For more details about the training and calibration of ROGER please read https://arxiv.org/abs/2010.11959v1
This software make an extensive usage of the following packages that must be installed: caret, randomForest, kernlab
You can install this packages inside an R-session with:
install.packages(c('caret', 'randomForest', 'kernlab'))
You can install the ROGER package directly from your R session using the install_github
function from the devtools
package.
library('devtools')
install_github('MartindelosRios/ROGER')
# Loading the ROGER library.
library('ROGER')
# Loading the data
data('testset')
# Let's see the structure of this dataset
str(testset)
# Let's keep only with the 'r' and 'v' columns that will be used to predict, and
# save the real classification for future comparison.
cat <- testset[, c(1,2)]
real_class <- testset$flag1
# Let's predict the proabability of being of each class using our ML
pred_prob <- get_class(cat, knn)
In the Examples section you can find more examples!
There is also a web interface where you can use the software online and analyze you galaxies!.
Martín de los Rios (ICTP-SAIFR/IFT-UNESP)
Héctor J. Martínez (IATE-OAC-UNC)
Valeria Coenda (IATE-OAC-UNC)
Hernán Muriel (IATE-OAC-UNC)
Andrés N. Ruiz (IATE-OAC-UNC)
Cristian Vega (UNLS)
Sofía Cora (CCT-UNLP)
If you use this software please cite https://arxiv.org/abs/2010.11959v1
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.