The inputs of the ringnorm problem are points from two Gaussian
distributions. Class 1 is multivariate normal with mean 0 and
covariance 4 times the identity matrix. Class 2 has unit covariance
and mean \((a,a,\ldots,a)\), \(a=d^{-0.5}\).
Usage
mlbench.ringnorm(n, d=20)
Value
Returns an object of class "mlbench.ringnorm" with components
x
input values
classes
factor vector of length n with target classes
Arguments
n
number of patterns to create
d
dimension of the ringnorm problem
References
Breiman, L. (1996). Bias, variance, and arcing classifiers.
Tech. Rep. 460, Statistics Department, University of California,
Berkeley, CA, USA.