Skip to main content

Weighted KMeans Clustering for Geolocational Problem

Project description

Repo for weighted k means clustering for specifically geo locational problems.

For an example and mathematical explanation:

https://emrahcimren.github.io/data%20science/Greenfield-Analysis-with-Weighted-Clustering/

Prerequisites

Install environment.yml for prerequisites.

` conda env create -f environment.yml `

To recreate environment.yml

` conda env export > environment.yml `

To create requirements.txt from environment.yml

` pip freeze > requirements.txt `

Installation

` pip install cimren-wkmeans-geo `

Inputs

input_locations is a pandas dataframe with the following format.

LOCATION_NAME | LATITUDE | LONGITUDE | WEIGHT — | — | — | — LOC 0 | -27.0065 | 170.583 | 1

number_of_clusters: Number of clusters to be created

minimum_elements_in_a_cluster: Minimum elements in a cluster

maximum_elements_in_a_cluster: Maximum elements in a cluster

maximum_iteration: How many maximum number of steps the algorithm takes to stop if it does not find the solution

enable_minimum_maximum_elements_in_a_cluster: True/False to enable minimum and maximum cluster size

objective_range: Acceptable difference between objectives at each iteration

How to use

Create initial clusters from a given solution. - Number of clusters to be generated should be bigger than number of clusters in the input solution

` from wkmeans_geo import clusters_from_input as cf initial_solution = cf.create_initial_clusters_from_given_input(number_of_clusters, input_locations_with_clusters) `

Create clusters.

` from wkmeans_geo import wkmeans_clustering as wkc clusters, locations_with_clusters = wkc.calculate_clusters(...) `

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

cimren-wkmeans-geo-1.1.0.tar.gz (6.8 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page