Description Usage Arguments Value See Also Examples
Construct a new kernel by combining existing kernels, either by summation, multiplication or the kronecker product.
Summation and multiplication require that the covariance matrices produced the two kernels have the same dimension (same number of rows in the input columns) and result in a kernel with the same dimension as its inputs.
The kronecker product doesn't require that the input functions have the same dimension, and the dimension of the output is the product of the dimensions of the inputs (i.e. an m-by-m matrix kroneckered with an n-by-n matrix gives rise to an nm-by-nm matrix).
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 | ## S3 method for class 'kernel'
sum(..., na.rm = FALSE)
## S3 method for class 'kernel'
kernel1 + kernel2
## S3 method for class 'kernel'
prod(..., na.rm = FALSE)
## S3 method for class 'kernel'
kernel1 * kernel2
## S3 method for class 'kernel'
kernel ^ power
kron(kernel1, kernel2)
kernel1 %x% kernel2
|
na.rm |
an unused argument for consistency with the generic sum function |
kernel, kernel1, kernel2 |
kernel objects to be combined |
power |
an integer (or integer-esque numeric) giving the power to which
to raise the kernel function. If |
... |
several kernel objects to be combined |
A kernel object for which there are a range of associated functions, see kernel
and access
for details.
Other kernel.constructors: expo
,
iid
, int
, lin
,
mat32
, mat52
,
per
, rbf
, rq
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 | # construct a kernel with one feature
k1 <- rbf('temperature')
# and another with two features
k2 <- rbf(c('temperature', 'pressure'))
# sum lots of kernels
k_sum <- sum(k1, k2, k1)
# evaluate the function and look at the matrix
image(k_sum(pressure))
# sum two kernels
k_1p2 <- k1 + k2
# evaluate the function and look at the matrix
image(k_1p2(pressure))
# multiply lots of kernels
k_prod <- prod(k1, k2, k1)
# evaluate the function and look at the matrix
image(k_prod(pressure))
# multiply two kernels
k_1_2 <- k1 * k2
# evaluate the function and look at the matrix
image(k_1_2(pressure))
# get a cubic kernel
k <- int() + lin('pressure', c = 400, sigma = 0.003)
k_cu <- k ^ 3
# evaluate the function and look at the matrix
image(k_cu(pressure))
# look at example draws from the original and cubed kernel
demoKernel(k, pressure)
demoKernel(k_cu, pressure)
# get the kronecker product of two kernels
k_kron <- kron(k1, k2)
# evaluate the function and look at the matrix
image(k_kron(pressure))
# get the kronecker product of two kernels again
k_kron <- k1 %x% k2
# evaluate the function and look at the matrix
image(k_kron(pressure))
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.