PGMDFT¶
PGMDFT
specialises PGM
to solve problems of
the form
\[\mathrm{argmin}_{\mathbf{x}} \;
f(\mathbf{x}) + g(\mathbf{x}) \;\;,\]
where \(f, g\) are convex functions and \(f\) is smooth, and all the updates and gradients are computed in DFT domain.
Classes derived from PGMDFT
should override/define the
same methods and attributes described in the PGM class
documentation.
See pgm.cbpdn.ConvBPDN
as an example of a class derived
from PGMDFT
.