clustering-scikitlearn.ipynb

[asif-nuaa]

When I change number of samples to 20 in the code,
db = DBSCAN(eps=epsilon, min_samples=20, algorithm='ball_tree', metric='haversine').fit(np.radians(coords))

I am getting the error, IndexError: list index out of range on the line in get_centermost_point(cluster)
----> 2 centroid = (MultiPoint(cluster).centroid.x, MultiPoint(cluster).centroid.y)

[SparkyBoilermaker]

You need to remove the cluster with label -1. When using minimum cluster size > 1, some points are classified as outliers and get cluster label -1. This needs to be removed from cluster_label and the number of clusters must be one less for consistency.

Added to remove the last cluster -1 (outliers)

clusters = clusters[0:-1]
cluster_list = list(set(cluster_labels))
cluster_list = cluster_list[0:-1]
num_clusters = len(cluster_list)

Additionally, you could add an if condition to check if minimum cluster size is greater than one (or if the last cluster is an empty vector) and put the above segment of code inside the if condition.