Collection of clustering algorithms for spatial analysis.

Let's set up our environment first. I prefer to use pyenv, but you can use whatever environment manager you'd like.

https://realpython.com/intro-to-pyenv/

Install Python

pyenv install python 3.11.6

Create a new pyenv environment

pyenv virtualenv 3.11.6 clustering_algs

Activate the new environment

pyenv activate clustering_algs

Install the required modules

pip install -r requirements.txt

data_path

"input_file/test_points.shp"

DBSCAN Clustering

clustered_data = dbscan_cluster(data_path, 'dbscan_clusters')

K-Means Clustering

kmeans_cluster(data_path, 5, "kmeans_clusters")

K-Menoids PAM Clustering

df_clustered = run_kmedoids_pam_clustering(data_path, 5, "kmenoids_clusters")

Evalute Clusters (K-Means)

print(evaluate_clusters(data_path, 5))

Here are a few example visualizations of the output from using the different algorithms against a dataset containing US Nuclear Power Plants.

K-Means for the Nuclear Powerpoint Dataset with 5 clusters (means in white):

K-Menoids for the Nuclear Powerpoint Dataset with 5 clusters (menoids in white):

Rilpey's K against the Nuclear Powerpoint Dataset (in the input_file folder) and the synthetically generated, evenly distributed data: