10. Change Log

Changes to each release will be added here

10.1. Release: 0.0.4

Date: 2021 April 16

Description: Fixes to the batch.spatial_crop function

  • Added gdf_shft_e_m and gdf_shft_n_m to spatial_mod.crop_single function. This adds flexibility for shifting a crop plot in some direction so it is better centered within the gdf plot boundary.

  • Changed column name requirement from “plot” to “plot_id” to indicate plot ID in gdf (relevant for spatial_mod.crop_single and spatial_mod.crop_many_gdf functions).

  • When cropping by gdf bounds in spatial_mod, if gdf bounds are outside image extent, the cropped image had an extra “buffer” equal to the length between the gdf bound and corresponding extent of the image. The new behavior is such that the image is croped at the edge of the image and geogransform is shifted according to distance gdf is outside of image extent.

  • Renamed spatial_mod.crop_single plot_id parameter to plot_id_ref to be consistent with spatial_mod.crop_many_gdf.

  • Many bug fixes making the spatial_crop function more robust to various scenarios and to perform quality control checks on input geodataframe of plot boundaries for spatial cropping.

  • During batch processing, if out_force is False and files exist in output directory, hs_process will skip over files instead of raising an error that the file already exists. This is useful if many files have already been processed and you’d like to process the remaining files without reprocessing all those that are completed.

  • Added an option to create a mask between two thresholds or percentiles (applies to hstools.mask_array and batch.segment_create_mask).

  • Added options to batch.segment_create_mask to provide a choice whether the datacube and .spec files are saved to file. This potentially saves disk space if the datacube isn’t needed in subsequent analyses.

  • User can optionally pass a lists for mask_dir, mask_side, and mask_thresh/mask_percentile to batch.segment_create_mask

  • Added a progress bar the following batch functions: cube_to_spectra, segment_composite_band, segment_band_math, segment_create_mask, spectral_clip, and spectral_smooth.

  • Added get_wavelength_range to utilities class.

  • Added spec_mod.spec_derivative, spec_mod.spectral_mimic, and spec_mod.spectral_resample functions.

10.2. Release: 0.0.3

Date: 2020 February 17

Description: Trying to build so hs_process works with Anaconda builds.

  • Adjusted the requirements to be only geopandas, seaborn, and spectral (others are dependencies of these three).

  • data, tests, and example folders added to installation

  • shortened name of sample datacube because pip won’t copy it with such a long filename

10.3. Release: 0.0.2

Date: 2020 February 17

Description: Uploading a source distribution to PyPI

  • Simplified the tests to use a small datacube

10.4. Release: 0.0.1

Date: 2020 February 16

Description: Initial release