BaggingClassifier.__init__() got an unexpected keyword argument 'base_estimator'

[PoojaPrasd]

Dear Team,

I am running the titanic demo notebook and getting the below error in the code

skope_rules_clf.fit(X_train, y_train)

in SkopeRules.fit(self, X, y, sample_weight)
265 self._max_depths = self.max_depth
266 if isinstance(self.max_depth, Iterable) else [self.max_depth]
268 for max_depth in self.max_depths:
--> 269 bagging_clf = BaggingClassifier(
270 base_estimator=DecisionTreeClassifier(
271 max_depth=max_depth,
272 max_features=self.max_features,
273 min_samples_split=self.min_samples_split),
274 n_estimators=self.n_estimators,
275 max_samples=self.max_samples,
276 max_features=self.max_samples_features,
277 bootstrap=self.bootstrap,
278 bootstrap_features=self.bootstrap_features,
279 # oob_score=... XXX may be added
280 # if selection on tree perf needed.
281 # warm_start=... XXX may be added to increase computation perf.
282 n_jobs=self.n_jobs,
283 random_state=self.random_state,
284 verbose=self.verbose)
286 bagging_reg = BaggingRegressor(
287 base_estimator=DecisionTreeRegressor(
288 max_depth=max_depth,
(...)
300 random_state=self.random_state,
301 verbose=self.verbose)
303 clfs.append(bagging_clf)

TypeError: BaggingClassifier.init() got an unexpected keyword argument 'base_estimator'

[Lapsis-glitch]

I ran into the same issue, it seems to stem from sklearn modifying the function parameters.

This can be fixed if you go into the code and change "base_estimator" to "estimator".

on linux you can do this with:

sed -i 's/base_estimator=/estimator=/g' path/to/your/environment/lib/python3.XX/site-packages/skrules/skope_rules.py

[liang623589326]

I also encounter with this problem. Can you tell me how to fixed the parameter ("estimator") function or test demo? Thank you very much!

[Lapsis-glitch]

I also encounter with this problem. Can you tell me how to fixed the parameter ("estimator") function or test demo? Thank you very much!

You have to locate the skope_rules.py file in your python environment (the path to it is usually written in the error message you get when trying to run this function).

Open the file in a text editor of your choice and replace every instance of the base_estimator keyword with estimator and save it. This should then allow the code to run as intended.

[liang623589326]

Thank you so much. I've done with you help.

[yousseftarhri]

base_estimator was renamed to estimator in version 1.2 and will be removed in 1.4.