eli5 has XGBoost support -
shows feature importances,
eli5.explain_prediction() explains predictions by showing feature weights.
Both functions work for XGBClassifier and XGBRegressor.
explain_prediction_xgboost(xgb, doc, vec=None, top=None, top_targets=None, target_names=None, targets=None, feature_names=None, feature_re=None, feature_filter=None, vectorized=False)¶
Return an explanation of XGBoost prediction (via scikit-learn wrapper XGBClassifier or XGBRegressor) as feature weights.
eli5.explain_prediction()for description of
vecis a vectorizer instance used to transform raw features to the input of the estimator
xgb(e.g. a fitted CountVectorizer instance); you can pass it instead of
vectorizedis a flag which tells eli5 if
docshould be passed through
vecor not. By default it is False, meaning that if
vecis not None,
vec.transform([doc])is passed to the estimator. Set it to True if you’re passing
docis already vectorized.
Method for determining feature importances follows an idea from http://blog.datadive.net/interpreting-random-forests/. Feature weights are calculated by following decision paths in trees of an ensemble. Each leaf has an output score, and expected scores can also be assigned to parent nodes. Contribution of one feature on the decision path is how much expected score changes from parent to child. Weights of all features sum to the output score of the estimator.
explain_weights_xgboost(xgb, vec=None, top=20, target_names=None, targets=None, feature_names=None, feature_re=None, feature_filter=None, importance_type='gain')¶
Return an explanation of an XGBoost estimator (via scikit-learn wrapper XGBClassifier or XGBRegressor) as feature importances.
eli5.explain_weights()for description of
targetsparameters are ignored.
importance_type (str, optional) – A way to get feature importance. Possible values are:
- ‘gain’ - the average gain of the feature when it is used in trees (default)
- ‘weight’ - the number of times a feature is used to split the data across all trees
- ‘cover’ - the average coverage of the feature when it is used in trees