elpidiofilho/labgeo: Collection of functions to fit models with emphasis in land use and soil mapping

Labgeo is a R package to fit classifications and regressions models in an easy way. This package addresses the teaching and research activities in digital soil mapping and image classification. The main purpose is to encapsulate within the package the steps of adjustment of models, such as: 1- Reduction of data dimensionality through the removal of variables highly correlated and/or removal of few importance variables. 2- Comparation between several models and selection of the best model for a certain set of data, automatically. 3- Model a large number of response variables (outcome) by comparing the several models for each one. 4 - Carry out the prediction of the best adjusted models using co-variables in the form of rasters files and satellite images. The functions were written for automatic execution in parallel whenever possible, taking advantage of multi-core CPUs, aiming agility in more complex modeling. The model settings are made using the Caret package written by Max Kuhn (https://github.com/topepo/caret). The Caret package creates a standard interface for accessing a few hundred models, is a Swiss Army knife of data modeling.

Getting started

Package details

AuthorElpidio I. Fernandes Filho
MaintainerElpidio Filho <[email protected]>
Package repositoryView on GitHub
Installation Install the latest version of this package by entering the following in R:
elpidiofilho/labgeo documentation built on Aug. 16, 2018, 4:41 a.m.