RandomUnderSampler' object has no attribute 'fit_resample'

Question

I am using RandomUnderSampler from imblearn, but I get the following error. Any ideas? Thanks

from imblearn.under_sampling import RandomUnderSampler
print('Initial dataset shape %s' % Counter(y.values.squeeze()))
rus = RandomUnderSampler(random_state=42)
X_undersampled, y_undersampled = rus.fit_resample(X, y)
y_undersampled = y_undersampled.squeeze()

output:

Initial dataset shape Counter({0: 2499739, 1: 1558})

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-18-4fe9dcfbd68e> in <module>
      1 print('Initial dataset shape %s' % Counter(y.values.squeeze()))
      2 rus = RandomUnderSampler(random_state=42)
----> 3 X_undersampled, y_undersampled = rus.fit_resample(X, y)
      4 y_undersampled = y_undersampled.squeeze()
      5 

AttributeError: 'RandomUnderSampler' object has no attribute 'fit_resample'

main libraries I am using:

imbalanced-learn==0.3.3
pandas==0.24.2
numpy==1.15.4
scikit-learn==0.19.2

Solution

The method fit_resample was introduced lately to imbalanced-learn API. Either update imbalanced-learn or use fit_sample instead.

Comments

[Pramit]

Had a similar problem few days back. Had to use fit_sample from version 0.6.2

[Miguel Rueda]

@Pramit Same issue here using the version 0.7.0 and 0.22.1 sci-kit version.

[Miguel Rueda]

I solved this problem by updating the scikit-learn version to 0.23.1