Location:
Search - FGPLVM
Search list
Description: fast Gaussian process latent variable model Software (FGPLVM). This toolbox allows for larger GP-LVM models through using the sparse approximations suggested in papers by authors including Titsias, Snelson, Ghahramani, Seeger, and Lawrence
Platform: |
Size: 13067689 |
Author: 347289298@qq.com |
Hits:
Description: 这是一个用于高斯过程隐变量模型的工具箱，其中包含了MATLAB/C/PYTHON三种语言版本-As of July 2005 a C++ implementation of the GPLVM exists which has most of the flexibility of this software but runs much faster. However as of this time it cannot handle very large data sets as the sparsification algorithm (i.e. the IVM) is not implemented.
Expected code updates in the near future: different kernels associated with each dimension of the IVM code (update to take place in IVM toolbox).
A note on speed issues:
This toolbox is designed to be flexible, not fast. Large improvements in running speed can be obtained by removing that flexibility (e.g. by fixing the kernel type and the noise model etc.).
Version 2.1 Release Notes
-
As part of releasing the new FGPLVM toolbox, which uses a different sparsification approach for dealing with large data sets, large portions of the original GPLVM code were moved into two new toolboxes: the datasets toolbox and the mltools toolbox. These toolboxes are now required for running the GPLVM toolbox.
Version
Platform: |
Size: 1377280 |
Author: 李 |
Hits: