Introduction

Spatial Cluster is an open-source library that aims to facilitate the generation of urban attribute maps that group data into different disjoint sets (clustering). The project consists of 5 main modules, which are as follows:

• Preprocessing: It offers various data preprocessing functions to make them usable by the methods included in the library.

• Methods: Currently, it features 5 different methods that perform the clustering task.

• Visualization: Currently, it provides two functions to generate corresponding maps using the results obtained after applying a clustering algorithm.

• Evaluation: Currently, it includes two evaluation metrics for comparing pairs of maps (ARI and AMI).

• Dataset: It offers a dataset collected from the Metropolitan Region of Chile, which meets the necessary requirements to utilize any method offered by the library.

Installation

To use SpatialCluster, it must first be installed using pip:

(.venv) $ pip install SpatialCluster

Next Steps

To use the library, it is recommended to visit the Methods and Visualization sections to see usage examples. It is also advisable to review the Preprocessing section as it can be quite useful in obtaining good results.