Add warning if numbers of variables in vars does not equal number of model variables

Author: carsten-j

pymc_experimental/inference/laplace.py

@@ -12,6 +12,7 @@
 #   See the License for the specific language governing permissions and
 #   limitations under the License.
 
+import warnings
 from collections.abc import Sequence
 from typing import Optional
 
@@ -98,6 +99,13 @@ def laplace(
     rng = np.random.default_rng(seed=random_seed)
 
     transformed_m = pm.modelcontext(model)
+
+    if len(vars) != len(transformed_m.free_RVs):
+        warnings.warn(
+            "Number of variables in vars does not eqaul the number of variables in the model.",
+            UserWarning,
+        )
+
     map = pm.find_MAP(vars=vars, progressbar=progressbar, model=transformed_m)
 
     # See https://www.pymc.io/projects/docs/en/stable/api/model/generated/pymc.model.transform.conditioning.remove_value_transforms.html

pymc_experimental/tests/test_laplace.py

@@ -87,6 +87,7 @@ def test_laplace_only_fit():
     idata = pmx.fit(
         method="laplace",
         vars=vars,
+        draws=None,
         model=m,
         random_seed=173300,
     )
@@ -99,3 +100,41 @@ def test_laplace_only_fit():
 
     assert np.allclose(idata.fit["mean_vector"].values, bda_map)
     assert np.allclose(idata.fit["covariance_matrix"].values, bda_cov, atol=1e-4)
+
+
+@pytest.mark.filterwarnings(
+    "ignore:Model.model property is deprecated. Just use Model.:FutureWarning",
+    "ignore:hessian will stop negating the output in a future version of PyMC.\n"
+    + "To suppress this warning set `negate_output=False`:FutureWarning",
+)
+def test_laplace_subset_of_rv(recwarn):
+
+    # Example originates from Bayesian Data Analyses, 3rd Edition
+    # By Andrew Gelman, John Carlin, Hal Stern, David Dunson,
+    # Aki Vehtari, and Donald Rubin.
+    # See section. 4.1
+
+    y = np.array([2642, 3503, 4358], dtype=np.float64)
+    n = y.size
+
+    with pm.Model() as m:
+        logsigma = pm.Uniform("logsigma", 1, 100)
+        mu = pm.Uniform("mu", -10000, 10000)
+        yobs = pm.Normal("y", mu=mu, sigma=pm.math.exp(logsigma), observed=y)
+        vars = [mu]
+
+    idata = pmx.fit(
+        method="laplace",
+        vars=vars,
+        draws=None,
+        model=m,
+        random_seed=173300,
+    )
+
+    assert len(recwarn) == 4
+    w = recwarn.pop(UserWarning)
+    assert issubclass(w.category, UserWarning)
+    assert (
+        str(w.message)
+        == "Number of variables in vars does not eqaul the number of variables in the model."
+    )