AUXIL\
gpugausskernel_matrix.pro
top gpugausskernel_matrix
result = gpugausskernel_matrix(G1_gpu [, G2_gpu] [, GMA=float] [, NSCALE=float])
Returns array of Gaussian kernel functions for data matrices G1_gpu and G2_gpu. G1_gpu is an N x m array, G2_gpu is an N x n array. If G2_gpu not present then returns a symmetric, positive definite Guassian kernel matrix
Return value
{GPUHANDLE} to n x m array
Parameters
- G1_gpu in required type={GPUHANDLE}
data matrix
- G2_gpu in optional type={GPUHANDLE}
data matrix
Keywords
- GMA in out optional type=float
if not given, calculated from the data: GMA=1/(2*(nscale*scale)^2) where scale = average distance between observations in the feature space
- NSCALE in optional type=float
multiple of scale for GMA (default 1.0)
Author information
- Author:
Mort Canty (2009)
Other attributes
- Uses:
GPULib
File attributes
| Modification date: | Mon Nov 22 21:38:20 2010 |
| Lines: | 74 |
| Docformat: | rst rst |