NEWS.md

In avrodrigues/naturaList: Classify Occurrences by Confidence Levels in the Species ID

naturaList 0.5.2

• remove dependencies on rgdal, rgeos and maptools.

naturaList 0.5.1

• it includes the env_grid_filter function. The function select the best classified species occurrence regarding the species identification reliability in the environmental space.

naturaList 0.5.0

• grid_filter function was re-coded. The new code is faster than the older version, but it could produce different results from older version.

naturaList 0.4.3

• documentation update

naturaList 0.4.2

• documentation update

naturaList 0.4.1

• bug solved in the classify_occ function

naturaList 0.4.0

• some arguments for column names of the occurrence dataset was changed. They are still working, with warning message informing they are deprecated and what is the current argument accepted.

• bugs solved in create_spec_df():

  • warning message included for specialists names without abbreviation letters
• bug solved in classify_occ():
  • it is possible to use a list of specialists with only 1 row

• Classification level taxonomy was renamed to not_spec_name

• bug solved in map_module():

  • Before if a point was selected, when Delete points with click was activated, the selected points was deleted. Now Delete points with click always wait a click to delete a point.
• new argument action in map_module(), which allows to choose if the output dataset will contain only the selected occurrences, or if the occurrences output dataset will be flagged about the choices of the user in the application.

naturaList 0.3.1

• bugs solved in classify_occ(). Inclusion of a new dataset cyathea.br in the package.

naturaList 0.3.0

• Added functions clean_eval() and define_env_space. These functions implements evaluation of cleaning procedures based in classification made by classify_occ(). The evaluation is based on the change of convex hull area of species occurrences between cleaned and raw occurrences dataset.

avrodrigues/naturaList documentation built on Feb. 9, 2024, 1:13 a.m.