NEWS.md
In benitezfj/rmoo: Multi-Objective Optimization in R

rmoo 0.2.0

Considerable changes made to the rmoo package:
All the algorithms implemented in rmoo will now be called through the rmoo() function, passing the strategy to be used as a parameter.
New s4 methods created, the functions scatter(), pcp(), heatmap() and polar() will be called by the plot method and passing the object and plot type.
s4 methods: summary() and print() for brief verification of returned results.
Summary() function deprecated: Now it is executed by the s4 method progress(), the method is used within the execution.
Other s4 methods created are getFitness(), getPopulation(), getDummyFitness(), getCrowdingDistance(), getMetrics(), they were created so that the user does not need to directly access the object slots.
the structure of classes and inheritance were improved, following the rules of object-oriented programming.
Bugs Fixed:
The non-dominance process was solved, an error was generated when the first front was only made up of one individual.
ecr, cdata, dplyr, reshape2 packages are now suggested and not imported. Only the packages that are exactly necessary or that generate errors if they are not imported are found in import.

Minor inconvenient the documentation fixed.
The structure of the non-dominated in the main functions modified: They did not evaluate fronts with one individual.
Condition on validation into the main functions deleted: It was not necessary.

Dataset: Add KRO dataset from TSPLIB to permutation operation in data and inst.
Change donttest to dontrun at the documentation.
Scatter Plotting: scatter() functions for plotting points in 2-D, 3-D and N-D planes, where the dimensions represent the objectives.
Parallel Coordinate Plots: pcp() functions show a fitness set of points in an objectives n-dimensional space.
Heat Map: head_map() function to show a set of fitness values as colors in two dimensions.
rgl packages change to suggests in DESCRIPTION.