sim_circle: Simulate Data from Hyper-Sphere for Classification Problems
Description
The classification problem Circle is described in the BARK paper (2008).
Inputs are dim independent variables uniformly
distributed on the interval \([-1,1]\), only the first 2
out of these dim are actually signals.
Outputs are created according to the formula
$$y = 1(x1^2+x2^2 \le 2/\pi)$$
Usage
sim_circle(n, dim = 5)
Value
Returns a list with components
x
input values (independent variables)
y
0/1 output values (dependent variable)
Arguments
n
number of data points to generate
dim
number of dimension of the problem, no less than 2