XGBoost Multi-class Example

In [1]:

    

import sklearn
from sklearn.model_selection import train_test_split
import numpy as np
import shap
import time
import xgboost

X_train,X_test,Y_train,Y_test = train_test_split(*shap.datasets.iris(), test_size=0.2, random_state=0)

shap.initjs()


    

In [2]:

    

model = xgboost.XGBClassifier(objective="binary:logistic", max_depth=4, n_estimators=10)
model.fit(X_train, Y_train)


    

    
Out[2]:





XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,
       colsample_bytree=1, gamma=0, learning_rate=0.1, max_delta_step=0,
       max_depth=4, min_child_weight=1, missing=None, n_estimators=10,
       n_jobs=1, nthread=None, objective='multi:softprob', random_state=0,
       reg_alpha=0, reg_lambda=1, scale_pos_weight=1, seed=None,
       silent=True, subsample=1)

In [3]:

    

shap_values = shap.TreeExplainer(model).shap_values(X_test)
shap.summary_plot(shap_values, X_test)