I have created four random forest models they have the same X data, but their y data are four different response variables. The sklearn random forest feature importance is identical for all four. All four models achieve their purpose and make different predictions, but their random forest feature importance is the same.

Has anyone experienced this before?

I created the models with a series of nested objects like illustrated below. I used the same code before without having identical random forest feature importances, there was however the difference that inside each object I ran a 3-fold CV to determine max_features, whereas here I just used the default which is all of them.

Current code:

class NoCVMethod:
    def __init__(self, X_train, y_train, X_test, y_test, y, Method):
        self.clf = Method
        self.clf.fit(X_train, y_train)
        self.predictions = self.clf.predict(X_test)
        self.rev_preds = rev_pred(y[-(13978+97):].values,self.predictions)
        self.residuals = y_test - self.rev_preds
        self.RMSE = np.mean((self.residuals)**2)**0.5
class Different_variables:
    def __init__(self, X_train, y_train, X_test, y_test, Method):
        self.TSS = NoCVMethod(X_train, y_train[y_train.columns.tolist()[0]], X_test, y_test[y_test.columns.tolist()[0]], y[y.columns.tolist()[0]], Method)
        self.NOx = NoCVMethod(X_train, y_train[y_train.columns.tolist()[1]], X_test, y_test[y_test.columns.tolist()[1]], y[y.columns.tolist()[1]], Method)
        self.NH4 = NoCVMethod(X_train, y_train[y_train.columns.tolist()[2]], X_test, y_test[y_test.columns.tolist()[2]], y[y.columns.tolist()[2]], Method)
        self.PO4 = NoCVMethod(X_train, y_train[y_train.columns.tolist()[3]], X_test, y_test[y_test.columns.tolist()[3]], y[y.columns.tolist()[3]], Method)

Old code:

class CVMethod:
    def __init__(self, X_train, y_train, X_test, y_test, y, param_dict, Method):
        self.pipeline = Pipeline([
            ('scale', StandardScaler()),
            ('clf', Method)
        self.param_grid = param_dict
        self.grid = GridSearchCV(self.pipeline, param_grid = self.param_grid, cv = 3, verbose = False, n_jobs = -1)
        self.grid.fit(X_train, y_train)
        self.predictions = self.grid.predict(X_test).ravel()
        self.rev_preds = rev_pred(y[-(13978+97):].values,self.predictions)
        self.residuals = y_test - self.rev_preds
        self.RMSE = np.mean((self.residuals)**2)**0.5
class CVDifferent_variables:
    def __init__(self, X_train, y_train, X_test, y_test, param_dict, Method):
        self.TSS = CVMethod(X_train, y_train[y_train.columns.tolist()[0]], X_test, y_test[y_test.columns.tolist()[0]], y[y.columns.tolist()[0]], param_dict, Method)
        self.NOx = CVMethod(X_train, y_train[y_train.columns.tolist()[1]], X_test, y_test[y_test.columns.tolist()[1]], y[y.columns.tolist()[1]], param_dict, Method)
        self.NH4 = CVMethod(X_train, y_train[y_train.columns.tolist()[2]], X_test, y_test[y_test.columns.tolist()[2]], y[y.columns.tolist()[2]], param_dict, Method)
        self.PO4 = CVMethod(X_train, y_train[y_train.columns.tolist()[3]], X_test, y_test[y_test.columns.tolist()[3]], y[y.columns.tolist()[3]], param_dict, Method)
  • $\begingroup$ could you give details on the four y ? What do you mean by different ? $\endgroup$
    – etiennedm
    Aug 6, 2020 at 8:33
  • $\begingroup$ @etiennedm the four different y are four different response variables (TSS, NOx, NH4, PO4). They differ in values and distribution. Update: I think the problem comes with the self.clf property being overwritten, but I do not know why $\endgroup$ Aug 6, 2020 at 8:38
  • $\begingroup$ in the __init__ of your class Method, you are passing another Method as parameter, isn't it weird ? is that also a type Method ? (you should use lower case for argument if the class name exists) $\endgroup$
    – etiennedm
    Aug 6, 2020 at 8:44
  • $\begingroup$ Thanks, I understand better now. I think your previous comment might be the reason of the unexpected behaviour. I have just proposed a solution, hope it helps. $\endgroup$
    – etiennedm
    Aug 6, 2020 at 10:23

1 Answer 1


It seems that your self.clf points to your Method. At the end, you are probably printing the features importance of a unique classifier.

Maybe you should copy it:

from sklearn.base import clone

class NoCVMethod:
    self.clf = clone(Method) # only copy the estimator
    # OR
    self.clf = deepcopy(Method) # if you want to also copy the data estimator

See here (or here as you suggested) for more details about copying an sklearn estimator.

  • $\begingroup$ I had some errors with this method, something with an issue with cloning a method that does not take a set of parameters?? Anyway I tried a similar method from: stackoverflow.com/questions/33576024/… and it worked. - Thank you so much :) $\endgroup$ Aug 7, 2020 at 11:31
  • $\begingroup$ Hi @SebastianTopalian, I'm glad it helped. I have updated the answer to include your comment. If this or any answer has solved your question please consider accepting it by clicking the check-mark. This indicates to the wider community that you've found a solution. There is no obligation to do this. $\endgroup$
    – etiennedm
    Aug 7, 2020 at 12:21
  • $\begingroup$ Haha thank you for teaching me the way of the community. I have been lurking for quite some time, but I figured it was time to get in on the action. :) $\endgroup$ Aug 7, 2020 at 12:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.