The kernel generating functions provided in kernlab.
The Gaussian kernel $k(x,x') = \exp(-\sigma \|x - x'\|^2)$
the Polynomial kernel $k(x,x') = (scale + offset)^degree$.
the Linear kernel $k(x,x') = $
and the Hyperbolic tangent kernel $k(x, x') = \tanh(scale + offset)$
Usage
rbfdot(sigma = 1)
polydot(degree = 1, scale = 1, offset = 1)
tanhdot(scale = 1, offset = 1)
vanilladot()
Arguments
sigma
The inverse kernel width used by the Gaussian kernel
degree
The degree of the polynomial kernel. This has to be an integer.
scale
The scaling parameter is a convenient way of normalizing
patterns without the need to modify the data itself
offset
The offset used in a polynomial or hyperbolic tangent kernel
Value
Return an S4 object of class kernel which extents the
function class. The resulting function implements the given
kernel calculating the inner (dot) product between two vectors.
kpara list containing the kernel parameters (hyperparameters)
used.
the kernel parameters can be accessed by the kpar function.
Details
The kernel generating function are used to initialize a kernel
function
which calculates the dot (inner) product between two feature vectors in a
Hilbert Space. These functions can be based as a kernel argument on almost all
functions in kernlab (eg. ksvm, kpca etc).
Although using one of the above mentioned existing kernel functions as a
kernel argument in various functions in kernlab has the
advantage that use of optimize kernel utilities methods are used
any other function implementing a dot product of class kernel can also be used as a kernel
argument. This allows the user to use test an develop special kernels
for a given data set and algorithm.