felix-last / kmeans_smote Goto Github PK

Hello, why is there no result after I run the program

The input space is the entire data set or just a few class data sets in the data set?

It seems that the latest version is not compatible with Python2:

Clearly, Python2's open() function does not have the argument 'encoding', which causes the error in installation. Is there any previous version of the library compatible with python2?

Hello, can you tell me?how to run test_kmeans_smote.py,and how to call those test functions, I have already said that the value of the plot is changed to True, but the result is the following error.---------------------------------------------------------------------------

TypeError Traceback (most recent call last)
in
----> 1 test_smoke(True)

in test_smoke(plot)
48
49 """Execute k-means SMOTE with default parameters"""
---> 50 kmeans_smote = KMeansSMOTE(random_state=RND_SEED)
51 X_resampled, y_resampled = kmeans_smote.fit_sample(X, Y)
52

TypeError: Can't instantiate abstract class KMeansSMOTE with abstract methods _sample

your proposed hybrid approach is it applicable for multiclass dataset?

It seems that the results cannot be run on the pycharm platform

where can I download this package---'imbtools'

Hi, @felix-last Thanks for sharing your great job. When I run the following codes, I met a Type Error. It seems that KMeansSMOTE didn't implement the abstract method _fit_resample. Can you help to fix it?

Using TensorFlow backend.
Class -1 has 896 instances
Class 1 has 41 instances
Traceback (most recent call last):
File "test_kmeans_smote.py", line 12, in
smote_args={'k_neighbors': 10})
TypeError: Can't instantiate abstract class KMeansSMOTE with abstract methods _fit_resample

import numpy as np
from imblearn.datasets import fetch_datasets
from kmeans_smote import KMeansSMOTE

datasets = fetch_datasets(filter_data=['oil'])
X, y = datasets['oil']['data'], datasets['oil']['target']

[print('Class {} has {} instances'.format(label, count))
 for label, count in zip(*np.unique(y, return_counts=True))]

kmeans_smote = KMeansSMOTE(kmeans_args={'n_clusters': 100},
    smote_args={'k_neighbors': 10})
X_resampled, y_resampled = kmeans_smote.fit_sample(X, y)

[print('Class {} has {} instances after oversampling'.format(label, count))
 for label, count in zip(*np.unique(y_resampled, return_counts=True))]

Why didn't I get the visualization after running the test program? I only got the results in the example, but there was no visualization after test_kemans_smote.py