Fix bug in compute_deterministics

pymc/sampling/deterministic.py

@@ -84,6 +84,7 @@ def compute_deterministics(
 
     if var_names is None:
         deterministics = model.deterministics
+        var_names = [det.name for det in deterministics]
     else:
         deterministics = [model[var_name] for var_name in var_names]
         if not set(deterministics).issubset(set(model.deterministics)):
@@ -101,7 +102,7 @@ def compute_deterministics(
     new_dataset = apply_function_over_dataset(
         fn,
         dataset[[rv.name for rv in model.free_RVs]],
-        output_var_names=[det.name for det in model.deterministics],
+        output_var_names=var_names,
         dims=dims,
         coords=coords,
         sample_dims=sample_dims,

tests/sampling/test_deterministic.py

@@ -59,6 +59,11 @@ def test_compute_deterministics():
     assert extended_with_mu["mu"].dims == ("chain", "draw", "group")
     assert_allclose(extended_with_mu["mu"], dataset["mu_raw"].cumsum("group"))
 
+    only_sigma = compute_deterministics(dataset, var_names=["sigma"], model=m, progressbar=False)
+    assert set(only_sigma.data_vars.variables) == {"sigma"}
+    assert only_sigma["sigma"].dims == ("chain", "draw")
+    assert_allclose(only_sigma["sigma"], np.exp(dataset["sigma_raw"]))
+
 
 def test_docstring_example():
     import pymc as pm