Skip to content

Releases: Astroua/TurbuStat

v1.3

05 Dec 15:06
v1.3
Compare
Choose a tag to compare

What's Changed

  • Fix for Tsallis chi squared values with scipy 1.7 by @e-koch in #236
  • Fix Delta Variance fill value by @e-koch in #235
  • Bug fixes from distance metric classes by @e-koch in #237
  • Update testing to drop 3.6 and add 3.9, 3.10 by @e-koch in #239
  • BUG FIX: Fixed name of smoothed image list by @gabrielb09 in #240
  • Updates and fixes to Genus + numpy slice fixes by @e-koch in #241
  • Add unbinned power spectrum fitting by @e-koch in #248
  • Fix check for dendrogram deltas when given as a string by @e-koch in #249

New Contributors

Full Changelog: v1.2.1...v1.3

Fixes for scipy 1.14

01 Dec 01:33
v1.2
99832f1
Compare
Choose a tag to compare

As of scipy 1.14, non-finite values are not ignored by default which causes issues for the azimuthal averaging used for several of the statistics. See #227 for further information.

This release also has large differences in the package infrastructure and a switch from travis to github actions for CI.

Fix compiling cython code in release

27 Nov 00:20
Compare
Choose a tag to compare

The cython code for turbustat.simulator.spectrum was not compiling in the v1.0 release on pypi. This new release fixes that.

First Major Release

20 Apr 23:31
v1.0.0
Compare
Choose a tag to compare

TurbuStat's first major release, with stable implementations and API. This version is described in the accompanying paper to be published in the near future.

Koch+2017 Version

03 Jul 17:44
Compare
Choose a tag to compare

The version used for the results in Koch et al. (2017). The functionality for the methods is tested and ready for use, but the documentation is incomplete.

Initial Release

13 Oct 17:14
Compare
Choose a tag to compare

First release of the Turbustat package: A python package developed to provide an accessible set of statistical measures for comparing observations and simulations of molecular clouds. Also provided are examples of R code used for fitting the models and wrapper scripts for running large sets of data.

Portions of this package are still under current development, but the statistics and distance metrics are in a stable form that is unlikely to have major changes.

This release indicates the version for our submitted paper.