EarthPy Release Notes

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.6.6]

  • Add sphinx gallery for vignettes and update get started page (@lwasser #279, #203)
  • Add two example vignettes for using EarthPy with raster data (@lwasser)
  • Fix bug in bytescale() - ensure math to calc range is floating point vals (@lwasser #282)
  • Fix tests for bytescale() to ensure the bug raised in #282 is fixed / tested; also added comment to plot_rgb docs to ensure users consider nodata values before plotting (@lwasser #282)

[0.6.5] - 2019-03-37

  • Add continuous integration testing on osx via Travis CI (@mbjoseph #228)
  • Add cbar legend to plot_bands() and scaling parameters (@lwasser #274)
  • BUGFIX: plot_bands() doesn’t plot single string titles properly + add test (@lwasser #258)
  • Remove dependency on download library (@mbjoseph #249)
  • BUGFIX: draw_legend() fails when classes are provided (@lwasser #253)
  • Remove earthpy.utils.fix_paths() as it is not used in the package (@lwasser #259)
  • Adding tests for hillshade() and improved docs (@jpalomino #260)
  • Closing plots in tests (@lwasser #257)
  • Added a code of conduct (@mbjoseph, #27)
  • Added CI testing across python versions and on Windows (@mbjoseph, #228)
  • Added tests for EarthlabData class (@mbjoseph, #37)

[0.6.2] - 2019-02-19

We have made significant changes in preparation for a 1.0 release on PyPI. If you have used EarthPy previously, please review the changes below. You may need to update your code accordingly.

Changed

  • normalized_diff() function arguments have been flipped. Please update all of your old code accordingly. Changes include:
    • Arguments are now provided as normalized_diff(b1, b2)
    • Math will be calculated (b1-b2) / (b1+b2)
    • Example: ndvi = es.normalized_diff(b1=nir_band, b2=red_band)
  • normalized_diff() now:
    • returns unmasked array by default
    • returns masked array if there are nan values present
    • converts infinity values that result from division by zero to nan values
  • stack() replaces stack_raster_tifs(), which is now deprecated. The new stack() function works similarly to stack_raster_tifs. Inputs parameters are now a list of filepaths and an optional filepath parameter that when populated saves the stacked raster array as a geotiff file. The default return is a numpy ndarray.
  • The parameter order for bytescale has changed:
    • PREVIOUS ORDER: data, cmin=None, cmax=None, high=255, low=0
    • NEW ORDER: data, high=255, low=0, cmin=None, cmax=None
  • All plot functions moved to a new earthpy.plot module. To import plot functions first import the plot module: import earthpy.plot as ep. Then you can call functions as follows:
    • ep.plot_rgb()
    • ep.draw_legend()
    • ep.hist()
    • ep.colorbar()
    • ep.plot_bands()

Added

  • draw_legend() works now on different types of categorical raster plots.
  • colorbar() has also been fixed to work given updates to Matplotlib
  • A new mask function has been released as mask_pixels().
  • We now have tests through pytest that run on Travis CI.
  • API documentation is now generated automatically from docstrings with autodoc.
  • Example code in documentation is tested using doctest

Deprecated

  • stack_raster_tifs() has been deprecated and replaced with stack().