fix default prior variable names (#3591)

* fix default prior variable names

* update docs and add test on pm.Data not being in prior

* Add model.potentials to prior_vars

pymc3/sampling.py

@@ -1292,7 +1292,7 @@ def sample_prior_predictive(samples=500,
         samples.  *DEPRECATED* - Use ``var_names`` argument instead.
     var_names : Iterable[str]
         A list of names of variables for which to compute the posterior predictive
-        samples. Defaults to ``model.named_vars``.
+        samples. Defaults to both observed and unobserved RVs.
     random_seed : int
         Seed for the random number generator.
 
@@ -1305,8 +1305,13 @@ def sample_prior_predictive(samples=500,
     model = modelcontext(model)
 
     if vars is None and var_names is None:
-        vars = set(model.named_vars.keys())
-        vars_ = model.named_vars
+        prior_pred_vars = model.observed_RVs
+        prior_vars = (
+            get_default_varnames(model.unobserved_RVs, include_transformed=True) +
+            model.potentials
+        )
+        vars_ = [var.name for var in prior_vars + prior_pred_vars]
+        vars = set(vars_)
     elif vars is None:
         vars = var_names
         vars_ = vars

pymc3/tests/test_sampling.py

@@ -505,12 +505,14 @@ def test_ignores_observed(self):
         observed = np.random.normal(10, 1, size=200)
         with pm.Model():
             # Use a prior that's way off to show we're ignoring the observed variables
+            observed_data = pm.Data("observed_data", observed)
             mu = pm.Normal("mu", mu=-100, sigma=1)
             positive_mu = pm.Deterministic("positive_mu", np.abs(mu))
             z = -1 - positive_mu
-            pm.Normal("x_obs", mu=z, sigma=1, observed=observed)
+            pm.Normal("x_obs", mu=z, sigma=1, observed=observed_data)
             prior = pm.sample_prior_predictive()
 
+        assert "observed_data" not in prior
         assert (prior["mu"] < 90).all()
         assert (prior["positive_mu"] > 90).all()
         assert (prior["x_obs"] < 90).all()