A distance metric learning algorithm for unsupervised dimensionality reduction, obtaining orthogonal directions that maximize the variance.
PCA(num_dims = NULL, thres = NULL)
| num_dims | Number of components for dimensionality reduction. If NULL, all the principal components will be taken. Ignored if thres is provided. Integer. |
|---|---|
| Fraction | of variability to keep, from 0 to 1. Data dimension will be reduced until the lowest dimension that keeps 'thres' explained variance. Float. |
The PCA transformer, structured as a named list.