snlpca
(Package: onlinePCA) :
Subspace Network Learning PCA
Online PCA with the SNL algorithm of Oja (1992).
● Data Source:
CranContrib
● Keywords:
● Alias: snlpca
●
0 images
|
Online PCA algorithms using perturbation methods (perturbationRpca), secular equations (secularRpca), incremental PCA (incRpca, incRpca.block, incRpca.rc), and stochastic optimization (bsoipca,
ccipca, ghapca, sgapca, snlpca). impute handles missing data with the regression approach of Brand (2002). batchpca performs fast batch (offline) PCA using iterative methods. create.basis, coef2fd, fd2coef respectively create B-spline basis sets for functional data (FD), convert FD to basis coefficients, and convert basis coefficients back to FD. updateMean and updateCovariance update the sample mean and sample covariance.
● Data Source:
CranContrib
● Keywords: package
● Alias: onlinePCA-package
●
0 images
|
|
This function recursively updates the PCA with respect to a single new data vector, using the (fast) perturbation method of Hegde et al. (2006).
● Data Source:
CranContrib
● Keywords:
● Alias: perturbationRpca
●
0 images
|
|
Recursive update of the sample mean vector.
● Data Source:
CranContrib
● Keywords:
● Alias: updateMean
●
0 images
|
incRpca.rc
(Package: onlinePCA) :
Incremental PCA With Reduced Complexity
The incremental PCA is computed without rotating the updated projection space (Brand, 2002; Arora et al., 2012). Specifically, PCs are specified through a matrix of orthogonal vectors Ut that spans the PC space and a rotation matrix Us such that the PC matrix is UtUs. Given a new data vector, the PCA is updated by adding one column to Ut and recalculating the low-dimensional rotation matrix Us. This reduces complexity and helps preserving orthogonality. Eigenvalues are updated as the usual incremental PCA algorithm.
● Data Source:
CranContrib
● Keywords:
● Alias: incRpca.rc
●
0 images
|
|
Online PCA using the incremental SVD method of Brand (2002) and Arora et al. (2012).
● Data Source:
CranContrib
● Keywords:
● Alias: incRpca
●
0 images
|
|
This function computes functional data from their coefficients in a B-spline basis.
● Data Source:
CranContrib
● Keywords:
● Alias: coef2fd
●
0 images
|
ccipca
(Package: onlinePCA) :
Candid Covariance-Free Incremental PCA
Stochastic gradient ascent algorithm CCIPCA of Weng et al. (2003).
● Data Source:
CranContrib
● Keywords:
● Alias: ccipca
●
0 images
|
|
This function recursively updates a covariance matrix without entirely recomputing it when new observations arrive.
● Data Source:
CranContrib
● Keywords:
● Alias: updateCovariance
●
0 images
|
ghapca
(Package: onlinePCA) :
Generalized Hebbian Algorithm for PCA
Online PCA with the GHA of Sanger (1989).
● Data Source:
CranContrib
● Keywords:
● Alias: ghapca
●
0 images
|
providing 
.