kernDiagGradX {tigre} | R Documentation |
computes the gradient of the (diagonal of the) kernel matrix with respect to the elements of the design matrix given in X.
kernDiagGradX(kern, x) kernGradX(kern, x, x2)
kern |
the kernel structure for which gradients are being computed. |
x |
if only argument: the input data in the form of a design matrix, if two arguments: row locations against which gradients are being computed. |
x2 |
(optional) column locations against which gradients are being computed. |
gX |
the gradients of the diagonal with respect to each element of X. The returned matrix has the same dimensions as X. |
gX2 |
the returned gradients. The gradients are returned in a matrix which is numData x numInputs x numData. Where numData is the number of data points and numInputs is the number of input dimensions in X. |
kern <- kernCreate(1, 'mlp') g <- kernDiagGradX(kern, as.matrix(3:8))