inf_kwargs, **kwargs)
    883                         'or Inference instance' %
    884                         set(_select.keys()))
--> 885     return inference.fit(n, **kwargs)

~/anaconda3/lib/python3.6/site-packages/pymc3/variational/inference.py in fit(self, n, score, callbacks, progressbar, **kwargs)
    129         progress = tqdm.trange(n, disable=not progressbar)
    130         if score:
--> 131             self._iterate_with_loss(n, step_func, progress, callbacks)
    132         else:
    133             self._iterate_without_loss(n, step_func, progress, callbacks)

~/anaconda3/lib/python3.6/site-packages/pymc3/variational/inference.py in _iterate_with_loss(self, n, step_func, progress, callbacks)
    170                     scores = scores[:i]
    171                     self.hist = np.concatenate([self.hist, scores])
--> 172                     raise FloatingPointError('NaN occurred in optimization.')
    173                 scores[i] = e
    174                 if i % 10 == 0:

FloatingPointError: NaN occurred in optimization.