The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
This package includes source code and documentation of gramEvol: Grammatical Evolution for R.
gramEvol implements grammatical evolution (GE) in native R syntax. It allows discovering programs that can achieve a desired goal, by performing an evolutionary optimization over a population of R expressions generated via a user-defined grammar. Functions are provided for creating and manipulating context-free grammars (CFGs), random search, exhaustive search, and evolutionary optimization. Users are only required to define their program structure via a grammar, and a cost function to evaluate the fitness of each program.
You can install this package from CRAN:
install.packages("gramEvol")You can install the latest version from Github:
if (!require("devtools")) install.packages("devtools")
devtools::install_github("fnoorian/gramEvol")A tutorial on implementing GE programs is included in the package’s vignette (PDF version).
More information regarding GE and its application in parameter optimization can be found in this paper in the Journal of Statistical Software.
This example implements the Kepler law rediscovery problem, as discussed in section 3.1 of the vignette.
library("gramEvol")
# grammar definition for generic symbolic regression
grammarDef <- CreateGrammar(list(
expr = grule(op(expr, expr), func(expr), var),
func = grule(sin, cos, log, sqrt),
op = grule(`+`, `-`, `*`), # define unary operators
var = grule(distance, distance^n, n),
n = gvrule(1:4) # this is shorthand for grule(1,2,3,4)
))
# cost function and data
planets <- c("Venus", "Earth", "Mars", "Jupiter", "Saturn", "Uranus")
distance <- c(0.72, 1.00, 1.52, 5.20, 9.53, 19.10)
period <- c(0.61, 1.00, 1.84, 11.90, 29.40, 83.50)
SymRegCostFunc <- function(expr) {
result <- eval(expr)
if (any(is.nan(result)))
return(Inf)
return (mean(log(1 + abs(period - result))))
}
# run GE
ge <- GrammaticalEvolution(grammarDef, SymRegCostFunc, iterations = 50)
print(ge)
# use the best expression
best.expression <- ge$best$expression
print(ge$best$expressions)
print(data.frame(distance, period, Kepler = sqrt(distance^3), GE = eval(best.expression)))The latest release and developmental versions of this package are available on: https://github.com/fnoorian/gramEvol
All files in this package, including the documentation and vignettes, are distributed under GNU GPL v2.0 or later license. For full terms of this license visit https://www.gnu.org/licenses/gpl-2.0.html.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.