Home > lib > data_transform_pcalda.m

data_transform_pcalda

PURPOSE ^

d2 = data_transform_pcalda(data, no_factors = 0, flag_rotate_factors = 1, flag_modified_s_b = 0)

SYNOPSIS ^

function d2 = data_transform_pcalda(data, no_factors, flag_rotate_factors, flag_sphere, flag_modified_s_b)

DESCRIPTION ^

 d2 = data_transform_pcalda(data, no_factors = 0, flag_rotate_factors = 1, flag_modified_s_b = 0)

 Combo PCA->LDA transformation. Scores are no different from applying PCA
 and then LDA through the provided data_transform_pca() and
 data_transform_lda() functions. The difference here will solely be in the
 loadings.

 Inputs
   no_factors: number of PCA-domain features to be kept after PCA loadings
     is calculated. 0 mean all. Obs: if you are rotating factors you'd
     typically restrain to only a few factors.
   flag_modified_s_b: see calculate_scatters()
   flag_sphere: see fisher_ld()

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:
Generated on Thu 18-Feb-2010 12:47:47 by m2html © 2003