Use shap.force_plot with the hiclass's explainer

[RamSnoussi]

Hi

1. how can I use shap.force_plot with a current definition of explainer in order to obtain the following figure? The example below illustrates this use but cannot work correctly. How can I correct this example?
2. shap.force_plot function have a base value as parameter. This parameter equals to explainer.expected_value
   how can I calculate expected_value with current definition of Explainer?

from hiclass import LocalClassifierPerParentNode, Explainer
from sklearn.ensemble import RandomForestClassifier
import numpy as np
import shap
X_train = np.array([
    [40.7,  1. ,  1. ,  2. ,  5. ,  2. ,  1. ,  5. , 34.3],
    [39.2,  0. ,  2. ,  4. ,  1. ,  3. ,  1. ,  2. , 34.1],
    [40.6,  0. ,  3. ,  1. ,  4. ,  5. ,  0. ,  6. , 27.7],
    [36.5,  0. ,  3. ,  1. ,  2. ,  2. ,  0. ,  2. , 39.9],
])

Y_train = np.array([
    ['Gastrointestinal', 'Norovirus', ''],
    ['Respiratory', 'Covid', ''],
    ['Allergy', 'External', 'Bee Allergy'],
    ['Respiratory', 'Cold', ''],
])

test_sample = np.array([[35.5,  0. ,  1. ,  1. ,  3. ,  3. ,  0. ,  2. , 37.5]])

classifier = LocalClassifierPerParentNode(local_classifier=RandomForestClassifier(), replace_classifiers=False)
classifier.fit(X_train, Y_train)
explainer = Explainer(classifier, data=X_train, mode="tree")
explanations = explainer.explain(test_sample)
mask = {'class': 'Cold', 'level':1}
shap_values = explanations.sel(mask).shap_values
shap.force_plot(shap_values, test_sample)


---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
File ~/anaconda3/envs/hiclass/lib/python3.8/site-packages/IPython/core/formatters.py:344, in BaseFormatter.__call__(self, obj)
    342     method = get_real_method(obj, self.print_method)
    343     if method is not None:
--> 344         return method()
    345     return None
    346 else:

File ~/anaconda3/envs/hiclass/lib/python3.8/site-packages/shap/plots/_force.py:531, in AdditiveForceVisualizer._repr_html_(self)
    530 def _repr_html_(self):
--> 531     return self.html()

File ~/anaconda3/envs/hiclass/lib/python3.8/site-packages/shap/plots/_force.py:516, in AdditiveForceVisualizer.html(self, label_margin)
    510         self.data["labelMargin"] = label_margin
    511         generated_id = id_generator()
    512         return f"""
    513 <div id='{generated_id}'>{err_msg}</div>
    514  <script>
    515    if (window.SHAP) SHAP.ReactDom.render(
--> 516     SHAP.React.createElement(SHAP.AdditiveForceVisualizer, {json.dumps(self.data)}),
    517     document.getElementById('{generated_id}')
    518   );
    519 </script>"""

File ~/anaconda3/envs/hiclass/lib/python3.8/json/__init__.py:231, in dumps(obj, skipkeys, ensure_ascii, check_circular, allow_nan, cls, indent, separators, default, sort_keys, **kw)
    226 # cached encoder
    227 if (not skipkeys and ensure_ascii and
    228     check_circular and allow_nan and
    229     cls is None and indent is None and separators is None and
    230     default is None and not sort_keys and not kw):
--> 231     return _default_encoder.encode(obj)
    232 if cls is None:
    233     cls = JSONEncoder

File ~/anaconda3/envs/hiclass/lib/python3.8/json/encoder.py:199, in JSONEncoder.encode(self, o)
    195         return encode_basestring(o)
    196 # This doesn't pass the iterator directly to ''.join() because the
    197 # exceptions aren't as detailed.  The list call should be roughly
    198 # equivalent to the PySequence_Fast that ''.join() would do.
--> 199 chunks = self.iterencode(o, _one_shot=True)
    200 if not isinstance(chunks, (list, tuple)):
    201     chunks = list(chunks)

File ~/anaconda3/envs/hiclass/lib/python3.8/json/encoder.py:257, in JSONEncoder.iterencode(self, o, _one_shot)
    252 else:
    253     _iterencode = _make_iterencode(
    254         markers, self.default, _encoder, self.indent, floatstr,
    255         self.key_separator, self.item_separator, self.sort_keys,
    256         self.skipkeys, _one_shot)
--> 257 return _iterencode(o, 0)

File ~/anaconda3/envs/hiclass/lib/python3.8/json/encoder.py:179, in JSONEncoder.default(self, o)
    160 def default(self, o):
    161     """Implement this method in a subclass such that it returns
    162     a serializable object for ``o``, or calls the base implementation
    163     (to raise a ``TypeError``).
   (...)
    177 
    178     """
--> 179     raise TypeError(f'Object of type {o.__class__.__name__} '
    180                     f'is not JSON serializable')

TypeError: Object of type DataArray is not JSON serializable

[RamSnoussi]

hi @mirand863,
Any suggestion how I can use shap.force_plot with a hiclass framework? Thanks