Imbalanced-learn

0.14.1 · active · verified Wed Apr 08

A Python library for handling imbalanced datasets in machine learning, currently at version 0.14.1, with a release cadence of approximately every 6 months.

Warnings

breaking Compatibility issues with scikit-learn 1.6.0 and sklearn-compat 0.1.3 in imbalanced-learn 0.13.0.
Fix: Upgrade to imbalanced-learn 0.14.1 or later.
gotcha ImportError: cannot import name 'MultiOutputMixin' from 'sklearn.base'.
Fix: Restart the Python kernel after installation to resolve import errors. ([stackoverflow.com](https://stackoverflow.com/questions/60254805/imbalanced-learn-import-error-cannot-import-name-multioutputmixin?utm_source=openai))
gotcha ImportError: cannot import name 'validate_data' from 'sklearn.utils.validation'.
Fix: Upgrade to imbalanced-learn 0.14.1 or later. ([github.com](https://github.com/scikit-learn-contrib/imbalanced-learn/issues/1119?utm_source=openai))

Install

pip install imbalanced-learn Install imbalanced-learn

Imports

RandomOverSampler
```
from imblearn.over_sampling import RandomOverSampler
```
Ensure correct import path to avoid ImportError.

Quickstart

A quickstart example demonstrating the use of RandomOverSampler to balance an imbalanced dataset.

import numpy as np
from imblearn.over_sampling import RandomOverSampler

# Sample data
X = np.array([[1, 2], [1, 3], [2, 3], [3, 4], [5, 6], [7, 8], [8, 9], [8, 10], [9, 10], [10, 11]])
y = np.array([0, 0, 0, 0, 1, 1, 1, 1, 1, 1])

# Initialize RandomOverSampler
ros = RandomOverSampler(random_state=42)

# Fit and resample
X_res, y_res = ros.fit_resample(X, y)

print(f'Resampled dataset shape: {np.bincount(y_res)}')

view raw JSON →