First try of a Python implementation of the Generalized DBSCAN algorithm from [1], enabling clustering of point objects using both spatial and nonspatial attributes.
[1]: "Density-based clustering in spatial databases: The algorithm gdbscan and its applications." by Sander, Jörg, et al.
- Improve testing to ensure correctness.
- Improve time complexity (currently O(n^2)) using spatial indexing techniques.
- Use less of an object-oriented style.