AUXIL\
gpukernelproject.pro
top gpukernelProject
result = gpukernelProject(alpha, gma, G, GG, num_cols, num_rows, image [, kernel=integer] [, center_train=center_train] [, nscale=float])
Centers and projects test data for kernel PCA or kernel MAF (linear or gaussian):
Parameters
- alpha in required
normalized eigenvectors of kernel matrix
- gma in required
radial basis kernel parameter
- G in required
training data matrix
- GG in required
test data matrix to be projected
- num_cols in required
image dimensions
- num_rows
- image out
projected data as BSQ image
Keywords
- kernel in optional type=integer
the kernel used 0: linear 1: Gaussian (default)
- center_train in optional
if set, center on training means (default 0)
- nscale in optional type=float
multiple of scale for GMA when KERNEL=1 (default 1.0)
Author information
- Author:
Mort Canty (2009)
Other attributes
- Uses:
ENVI:, ,
, CENTER:, ,, GPUKERNEL_MATRIX:COYOTE:GPULIB
File attributes
Modification date: | Fri May 17 08:32:29 2013 |
Lines: | 233 |
Docformat: | rst rst |