TODO.md

In BlasBenito/sdmflow: Streamlining the Development of Species Distribution Models

TO DO

v_{} functions

Functions to deal with raster data representing environmental information

import4D: rename to v_import

Changes in arguments:

• raster.template -> rename to v.template
• raster.template.crs -> rename to v.template.crs
• folder -> rename to v.folder
• dynamic.vars -> rename to v.dynamic
• static.vars -> rename to v.static
• times -> rename to v.times
• vars.crs -> rename to
• to.data.frame -> remove this option and keep it in a helper function

importTIF: include in v_import

importASC: include in v_import

o_{} functions

Functions to deal with occurrence data

o_make_training()

Changes in argumets:

• variables -> find proper name for this argument after v_{} functions are ready
• n -> rename to n.background
• presence.only -> remove this option
• background -> remove this option
• restricted.background -> remove this option, link it to non-null restricted.background.buffer
• restricted.background.buffer -> restriction.radius? restrict.to.buffer?
• thinning -> remove this option, link it to non-null minimum.distance
• minimum.distance -> rename to same as in o_thinning(), maybe min.dist

o_thinning()

This function should be paired with the function to test spatial correlation (testSpatialCorrelation()). Maybe o_thinning() should be s_thinning() and testSpatialCorrelation() should be s_spatial_cor(), so if there are variables with a high spatial correlation for the presences, the user can apply thinning and/or remove the offending variables. This should be done most likely before assessing multicollinearity.

Changes in arguments:

• variables -> find proper name for this argument after v_{} functions are ready
• minimum.distance -> rename to min.dist
• random.start -> remove this option, link to a non-null random.seed
• seed -> rename to random.seed

s_{} functions

Functions for variable selection and reduction of multicollinearity

BlasBenito/sdmflow documentation built on April 10, 2020, 2:31 a.m.