Class Hierarchy

object: The most base type
- parsimony.utils.utils.AnonymousClass: Used to create anonymous classes.
- parsimony.algorithms.bases.BaseAlgorithm
  - parsimony.algorithms.bases.ExplicitAlgorithm: Explicit algorithms are algorithms that minimises a given function.
    - parsimony.algorithms.proximal.ADMM: The alternating direction method of multipliers (ADMM).
    - parsimony.algorithms.utils.BacktrackingLineSearch: Finds a step length a that fulfills a given descent criterion.
    - parsimony.algorithms.utils.Bisection: Finds a root of the function assumed to be on the line between two points.
    - parsimony.algorithms.proximal.CONESTA: COntinuation with NEsterov smoothing in a Soft-Thresholding Algorithm, or CONESTA for short.
    - parsimony.algorithms.proximal.DykstrasProjectionAlgorithm: Dykstra's projection algorithm.
    - parsimony.algorithms.proximal.DykstrasProximalAlgorithm: Dykstra's proximal algorithm.
    - parsimony.algorithms.primaldual.ExcessiveGapMethod: Nesterov's excessive gap method for strongly convex functions.
    - parsimony.algorithms.proximal.FISTA: The fast iterative shrinkage-thresholding algorithm.
    - parsimony.algorithms.gradient.GradientDescent: The gradient descent algorithm.
    - parsimony.algorithms.proximal.ISTA: The iterative shrinkage-thresholding algorithm.
    - parsimony.algorithms.multiblock.MultiblockCONESTA: An alternating minimising multiblock algorithm that utilises CONESTA in the inner minimisation.
    - parsimony.algorithms.multiblock.MultiblockFISTA: The projected or proximal gradient algorithm with alternating minimisations in a multiblock setting.
    - parsimony.algorithms.utils.NewtonRaphson: Finds a root of the function assumed to be in the vicinity of a given point.
    - parsimony.algorithms.proximal.ParallelDykstrasProjectionAlgorithm: Dykstra's projection algorithm for two or more functions.
    - parsimony.algorithms.proximal.ParallelDykstrasProximalAlgorithm: Dykstra's projection algorithm for two or more functions.
    - parsimony.algorithms.proximal.StaticCONESTA: COntinuation with NEsterov smoothing in a Soft-Thresholding Algorithm, or CONESTA for short, with a statically decreasing sequence of eps and mu.
  - parsimony.algorithms.bases.ImplicitAlgorithm: Implicit algorithms are algorithms that do not utilise a loss function.
    - parsimony.algorithms.nipals.FastSVD
    - parsimony.algorithms.nipals.FastSVDProduct
    - parsimony.algorithms.nipals.FastSparseSVD
    - parsimony.algorithms.cluster.KMeans: The K-means clustering algorithm.
    - parsimony.algorithms.coordinate.LassoCoordinateDescent: A coordinate descent algorithm for the lasso.
    - parsimony.algorithms.nipals.PLSR: A NIPALS implementation for PLS regresison.
    - parsimony.algorithms.coordinate.ShootingAlgorithm: The shooting algorithm for the lasso.
    - parsimony.algorithms.nipals.SparsePLSR: A NIPALS implementation for Sparse PLS regresison.
- parsimony.estimators.BaseEstimator: Base class for estimators.
  - parsimony.estimators.Clustering: Estimator for the clustering problem, i.e.
  - parsimony.estimators.LogisticRegressionEstimator: Base estimator for logistic regression estimation
    - parsimony.estimators.ElasticNetLogisticRegression: Logistic regression (re-weighted log-likelihood aka.
    - parsimony.estimators.LogisticRegression: Logistic regression (re-weighted log-likelihood aka.
    - parsimony.estimators.LogisticRegressionL1L2GL: Logistic regression (re-weighted log-likelihood aka.
    - parsimony.estimators.LogisticRegressionL1L2TV: Logistic regression (re-weighted log-likelihood aka.
    - parsimony.estimators.RidgeLogisticRegression: Logistic regression (re-weighted log-likelihood aka.
  - parsimony.estimators.RegressionEstimator: Base estimator for regression estimation.
    - parsimony.estimators.ElasticNet: Linear regression with L1 and L2 penalties.
    - parsimony.estimators.Lasso: Linear regression with an L1 penalty:
    - parsimony.estimators.LinearRegression: Linear regression:
    - parsimony.estimators.LinearRegressionL1L2GL: Linear regression with L1, L2 and Group lasso penalties:
    - parsimony.estimators.LinearRegressionL1L2TV: Linear regression with L1, L2 and TV penalties:
    - parsimony.estimators.LinearRegressionL2SmoothedL1TV: Linear regression with L2 and simultaneously smoothed L1 and TV penalties:
    - parsimony.estimators.PLSRegression: Estimator for PLS regression
    - parsimony.estimators.RidgeRegression: Linear regression with an L2 penalty.
    - parsimony.estimators.SparsePLSRegression: Estimator for sparse PLS regression
- parsimony.utils.start_vectors.BaseStartVector: Base class for start vector generation.
  - parsimony.utils.start_vectors.IdentityStartVector: A pre-determined start vector.
  - parsimony.utils.start_vectors.OnesStartVector: A start vector of ones.
  - parsimony.utils.start_vectors.RandomStartVector: A start vector of uniformly distributed random values.
  - parsimony.utils.start_vectors.ZerosStartVector: A start vector of zeros.
- parsimony.datasets.simulate.utils.ConstantValue: Random-like number generator that returns a constant value.
- parsimony.functions.properties.Constraint: Represents a constraint of a function.
  - parsimony.functions.nesterov.gl.GroupLassoOverlap: Group L1-L2 function, with overlapping groups.
  - parsimony.functions.nesterov.grouptv.GroupTotalVariation: The smoothed Group total variation (Group TV) function
  - parsimony.functions.penalties.L0: The proximal operator of the "pseudo" L0 function
  - parsimony.functions.nesterov.l1.L1: The proximal operator of the smoothed L1 function
  - parsimony.functions.penalties.L1: The proximal operator of the L1 function with a penalty formulation
  - parsimony.functions.penalties.L2: The proximal operator of the L2 function with a penalty formulation
  - parsimony.functions.penalties.L2Squared: The proximal operator of the squared L2 function with a penalty formulation
  - parsimony.functions.penalties.LInf: The proximal operator of the L-infinity function
  - parsimony.functions.penalties.LinearVariableConstraint: Represents a linear constraint
  - parsimony.functions.penalties.QuadraticConstraint: The proximal operator of the quadratic function
    - parsimony.functions.penalties.RGCCAConstraint: Represents the quadratic function
  - parsimony.functions.penalties.SufficientDescentCondition
  - parsimony.functions.nesterov.tv.TotalVariation: The smoothed Total variation (TV) function
  - parsimony.functions.penalties.ZeroFunction
- parsimony.functions.properties.Continuation
  - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
  - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
  - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
- parsimony.algorithms.deflation.Deflation
  - parsimony.algorithms.deflation.ColumnProjectionDeflation
  - parsimony.algorithms.deflation.RankOneDeflation
  - parsimony.algorithms.deflation.RowProjectionDeflation
- parsimony.functions.properties.DualFunction
  - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
  - parsimony.functions.combinedfunctions.LinearRegressionL2SmoothedL1TV: Combination (sum) of Linear Regression, L2 and simultaneously smoothed L1 and TotalVariation.
  - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
- parsimony.functions.properties.Eigenvalues
  - parsimony.functions.nesterov.l1tv.L1TV: The proximal operator of the smoothed sum of the TV and L1 functions
  - parsimony.functions.properties.NesterovFunction: Abstract superclass of Nesterov functions.
    - parsimony.functions.nesterov.gl.GroupLassoOverlap: Group L1-L2 function, with overlapping groups.
    - parsimony.functions.nesterov.grouptv.GroupTotalVariation: The smoothed Group total variation (Group TV) function
    - parsimony.functions.nesterov.l1.L1: The proximal operator of the smoothed L1 function
    - parsimony.functions.nesterov.l1tv.L1TV: The proximal operator of the smoothed sum of the TV and L1 functions
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
      - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
        
        parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL2SmoothedL1TV: Combination (sum) of Linear Regression, L2 and simultaneously smoothed L1 and TotalVariation.
    - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
    - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
    - parsimony.functions.nesterov.tv.TotalVariation: The smoothed Total variation (TV) function
- parsimony.functions.properties.Function
  - parsimony.functions.properties.AtomicFunction: This is a function that is not in general supposed to be minimised by itself.
    - parsimony.functions.penalties.L0: The proximal operator of the "pseudo" L0 function
    - parsimony.functions.penalties.L1: The proximal operator of the L1 function with a penalty formulation
    - parsimony.functions.penalties.L1L2Squared: The proximal operator of the L1 function with an L2 constraint.
    - parsimony.functions.penalties.L2: The proximal operator of the L2 function with a penalty formulation
    - parsimony.functions.penalties.L2Squared: The proximal operator of the squared L2 function with a penalty formulation
    - parsimony.functions.penalties.LInf: The proximal operator of the L-infinity function
    - parsimony.functions.losses.LogisticRegression: The Logistic Regression loss function.
    - parsimony.functions.penalties.QuadraticConstraint: The proximal operator of the quadratic function
      - parsimony.functions.penalties.RGCCAConstraint: Represents the quadratic function
    - parsimony.functions.penalties.ZeroFunction
  - parsimony.functions.properties.CombinedProjectionOperator
  - parsimony.functions.properties.CompositeFunction: This is a function that is the combination (e.g.
    - parsimony.functions.combinedfunctions.CombinedFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
    - parsimony.functions.losses.LinearFunction: A linear function.
    - parsimony.functions.losses.LinearRegression: The Linear regression loss function.
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
      - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
        
        parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL2SmoothedL1TV: Combination (sum) of Linear Regression, L2 and simultaneously smoothed L1 and TotalVariation.
    - parsimony.functions.multiblock.properties.MultiblockFunction: This is a function that is the combination (i.e.
      - parsimony.functions.multiblock.losses.CombinedMultiblockFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
      - parsimony.functions.multiblock.losses.GeneralisedMultiblock
      - parsimony.functions.multiblock.losses.LatentVariableCovariance: Represents
      - parsimony.functions.multiblock.losses.LatentVariableCovarianceSquared: Represents
    - parsimony.functions.multiblock.losses.MultiblockFunctionWrapper
      - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
    - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
    - parsimony.functions.losses.RidgeLogisticRegression: The Logistic Regression loss function with a squared L2 penalty.
    - parsimony.functions.losses.RidgeRegression: The Ridge Regression function, i.e.
    - parsimony.functions.penalties.RidgeSquaredError: Represents a ridge squared error penalty, i.e.
  - parsimony.functions.properties.IndicatorFunction: Represents an indicator function.
    - parsimony.functions.penalties.LinearVariableConstraint: Represents a linear constraint
  - parsimony.functions.losses.LatentVariableVariance
  - parsimony.functions.properties.SplittableFunction: Represents a function that is the sum of two other functions such that
    - parsimony.functions.combinedfunctions.AugmentedLinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and 1D TotalVariation with a linear constraint.
  - parsimony.functions.penalties.SufficientDescentCondition
- parsimony.datasets.simulate.grad.Function
  - parsimony.datasets.simulate.grad.GroupLasso
  - parsimony.datasets.simulate.grad.L1
  - parsimony.datasets.simulate.grad.L2
  - parsimony.datasets.simulate.grad.L2Squared
  - parsimony.datasets.simulate.grad.NesterovFunction
  - parsimony.datasets.simulate.grad.SmoothedL1
  - parsimony.datasets.simulate.grad.TotalVariation
- parsimony.functions.properties.Gradient
  - parsimony.functions.combinedfunctions.CombinedFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
  - parsimony.functions.penalties.L2Squared: The proximal operator of the squared L2 function with a penalty formulation
  - parsimony.functions.losses.LatentVariableVariance
  - parsimony.functions.losses.LinearFunction: A linear function.
  - parsimony.functions.losses.LinearRegression: The Linear regression loss function.
  - parsimony.functions.losses.LogisticRegression: The Logistic Regression loss function.
  - parsimony.functions.multiblock.losses.MultiblockFunctionWrapper
    - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
  - parsimony.functions.properties.NesterovFunction: Abstract superclass of Nesterov functions.
    - parsimony.functions.nesterov.gl.GroupLassoOverlap: Group L1-L2 function, with overlapping groups.
    - parsimony.functions.nesterov.grouptv.GroupTotalVariation: The smoothed Group total variation (Group TV) function
    - parsimony.functions.nesterov.l1.L1: The proximal operator of the smoothed L1 function
    - parsimony.functions.nesterov.l1tv.L1TV: The proximal operator of the smoothed sum of the TV and L1 functions
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
      - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
        
        parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL2SmoothedL1TV: Combination (sum) of Linear Regression, L2 and simultaneously smoothed L1 and TotalVariation.
    - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
    - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
    - parsimony.functions.nesterov.tv.TotalVariation: The smoothed Total variation (TV) function
  - parsimony.functions.penalties.QuadraticConstraint: The proximal operator of the quadratic function
    - parsimony.functions.penalties.RGCCAConstraint: Represents the quadratic function
  - parsimony.functions.losses.RidgeLogisticRegression: The Logistic Regression loss function with a squared L2 penalty.
  - parsimony.functions.losses.RidgeRegression: The Ridge Regression function, i.e.
  - parsimony.functions.penalties.RidgeSquaredError: Represents a ridge squared error penalty, i.e.
  - parsimony.functions.penalties.ZeroFunction
- parsimony.functions.properties.GradientMap
  - parsimony.functions.combinedfunctions.LinearRegressionL2SmoothedL1TV: Combination (sum) of Linear Regression, L2 and simultaneously smoothed L1 and TotalVariation.
- parsimony.functions.properties.Hessian
- parsimony.algorithms.cluster.KMeans.INIT
- parsimony.algorithms.utils.Info: Enum-like class for information constants.
- parsimony.algorithms.bases.InformationAlgorithm: Algorithms that produce information about their run.
  - parsimony.algorithms.proximal.ADMM: The alternating direction method of multipliers (ADMM).
  - parsimony.algorithms.utils.Bisection: Finds a root of the function assumed to be on the line between two points.
  - parsimony.algorithms.proximal.CONESTA: COntinuation with NEsterov smoothing in a Soft-Thresholding Algorithm, or CONESTA for short.
  - parsimony.algorithms.primaldual.ExcessiveGapMethod: Nesterov's excessive gap method for strongly convex functions.
  - parsimony.algorithms.proximal.FISTA: The fast iterative shrinkage-thresholding algorithm.
  - parsimony.algorithms.gradient.GradientDescent: The gradient descent algorithm.
  - parsimony.algorithms.proximal.ISTA: The iterative shrinkage-thresholding algorithm.
  - parsimony.algorithms.coordinate.LassoCoordinateDescent: A coordinate descent algorithm for the lasso.
  - parsimony.algorithms.multiblock.MultiblockCONESTA: An alternating minimising multiblock algorithm that utilises CONESTA in the inner minimisation.
  - parsimony.algorithms.multiblock.MultiblockFISTA: The projected or proximal gradient algorithm with alternating minimisations in a multiblock setting.
  - parsimony.algorithms.utils.NewtonRaphson: Finds a root of the function assumed to be in the vicinity of a given point.
  - parsimony.algorithms.coordinate.ShootingAlgorithm: The shooting algorithm for the lasso.
  - parsimony.algorithms.proximal.StaticCONESTA: COntinuation with NEsterov smoothing in a Soft-Thresholding Algorithm, or CONESTA for short, with a statically decreasing sequence of eps and mu.
- parsimony.algorithms.bases.IterativeAlgorithm: Algorithms that require iterative steps to achieve the goal.
  - parsimony.algorithms.proximal.ADMM: The alternating direction method of multipliers (ADMM).
  - parsimony.algorithms.utils.Bisection: Finds a root of the function assumed to be on the line between two points.
  - parsimony.algorithms.proximal.CONESTA: COntinuation with NEsterov smoothing in a Soft-Thresholding Algorithm, or CONESTA for short.
  - parsimony.algorithms.primaldual.ExcessiveGapMethod: Nesterov's excessive gap method for strongly convex functions.
  - parsimony.algorithms.proximal.FISTA: The fast iterative shrinkage-thresholding algorithm.
  - parsimony.algorithms.gradient.GradientDescent: The gradient descent algorithm.
  - parsimony.algorithms.proximal.ISTA: The iterative shrinkage-thresholding algorithm.
  - parsimony.algorithms.coordinate.LassoCoordinateDescent: A coordinate descent algorithm for the lasso.
  - parsimony.algorithms.multiblock.MultiblockCONESTA: An alternating minimising multiblock algorithm that utilises CONESTA in the inner minimisation.
  - parsimony.algorithms.multiblock.MultiblockFISTA: The projected or proximal gradient algorithm with alternating minimisations in a multiblock setting.
  - parsimony.algorithms.utils.NewtonRaphson: Finds a root of the function assumed to be in the vicinity of a given point.
  - parsimony.algorithms.nipals.PLSR: A NIPALS implementation for PLS regresison.
  - parsimony.algorithms.coordinate.ShootingAlgorithm: The shooting algorithm for the lasso.
  - parsimony.algorithms.nipals.SparsePLSR: A NIPALS implementation for Sparse PLS regresison.
  - parsimony.algorithms.proximal.StaticCONESTA: COntinuation with NEsterov smoothing in a Soft-Thresholding Algorithm, or CONESTA for short, with a statically decreasing sequence of eps and mu.
- parsimony.functions.properties.LipschitzContinuousGradient
  - parsimony.functions.penalties.L2Squared: The proximal operator of the squared L2 function with a penalty formulation
  - parsimony.functions.losses.LatentVariableVariance
  - parsimony.functions.losses.LinearFunction: A linear function.
  - parsimony.functions.losses.LinearRegression: The Linear regression loss function.
  - parsimony.functions.losses.LogisticRegression: The Logistic Regression loss function.
  - parsimony.functions.properties.NesterovFunction: Abstract superclass of Nesterov functions.
    - parsimony.functions.nesterov.gl.GroupLassoOverlap: Group L1-L2 function, with overlapping groups.
    - parsimony.functions.nesterov.grouptv.GroupTotalVariation: The smoothed Group total variation (Group TV) function
    - parsimony.functions.nesterov.l1.L1: The proximal operator of the smoothed L1 function
    - parsimony.functions.nesterov.l1tv.L1TV: The proximal operator of the smoothed sum of the TV and L1 functions
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
      - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
        
        parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL2SmoothedL1TV: Combination (sum) of Linear Regression, L2 and simultaneously smoothed L1 and TotalVariation.
    - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
    - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
    - parsimony.functions.nesterov.tv.TotalVariation: The smoothed Total variation (TV) function
  - parsimony.functions.losses.RidgeLogisticRegression: The Logistic Regression loss function with a squared L2 penalty.
  - parsimony.functions.losses.RidgeRegression: The Ridge Regression function, i.e.
- parsimony.functions.multiblock.properties.MultiblockContinuation
- parsimony.functions.multiblock.properties.MultiblockGradient
  - parsimony.functions.multiblock.losses.CombinedMultiblockFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
  - parsimony.functions.multiblock.losses.GeneralisedMultiblock
  - parsimony.functions.multiblock.losses.LatentVariableCovariance: Represents
  - parsimony.functions.multiblock.losses.LatentVariableCovarianceSquared: Represents
- parsimony.functions.multiblock.properties.MultiblockLipschitzContinuousGradient
  - parsimony.functions.multiblock.losses.LatentVariableCovariance: Represents
  - parsimony.functions.multiblock.losses.LatentVariableCovarianceSquared: Represents
- parsimony.functions.multiblock.properties.MultiblockProjectionOperator
  - parsimony.functions.multiblock.losses.CombinedMultiblockFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
  - parsimony.functions.multiblock.losses.GeneralisedMultiblock
- parsimony.functions.multiblock.properties.MultiblockProximalOperator
  - parsimony.functions.multiblock.losses.CombinedMultiblockFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
- parsimony.functions.multiblock.properties.MultiblockStepSize
  - parsimony.functions.multiblock.losses.CombinedMultiblockFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
- parsimony.utils.linalgs.MultipartArray
- parsimony.functions.properties.OR
- parsimony.datasets.utils.ObjImage: Parameters: ----------- mask: ???
  - parsimony.datasets.utils.Dimaond
  - parsimony.datasets.utils.Dot
  - parsimony.datasets.utils.Square
- parsimony.functions.properties.Penalty: Represents the penalisation of a function.
  - parsimony.functions.nesterov.gl.GroupLassoOverlap: Group L1-L2 function, with overlapping groups.
  - parsimony.functions.nesterov.grouptv.GroupTotalVariation: The smoothed Group total variation (Group TV) function
  - parsimony.functions.penalties.L0: The proximal operator of the "pseudo" L0 function
  - parsimony.functions.nesterov.l1.L1: The proximal operator of the smoothed L1 function
  - parsimony.functions.penalties.L1: The proximal operator of the L1 function with a penalty formulation
  - parsimony.functions.penalties.L1L2Squared: The proximal operator of the L1 function with an L2 constraint.
  - parsimony.functions.nesterov.l1tv.L1TV: The proximal operator of the smoothed sum of the TV and L1 functions
  - parsimony.functions.penalties.L2: The proximal operator of the L2 function with a penalty formulation
  - parsimony.functions.penalties.L2Squared: The proximal operator of the squared L2 function with a penalty formulation
  - parsimony.functions.penalties.LInf: The proximal operator of the L-infinity function
  - parsimony.functions.penalties.QuadraticConstraint: The proximal operator of the quadratic function
    - parsimony.functions.penalties.RGCCAConstraint: Represents the quadratic function
  - parsimony.functions.penalties.RidgeSquaredError: Represents a ridge squared error penalty, i.e.
  - parsimony.functions.nesterov.tv.TotalVariation: The smoothed Total variation (TV) function
  - parsimony.functions.penalties.ZeroFunction
- parsimony.functions.properties.ProjectionOperator
  - parsimony.functions.combinedfunctions.CombinedFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
  - parsimony.functions.properties.CombinedProjectionOperator
  - parsimony.functions.penalties.L0: The proximal operator of the "pseudo" L0 function
  - parsimony.functions.penalties.L1: The proximal operator of the L1 function with a penalty formulation
  - parsimony.functions.penalties.L2: The proximal operator of the L2 function with a penalty formulation
  - parsimony.functions.penalties.L2Squared: The proximal operator of the squared L2 function with a penalty formulation
  - parsimony.functions.penalties.LInf: The proximal operator of the L-infinity function
  - parsimony.functions.penalties.LinearVariableConstraint: Represents a linear constraint
  - parsimony.functions.penalties.RGCCAConstraint: Represents the quadratic function
  - parsimony.functions.penalties.ZeroFunction
- parsimony.functions.properties.ProximalOperator
  - parsimony.functions.properties.AugmentedProximalOperator: Given the problem
    - parsimony.functions.combinedfunctions.AugmentedLinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and 1D TotalVariation with a linear constraint.
  - parsimony.functions.combinedfunctions.CombinedFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
  - parsimony.functions.penalties.L0: The proximal operator of the "pseudo" L0 function
  - parsimony.functions.penalties.L1: The proximal operator of the L1 function with a penalty formulation
  - parsimony.functions.penalties.L1L2Squared: The proximal operator of the L1 function with an L2 constraint.
  - parsimony.functions.penalties.L2: The proximal operator of the L2 function with a penalty formulation
  - parsimony.functions.penalties.L2Squared: The proximal operator of the squared L2 function with a penalty formulation
  - parsimony.functions.penalties.LInf: The proximal operator of the L-infinity function
  - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
  - parsimony.functions.multiblock.losses.MultiblockFunctionWrapper
    - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
  - parsimony.functions.properties.NesterovFunction: Abstract superclass of Nesterov functions.
    - parsimony.functions.nesterov.gl.GroupLassoOverlap: Group L1-L2 function, with overlapping groups.
    - parsimony.functions.nesterov.grouptv.GroupTotalVariation: The smoothed Group total variation (Group TV) function
    - parsimony.functions.nesterov.l1.L1: The proximal operator of the smoothed L1 function
    - parsimony.functions.nesterov.l1tv.L1TV: The proximal operator of the smoothed sum of the TV and L1 functions
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
      - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
        
        parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL2SmoothedL1TV: Combination (sum) of Linear Regression, L2 and simultaneously smoothed L1 and TotalVariation.
    - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
    - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
    - parsimony.functions.nesterov.tv.TotalVariation: The smoothed Total variation (TV) function
  - parsimony.functions.penalties.RidgeSquaredError: Represents a ridge squared error penalty, i.e.
  - parsimony.functions.penalties.ZeroFunction
- parsimony.datasets.simulate.utils.RandomUniform: Example...
- parsimony.datasets.simulate.simulate.SimulatedData
  - parsimony.datasets.simulate.simulate.LinearRegressionData: Returns data for linear regression that is generated such that we know the exact solution.
- parsimony.utils.linalgs.Solver
  - parsimony.utils.linalgs.SparseSolver
  - parsimony.utils.linalgs.TridiagonalSolver
- parsimony.functions.properties.StepSize
  - parsimony.functions.combinedfunctions.CombinedFunction: Combines one or more loss functions, any number of penalties and zero or one proximal operator.
  - parsimony.functions.multiblock.losses.GeneralisedMultiblock
  - parsimony.functions.losses.LatentVariableVariance
  - parsimony.functions.losses.LinearFunction: A linear function.
  - parsimony.functions.losses.LinearRegression: The Linear regression loss function.
  - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
  - parsimony.functions.losses.LogisticRegression: The Logistic Regression loss function.
  - parsimony.functions.multiblock.losses.MultiblockFunctionWrapper
    - parsimony.functions.multiblock.losses.MultiblockNesterovFunctionWrapper
  - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
  - parsimony.functions.losses.RidgeLogisticRegression: The Logistic Regression loss function with a squared L2 penalty.
  - parsimony.functions.losses.RidgeRegression: The Ridge Regression function, i.e.
- parsimony.functions.properties.StronglyConvex: Represents strongly convex functions.
  - parsimony.functions.combinedfunctions.LinearRegressionL1L2TV: Combination (sum) of LinearRegression, L1, L2 and TotalVariation.
    - parsimony.functions.combinedfunctions.LinearRegressionL1L2GL: Combination (sum) of RidgeRegression, L1 and Overlapping Group Lasso.
      - parsimony.functions.combinedfunctions.LogisticRegressionL1L2GL: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
    - parsimony.functions.combinedfunctions.LogisticRegressionL1L2TV: Combination (sum) of RidgeLogisticRegression, L1 and TotalVariation.
  - parsimony.functions.combinedfunctions.LinearRegressionL2SmoothedL1TV: Combination (sum) of Linear Regression, L2 and simultaneously smoothed L1 and TotalVariation.
  - parsimony.functions.combinedfunctions.PrincipalComponentAnalysisL1TV: Combination (sum) of PCA (Variance), L1 and TotalVariation
  - parsimony.functions.losses.RidgeRegression: The Ridge Regression function, i.e.
  - parsimony.functions.penalties.RidgeSquaredError: Represents a ridge squared error penalty, i.e.
- parsimony.utils.linalgs.MultipartArray.__ops
- type: type(object) -> the object's type type(name, bases, dict) -> a new type
  - abc.ABCMeta: Metaclass for defining Abstract Base Classes (ABCs).