K Means using PyTorch

PyTorch implementation of kmeans for utilizing GPU

Alt Text

Getting Started


import torch
import numpy as np
from kmeans_pytorch import kmeans

# data
data_size, dims, num_clusters = 1000, 2, 3
x = np.random.randn(data_size, dims) / 6
x = torch.from_numpy(x)

# kmeans
cluster_ids_x, cluster_centers = kmeans(
    X=x, num_clusters=num_clusters, distance='euclidean', device=torch.device('cuda:0')
)

see example.ipynb for a more elaborate example

Requirements

PyTorch version >= 1.0.0
Python version >= 3.6

Installation

install with pip:

pip install kmeans-pytorch

Installing from source

To install from source and develop locally:

git clone https://github.com/subhadarship/kmeans_pytorch
cd kmeans_pytorch
pip install --editable .

CPU vs GPU

see cpu_vs_gpu.ipynb for a comparison between CPU and GPU

Notes

useful when clustering large number of samples
utilizes GPU for faster matrix computations
support euclidean and cosine distances (for now)

Credits

This implementation closely follows the style of this
Documentation is done using the awesome theme jekyllbook

kmeans PyTorch