Description Usage Arguments Value Examples
Method used to start the multi-objective algorithm portfolio selection
1 2 3 | selectPortfolio(data, var.cols, algo.col, repl.col, indicator = "hv",
ref.point = c(1.1, 1.1), lambda = 100, eta = 0.5, w = c(0.05, 0.95),
cp = 0.1, normalize = TRUE)
|
data |
[ |
var.cols |
[ |
algo.col |
[ |
repl.col |
[ |
indicator |
[ |
ref.point |
[ |
lambda |
[ |
eta |
[ |
w |
[ |
cp |
[ |
normalize |
[ |
Object of class frontTestResult
. Named list with the elements:
non.dominated.algos [logical
]Named vector, each element
corresponds to one algorithm. TRUE if algorithm is selected in step 1.
algos.domination.count [numeric
]Named vector, each element
corresponds to one algorithm. Number of replications with non-dominated
points for each algorithm.
relevant.algos [logical
]Named vector, each element
corresponds to one algorithm selected in step 1. TRUE if the algorithm
is selected in step 2.
algos.selection.vals [data.frame
]Data.frame with 4 cols and
2^(number of algorithms after in step 2) - 1 rows. First cols active in
this subset. Last 2 rows are numeric. First is the number of algos in this
subset, second one the optimality gap..
best.algo.order [Factor
]Vector gives the order of the algorithms
on the common Pareto front, algorithms low values of the first var.col first.
split.vals [numeric
]Split values between the algorithms given
in best.algo.order.
args [list
] List containing all input arguments
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | ## Not run:
# Load data - for the data with subsampling enabled use apprSubsampleSVMParetoFronts
data(apprSVMParetoFronts)
# Avaible datasets: codrna, mnist, protein, vehicle
data = subset(apprSVMParetoFronts, apprSVMParetoFronts$dataset == "mnist")
# Start the front analysis with the main procedure
res = selectPortfolio(
data = data,
var.cols = c("error", "execTime"),
algo.col = "solver",
repl.col = "repl",
indicator = "hv",
ref.point = c(1.1, 1.1),
eta = 0.5,
w = c(0.05, 0.95),
cp = 0.01,
normalize = TRUE
)
print(res)
plot(res, colors = c("turquoise", "green", "violet", "red", "black", "blue"))
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.