pcomp

class pydl.pcomp(x, standardize=False, covariance=False)[source]

Bases: object

Replicates the IDL PCOMP() function.

The attributes of this class are all read-only properties, implemented with lazyproperty.

Parameters:

x : array-like

A 2-D array with \(N\) rows and \(M\) columns.

standardize : bool, optional

If set to True, the input data will have its mean subtracted off and will be scaled to unit variance.

covariance : bool, optional.

If set to True, the covariance matrix of the data will be used for the computation. Otherwise the correlation matrix will be used.

References

http://www.exelisvis.com/docs/PCOMP.html

Attributes Summary

coefficients (ndarray) The principal components.
derived (ndarray) The derived variables.
eigenvalues (ndarray) The eigenvalues.
variance (ndarray) The variances of each derived variable.

Attributes Documentation

coefficients

(ndarray) The principal components. These are the coefficients of derived. Basically, they are a re-scaling of the eigenvectors.

derived

(ndarray) The derived variables.

eigenvalues

(ndarray) The eigenvalues. There is one eigenvalue for each principal component.

variance

(ndarray) The variances of each derived variable.