Use shap.force_plot with the hiclass's explainer (the first parameter base value)

[RamSnoussi]

Hi @mirand863
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? what can i write in xxxxxxxxx ? below is my code:

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( xxxxxxxxxxx, shap_values, test_sample, matplotlib=True, show=True)

[RamSnoussi]

@mirand863 i used hiclass v4.11.0