- Added :py:class:
pyproximal.optimization.palm.iPALM
solver - Added :py:func:
pyproximal.optimization.palm._backtracking
method to be used whengammaf=None
and/orgammag=None
- Added :py:func:
pyproximal.utils.gradtest.gradtest_proximal
and :py:func:pyproximal.utils.gradtest.gradtest_bilinear
methods - Added
tol
to :py:class:pyproximal.optimization.primal.ProximalPoint
and :py:class:pyproximal.optimization.primal.ProximalGradient
solvers - Modified :py:class:
pyproximal.ProxOperator.precomposition
to allowb
being also a vector
- Added
pyproximal.projection.L01BallProj
andpyproximal.proximal.L01Ball
operators - Added
eta
topyproximal.optimization.primal.ProximalGradient
- Added
eta
andweights
topyproximal.optimization.primal.GeneralizedProximalGradient
- Allow
eta
topyproximal.optimization.primal.ProximalGradient
to have iteration-dependentepsg
- Switched from
lsqr
tocg
inpyproximal.projection.AffineSetProj
- Added
pyproximal.proximal.RelaxedMumfordShah
operator - Added cuda version to the proximal operator of
pyproximal.proximal.Simplex
- Added bilinear update to
pyproximal.optimization.primal.ProximalGradient
- Modified
pyproximal.optimization.pnp.PlugAndPlay
function signature to allow using any proximal solver of choice - Fixed print in
pyproximal.optimization.primaldual.PrimalDual
when using cupy arrays - Fixed
pyproximal.utils.bilinear.LowRankFactorizedMatrix
whenn=m
🚦 🚦 This is the first release supporting PyLops v2. 🚦 🚦
- Added
grad
method topyproximal.utils.bilinear.LowRankFactorizedMatrix
operator - Allow passing optional arguments to solvers in
pyproximal.proximal.L2
- Modified codebase to integrate with pylops's
cupy
backend. - Modified codebase to integrate with
pylops
v2.
🚦 🚦 This is the latest release supporting PyLops v1. 🚦 🚦
- Added
pyproximal.proximal.Log1
operator - Allow
radius
parameter ofpyproximal.proximal.L0
to be a function - Allow
tau
parameter ofpyproximal.optimization.primal.HQS
to be a vector and change over iterations - Added
z0
topyproximal.optimization.primal.HQS
- Added
factorize
option todensesolver
ofpyproximal.proximal.L2
- Added
pyproximal.optimization.primal.ADMML2
,pyproximal.optimization.primal.HQS
, andpyproximal.optimization.pnp.PlugAndPlay
solvers - Added
pyproximal.proximal.ETP
,pyproximal.proximal.Geman
,pyproximal.proximal.L0`,
pyproximal.proximal.Log,
pyproximal.proximal.QuadraticEnvelopeCard,
pyproximal.proximal.SCAD`` operators. - Allow
tau
parameter of proximal operators to be a vector to handle problems with multiple right-hand sides.
- Added
pyproximal.optimization.palm.PALM
optimizer - Added
callback
topyproximal.optimization.proximal.ProximalPoint
optimizer - Added
pyproximal.utils.bilinear.BilinearOperator
andpyproximal.utils.bilinear.LowRankFactorizedMatrix
operators
- Added
pyproximal.proximal.L0Ball
,pyproximal.proximal.L1Ball
,pyproximal.proximal.L21_plus_L1
,pyproximal.proximal.Nuclear
,pyproximal.proximal.NuclearBall
, andpyproximal.proximal.Nonlinear
operators - Added
pyproximal.optimization.primal.TwIST
solver - Added
acceleration
inpyproximal.optimization.primal.AcceleratedProximalGradient
solver - Added classes standard deviation in
pyproximal.optimization.segmentation.Segment
solver - Added
chain
methodpyproximal.ProxOperator
- Fix
pyproximal.proximal.Orthogonal
by introducingalpha
in the proximal evaluation
- Added
pyproximal.optimization.sr3.SR3
solver - Added
pyproximal.projection.AffineSetProj
andpyproximal.AffineSet
operators - Fixed
pyproximal.Huber
operator
- First official release.