Skip to content

Releases: martinfleis/clustergram

Version v0.8.1

27 Mar 20:07
Compare
Choose a tag to compare
Version 0.8.1

Version 0.8.0

01 Aug 08:21
Compare
Choose a tag to compare

API changes:

  • The first argument in Clustergram.fit() is now X instead of data.

Enhancements:

  • API: follow scikit-learn API conventions (#59)
  • ENH: ensure that x ticks are only integers (#58)
  • ENH: format time in verbose mode

Full Changelog: v0.7.0...v0.8.0

Version 0.7.0

15 Jan 19:41
Compare
Choose a tag to compare

Enhancements:

  • ENH: allow weighting by a custom principal component (#35)

Compatibility notes:

  • clustergram now requires Python 3.8
  • RAPIDS.AI implementation has been tested with version 22.12.

Minor notes:

  • examples dictionary has been removed. Refer to the notebooks in the documentation.

Version v0.6.0

12 Nov 12:09
Compare
Choose a tag to compare

Enhancements:

  • ENH: optionally measure BIC during GMM (#21)

Bug fixes:

  • BUG: cuML non-weighted plot fix (#25)

Full Changelog: v0.5.1...v0.6.0

Version v0.5.1

24 May 12:39
Compare
Choose a tag to compare

Bugfix for from_data method with non-default indices.

Bugs:

  • BUG: cluster centers empty due to index mismatch (#19)

Version v0.5.0

11 May 21:07
Compare
Choose a tag to compare

Clustergram now supports interactive plotting using a new .bokeh() method based on BokehJS. It
can be handy for the exploration of larger and more complex clustergrams or those with significant outliers.

Enhancements:

  • ENH: support interactive bokeh plots (#14)
  • ENH: skip k=1 in K-Means implementations (#18)
  • documentation restructuring

Version v0.4.0

27 Apr 18:58
Compare
Choose a tag to compare

Spring comes with native hierarchical clustering and the ability to create clustergam from a manual input.

Enhancements:

  • ENH: support hierarchical clustering using scipy (#11)
  • ENH: from_data and from_centers methods (#12)

Version 0.3.0

31 Jan 20:22
Compare
Choose a tag to compare

API chages:

  • pca_weighted is now keyword of Clustergram.plot() not init.

Enhancements:

  • Support MiniBatchKMeans (sklearn)
  • Custom __repr__
  • Expose cluster labels obtained during the loop
  • Expose cluster centers
  • Silhouette score
  • Calinski and Harabasz score
  • Davies-Bouldin score

Version 0.2.1

23 Dec 15:02
Compare
Choose a tag to compare

Minor release updating documentation to acknowledge that clustergram can be installed from conda-forge.

Version 0.2.0

21 Dec 23:01
Compare
Choose a tag to compare

Version 0.2.0 brings support of Gaussian Mixture Models (using scikit-learn) and few minor changes.

Enhancements:

  • Gaussian Mixture Model support (#4)
  • Verbosity - Clustergram now indicates the progress
  • Additional arguments can be passed to the PCA object

Bug fixes:

  • BUG: avoid LinAlgError: singular matrix