Skip to content

Releases: materialsinnovation/pymks

Version 0.4.1

31 Jul 02:49
Compare
Choose a tag to compare
  • Update Nix build to 21.05 #561
  • New documentation design #559
  • Better API documentation #543
  • Use Nix to Build Docker instance #558
  • Fix bug with ipywidgets and Jupyter #557
  • Fix issue with train test split and flat arrays when 2D arrays are expected #552
  • Add Graspi notebook and Graspi as an optional dependency #553
  • Add a numpy only introduction notebook #548
  • Ensure functions and classes work with Numpy not just Dask #546
  • Use versioneer for versioning #542
  • Add pair correlations functionality #524
  • Add conda-forge installation to Circle CI #521
  • Fix bug with non-periodic auto-correlations #518

Version 0.4.1a1

29 Jun 20:32
5f42557
Compare
Choose a tag to compare
Merge pull request #548 from beyucel/HowToNotebook

How to Notebook for new users

Version 0.4

17 Aug 20:00
7780e16
Compare
Choose a tag to compare
  • New API that works with Scikit-learn pipelines
  • Use Circle CI for testing
  • Parallel, distributed and out-of-memory support using Dask and
    Dask-ML
  • Deprecated old API
  • Internals are now programmed in a functional manner rather than
    object oriented
  • Nix build support
  • Update for Zenodo

Version 0.3.4

28 Jul 15:02
Compare
Choose a tag to compare
  • Integrate PyMKS into Conda-forge
  • Notebook testing
  • Full Python 3 functionality

Version 0.3.3

06 Oct 20:38
Compare
Choose a tag to compare
v0.3.3

Fix mistake in setup.py version number

0.3.1

22 May 06:29
Compare
Choose a tag to compare

Fixed bugs with plotting spatial statistics
Fixed bugs with pyfftw
Update installations to include conda builds
Added binder to README.md

0.3.0

13 Feb 13:35
Compare
Choose a tag to compare

The highlights of this release inclue

  • The MKSStructureAnalysis class that can be used to do dimensionality reduction of 2-point statistics
  • The GSHBasisi class that can be used to represent orientations
  • The Fourer basis class that can be used to represent continuous one dimensional periodic local states
  • Improved memory management
  • Argument to parallelize pyFFTW
  • 5 new examples that demonstrate some of the new functionalities in PyMKS.

0.2.3

21 Aug 18:05
Compare
Choose a tag to compare

This release

  • resolves error when using SfePy 2015.2
  • adds installation instructions for OS X
  • fixes typos in documentation

0.2.1

25 Jun 04:22
Compare
Choose a tag to compare

Update documentation and fix bugs with numpy arrays with dtype complex128 in MKSLocalizationModel

0.2.0

16 Jun 13:22
Compare
Choose a tag to compare

Update version