sporco.dictlrn.onlinecdl

Online dictionary learning based on CBPDN sparse coding

Classes

OnlineConvBPDNDictLearn(*args, **kwargs)	Stochastic gradient descent (SGD) based online convolutional dictionary learning, as proposed in [33].
OnlineConvBPDNMaskDictLearn(*args, **kwargs)	Stochastic gradient descent (SGD) based online convolutional dictionary learning with a spatial mask, as proposed in [33].

Class Descriptions

class sporco.dictlrn.onlinecdl.OnlineConvBPDNDictLearn(*args, **kwargs)

Bases: sporco.common.IterativeSolver

Stochastic gradient descent (SGD) based online convolutional dictionary learning, as proposed in [33].

Parameters

D0array_like

Initial dictionary array

lmbdafloat

Regularisation parameter

optOnlineConvBPDNDictLearn.Options object

Algorithm options

dimK0, 1, or None, optional (default None)

Number of signal dimensions in signal array passed to solve. If there will only be a single input signal (e.g. if S is a 2D array representing a single image) dimK must be set to 0.

dimNint, optional (default 2)

Number of spatial/temporal dimensions

class Options(opt=None)

Bases: sporco.dictlrn.dictlrn.DictLearn.Options

Online CBPDN dictionary learning algorithm options.

Options:

Verbose : Flag determining whether iteration status is displayed.

StatusHeader : Flag determining whether status header and separator are displayed.

IterTimer : Label of the timer to use for iteration times.

DictSize : Dictionary size vector.

DataType : Specify data type for solution variables, e.g. np.float32.

ZeroMean : Flag indicating whether the solution dictionary \(\{\mathbf{d}_m\}\) should have zero-mean components.

eta_a, eta_b : Constants \(a\) and \(b\) used in setting the SGD step size, \(\eta\), which is set to \(a / (b + i)\) where \(i\) is the iteration index. See Sec. 3 (pg. 9) of [33].

CUDA_CBPDN : Flag indicating whether to use CUDA solver for CBPDN problem (see sporco.cuda package)

CBPDN : Options admm.cbpdn.ConvBPDN.Options.

Parameters

optdict or None, optional (default None)

OnlineConvBPDNDictLearn algorithm options

fwiter = 4

Field width for iteration count display column

fpothr = 2

Field precision for other display columns

itstat_fields_objfn = ('ObjFun', 'DFid', 'RegL1')

Fields in IterationStats associated with the objective function

itstat_fields_alg = ('PrimalRsdl', 'DualRsdl', 'Rho', 'Cnstr', 'DeltaD', 'Eta')

Fields in IterationStats associated with the specific solver algorithm

itstat_fields_extra = ()

Non-standard fields in IterationStats; see itstat_extra

solve(S, dimK=None)

Compute sparse coding and dictionary update for training data S.

init_vars(S, dimK)

Initalise variables required for sparse coding and dictionary update for training data S.

xstep(S, lmbda, dimK)

Solve CSC problem for training data S.

setcoef(Z)

Set coefficient array.

dstep()

Compute dictionary update for training data of preceding xstep.

manage_itstat()

Compute, record, and display iteration statistics.

getdict()

Get final dictionary.

itstat_extra()

Non-standard entries for the iteration stats record tuple.

classmethod hdrtxt()

Construct tuple of status display column title.

classmethod hdrval()

Construct dictionary mapping display column title to IterationStats entries.

iteration_stats()

Construct iteration stats record tuple.

getitstat()

Get iteration stats as named tuple of arrays instead of array of named tuples.

display_config()

Set up status display if option selected. NB: this method assumes that the first entry is the iteration count and the last is the rho value.

display_start()

Start status display if option selected.

display_status(fmtstr, itst)

Display current iteration status as selection of fields from iteration stats tuple.

display_end()

Terminate status display if option selected.

class sporco.dictlrn.onlinecdl.OnlineConvBPDNMaskDictLearn(*args, **kwargs)

Bases: sporco.dictlrn.onlinecdl.OnlineConvBPDNDictLearn

Stochastic gradient descent (SGD) based online convolutional dictionary learning with a spatial mask, as proposed in [33].

Parameters

D0array_like

Initial dictionary array

lmbdafloat

Regularisation parameter

optOnlineConvBPDNDictLearn.Options object

Algorithm options

dimK0, 1, or None, optional (default None)

Number of signal dimensions in signal array passed to solve. If there will only be a single input signal (e.g. if S is a 2D array representing a single image) dimK must be set to 0.

dimNint, optional (default 2)

Number of spatial/temporal dimensions

class Options(opt=None)

Bases: sporco.dictlrn.onlinecdl.OnlineConvBPDNDictLearn.Options

Online masked CBPDN dictionary learning algorithm options.

Options are the same as those of OnlineConvBPDNDictLearn.Options, except for

CBPDN : Options admm.cbpdn.ConvBPDNMaskDcpl.Options.

Parameters

optdict or None, optional (default None)

OnlineConvBPDNMaskDictLearn algorithm options

solve(S, W=None, dimK=None)

Compute sparse coding and dictionary update for training data S.

xstep(S, W, lmbda, dimK)

Solve CSC problem for training data S.

dstep(W)

Compute dictionary update for training data of preceding xstep.