BUG integrator keyword

[lorentzenchr]

Description of your problem

Since version 3.7, 'integrator' is not anymore accepted as a kwarg to sample. This might have to do with the deprication of nuts_kwargs and step_kwargs in v3.7.

Please provide a minimal, self-contained, and reproducible example.

with pm.Model() as model:
    mu = pm.Normal('mu', mu=0, sigma=1)
    obs = pm.Normal('obs', mu=mu, sigma=1, observed=np.random.randn(100))
    kwargs = {'integrator':'two-stage'}
    trace = pm.sample(1000, tune=500, **kwargs)

Please provide the full traceback.

Auto-assigning NUTS sampler...
Initializing NUTS using jitter+adapt_diag...
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-3-48064cbbd5a7> in <module>
      3     obs = pm.Normal('obs', mu=mu, sigma=1, observed=np.random.randn(100))
      4     kwargs = {'integrator':'two-stage'}
----> 5     trace = pm.sample(1000, tune=500, **kwargs) 

/python3.6/site-packages/pymc3/sampling.py in sample(draws, step, init, n_init, start, trace, chain_idx, chains, cores, tune, progressbar, model, random_seed, discard_tuned_samples, compute_convergence_checks, **kwargs)
    394                 start_, step = init_nuts(init=init, chains=chains, n_init=n_init,
    395                                          model=model, random_seed=random_seed,
--> 396                                          progressbar=progressbar, **kwargs)
    397                 if start is None:
    398                     start = start_

/python3.6/site-packages/pymc3/sampling.py in init_nuts(init, chains, n_init, model, random_seed, progressbar, **kwargs)
   1513             'Unknown initializer: {}.'.format(init))
   1514
-> 1515     step = pm.NUTS(potential=potential, model=model, **kwargs)
   1516
   1517     return start, step

/python3.6/site-packages/pymc3/step_methods/hmc/nuts.py in __init__(self, vars, max_treedepth, early_max_treedepth, **kwargs)
    150         `pm.sample` to the desired number of tuning steps.
    151         
--> 152         super().__init__(vars, **kwargs)
    153
    154         self.max_treedepth = max_treedepth
 
/python3.6/site-packages/pymc3/step_methods/hmc/base_hmc.py in __init__(self, vars, scaling, step_scale, is_cov, model, blocked, potential, dtype, Emax, target_accept, gamma, k, t0, adapt_step_size, step_rand, **theano_kwargs)
     70         vars = inputvars(vars)
     71
---> 72         super().__init__(vars, blocked=blocked, model=model, dtype=dtype, **theano_kwargs)
     73
     74         self.adapt_step_size = adapt_step_size
 
/python3.6/site-packages/pymc3/step_methods/arraystep.py in __init__(self, vars, model, blocked, dtype, **theano_kwargs)
    226
    227         func = model.logp_dlogp_function(
--> 228             vars, dtype=dtype, **theano_kwargs)
    229
    230         # handle edge case discovered in #2948

/python3.6/site-packages/pymc3/model.py in logp_dlogp_function(self, grad_vars, **kwargs)
    721         varnames = [var.name for var in grad_vars]
    722         extra_vars = [var for var in self.free_RVs if var.name not in varnames]
--> 723         return ValueGradFunction(self.logpt, grad_vars, extra_vars, **kwargs)
    724
    725     @property
 
/python3.6/site-packages/pymc3/model.py in __init__(self, cost, grad_vars, extra_vars, dtype, casting, **kwargs)
    460
    461         self._theano_function = theano.function(
--> 462             inputs, [self._cost_joined, grad], givens=givens, **kwargs)
    463
    464     def set_extra_values(self, extra_vars):
 
TypeError: function() got an unexpected keyword argument 'integrator'

Versions and main components

• PyMC3 Version: 3.7
• Theano Version: 1.0.4
• Python Version: 3.6.8
• Operating system: RHEL 7.6
• How did you install PyMC3: pip

[fonnesbeck]

Yes, this also fails when passing a nuts_kwargs that is deprecated, but should still work.

[ColCarroll]

@lorentzenchr I think I just pushed a fix -- is it right that it was just the leftover documentation, or was there something else that could help others avoid this problem?

[lorentzenchr]

@ColCarroll Thanks for addressing this issue.
Some time ago, I had a computational challenging analysis and I think I had found some hint on using integrator at https://discourse.pymc.io/, but I don't find that post anymore.
On the other hand, I can't find any note on dropping integrator in https://github.com/pymc-devs/pymc3/blob/master/RELEASE-NOTES.md neither 😏