Scikit-learn StackingClassifier with missing values - scikit-learn

Question

When I call model.predict_proba(X) on my StackingClassifier model the run execution crashes because the library is calling a method assert_all_finite() to check whether my dataframe contains missing values.
Since the estimators I stacked are able to handle missing values, I don't see the reason why this should happen and I didn't find anything in the documentation that says that the StackingClassifier requires data without missing values.
It's a bit hard for me to come up with a short reproducibile snippet of code given that it comes from several layers of model abstraction, but I can print out the model effectively raising the error call.
p = model.predict_proba(X_loyal)
where model is:
StackingClassifier(estimators=[('ExtraTreesClassifier_117',
ExtraTreesClassifier(bootstrap=True,
class_weight={0: 1, 1: 5},
criterion='entropy',
max_depth=11,
max_features='log2',
max_samples=0.5946040593595099,
min_samples_leaf=2,
n_estimators=163,
random_state=117)),
('RandomForestClassifier_117',
RandomForestClassifier(class_weight={0: 1,
1: 5},
criterion='entropy',
max_depth=11,
max_features='log2',
max_samples=0.5946040593595099,
min_samples_leaf=2,
n_estimators=163,
random_state=117)),
('LGBMClassifier_117',
LGBMClassifier(class_weight={0: 1, 1: 1},
deterministic=True, max_depth=9,
n_estimators=183, num_leaves=3,
subsample=0.2986274713775564,
verbose=-1))])
Error
Traceback (most recent call last):
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/IPython/core/interactiveshell.py", line 3417, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)
File "<ipython-input-12-eafa75c49322>", line 1, in <module>
model.predict_proba(X_loyal)
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/utils/metaestimators.py", line 120, in <lambda>
out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/ensemble/_stacking.py", line 485, in predict_proba
return self.final_estimator_.predict_proba(self.transform(X))
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/ensemble/_stacking.py", line 522, in transform
return self._transform(X)
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/ensemble/_stacking.py", line 215, in _transform
predictions = [
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/ensemble/_stacking.py", line 216, in <listcomp>
getattr(est, meth)(X)
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/ensemble/_forest.py", line 674, in predict_proba
X = self._validate_X_predict(X)
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/ensemble/_forest.py", line 422, in _validate_X_predict
return self.estimators_[0]._validate_X_predict(X, check_input=True)
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/tree/_classes.py", line 407, in _validate_X_predict
X = self._validate_data(X, dtype=DTYPE, accept_sparse="csr",
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/base.py", line 421, in _validate_data
X = check_array(X, **check_params)
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/utils/validation.py", line 63, in inner_f
return f(*args, **kwargs)
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/utils/validation.py", line 720, in check_array
_assert_all_finite(array,
File "/home/mlpoc/miniconda3/envs/churn/lib/python3.8/site-packages/sklearn/utils/validation.py", line 103, in _assert_all_finite
raise ValueError(
ValueError: Input contains NaN, infinity or a value too large for dtype('float32').
Versions
sklearn.__version__
Out[6]: '0.24.2'
lightgbm.__version__
Out[8]: '3.2.1'