ClassifierChain + StackingClassifier : AttributeError: 'StackingClassifier' object has no attribute 'final_estimator_'

Describe the bug

Getting an exception when chaining together StackingClassifiers using ClassifierChain.

Steps/Code to Reproduce

from sklearn.linear_model import LogisticRegression
from sklearn.datasets import make_multilabel_classification
from sklearn.ensemble import RandomForestClassifier, StackingClassifier
from sklearn.multioutput import ClassifierChain

X, y = make_multilabel_classification(n_classes=3, n_labels=3, random_state=0)
rfc = RandomForestClassifier(random_state=0, n_estimators=1, max_depth=1)
lr = LogisticRegression(random_state=0)
stacking_classifier = StackingClassifier([('rfc', rfc), ('lr', lr)])
classifier_chain = ClassifierChain(stacking_classifier)
classifier_chain = classifier_chain.fit(X, y)
proba = classifier_chain.predict_proba(X)
print(proba)

Expected Results

No error is thrown. Predicted probabilities for all classes of all outputs returned from clf.predict_proba(X)

Actual Results

Traceback (most recent call last):
  File "/Applications/PyCharm.app/Contents/plugins/python/helpers/pydev/pydevd.py", line 1477, in _exec
    pydev_imports.execfile(file, globals, locals)  # execute the script
  File "/Applications/PyCharm.app/Contents/plugins/python/helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile
    exec(compile(contents+"\n", file, 'exec'), glob, loc)
  File "/Users/vq/workspace/tennis-live-win-prob-all/src/bug_report.py", line 14, in <module>
    predictions = clf.predict_proba(X)
  File "/Users/vq/workspace/tennis-live-win-prob-all/venv/lib/python3.7/site-packages/sklearn/utils/metaestimators.py", line 114, in __get__
    getattr(delegate, self.attribute_name)
  File "/Users/vq/workspace/tennis-live-win-prob-all/venv/lib/python3.7/site-packages/sklearn/utils/metaestimators.py", line 117, in __get__
    attrgetter(self.delegate_names[-1])(obj)
AttributeError: 'StackingClassifier' object has no attribute 'final_estimator_'

Versions

System:
    python: 3.7.6 (default, Dec 30 2019, 19:38:28)  [Clang 11.0.0 (clang-1100.0.33.16)]
executable: /Users/vq/workspace/tennis-live-win-prob-all/venv/bin/python
   machine: Darwin-19.6.0-x86_64-i386-64bit
Python dependencies:
          pip: 21.0.1
   setuptools: 52.0.0
      sklearn: 0.24.1
        numpy: 1.20.0
        scipy: 1.6.0
       Cython: None
       pandas: 1.2.1
   matplotlib: 3.3.4
       joblib: 1.0.0
threadpoolctl: 2.1.0

[thomasjpfan]

@snaraya7 In this issue, the StackingClassifier is placed in a ClassifierChain, which does not require StackingClassifier.fit to be called. The internals of ClassifierChain.fit will clone StackingClassifier removing the fitted attributes anyways:

https://github.com/scikit-learn/scikit-learn/blob/986d8f2524ad06b87c73e0e1f80dd49a4c9ca1e6/sklearn/multioutput.py#L554

[snaraya7]

‘final_estimator_’ exception is thrown when StackingClassifier.fit( ) method is not called before predict( ).