The KMeans algorithm is commonly used to cluster unlabeled data. But with large datasets, scikit-learn takes plenty of time to train and predict. To speed-up KMeans, use Faiss by Facebook AI Research. It provides faster nearest-neighbor search and clustering.
Make Sklearn KMeans 20x times faster
Make Sklearn KMeans 20x times faster
Make Sklearn KMeans 20x times faster
The KMeans algorithm is commonly used to cluster unlabeled data. But with large datasets, scikit-learn takes plenty of time to train and predict. To speed-up KMeans, use Faiss by Facebook AI Research. It provides faster nearest-neighbor search and clustering.