kpca
fun kpca(data: DataFrame, kernel: MercerKernel<DoubleArray>, k: Int, threshold: Double = 1.0E-4, vararg columns: String): KernelPCA(source)
Kernel principal component analysis on a data frame.
Return
the KernelPCA model.
Parameters
data
training data.
kernel
Mercer kernel to compute kernel matrix.
k
choose top k principal components used for projection.
threshold
only principal components with eigenvalues larger than the threshold will be kept.
columns
the column names to use. If empty, all columns are used.