dtaidistance.clustering.medoids

Time series clustering using medoid-based methods.

author:Wannes Meert
copyright:Copyright 2020 KU Leuven, DTAI Research Group.
license:Apache License, Version 2.0, see LICENSE for details.
class dtaidistance.clustering.medoids.KMedoids(dists_fun, dists_options, k=None, initial_medoids=None, show_progress=True)

KMedoids using the PyClustering package.

Novikov, A., 2019. PyClustering: Data Mining Library. Journal of Open Source Software, 4(36), p.1230. Available at: http://dx.doi.org/10.21105/joss.01230.

https://pyclustering.github.io/docs/0.9.0/html/d0/dd3/classpyclustering_1_1cluster_1_1kmedoids_1_1kmedoids.html

fit(series)
plot(filename=None, axes=None, ts_height=10, bottom_margin=2, top_margin=2, ts_left_margin=0, ts_sample_length=1, tr_label_margin=3, tr_left_margin=2, ts_label_margin=0, show_ts_label=None, show_tr_label=None, cmap='viridis_r', ts_color=None)
class dtaidistance.clustering.medoids.Medoids(dists_fun, dists_options, k, show_progress=True)
Parameters:
  • dists_fun
  • dists_options
  • show_progress
plot(filename=None, axes=None, ts_height=10, bottom_margin=2, top_margin=2, ts_left_margin=0, ts_sample_length=1, tr_label_margin=3, tr_left_margin=2, ts_label_margin=0, show_ts_label=None, show_tr_label=None, cmap='viridis_r', ts_color=None)