proxtoolbox.Problems

Created on Mon Dec 29 15:46:43 2015

@author: rebecca

The “Problems”-module contains all mathematical problems that can already be solved by the ProxToolbox.

class proxtoolbox.Problems.Ptychography(new_config={})

Bases: proxtoolbox.Problems.problems.Problem

Ptychography Problem

References

[1]	R. Hesse, D.R. Luke, S. Sabach, and M.K. Tam, “Proximal Heterogeneous Block Implicit-Explicit Method and Application to Blind Ptychographic Diffraction Imaging”, SIAM J. Imaging Sciences, 8(1):426–457, 2015.

Initialization of Ptychography problem

Parameters:

config : dict, optional - Dictionary containing the problem configuration. If unspecified, Ptychography.config is used.

_create_blocks()

Create blocks for the current problem

_gaussu(rho, z, l, wnull)

Calculates the electric field component of a Gaussian beam

Parameters:

rho : array_like - Input image

z : int - Position on the Z axis

l : int/float - Wave length

wnull : number - Waist radius

Returns:

array_like - Field amplitude

_get_ptychography_data()

Generates an experiment on-the-fly. If you want to load existing data or make your own experiments, you can overload this.

Returns:

d1x : number - Pixel dimension

d1y : number - Pixel dimension

positions : array_like

probe : array_like

sample_plane : array_like

i_exp : array_like

_im2hsv(im, th)

method

_postsolve()

Processes the solution and generates the output

_presolve()

Prepares argument for actual solving routine

_prop_nf_pa(im, l, delta_z, dx, dy)

Propagates a 2D-wave field within a paraxial approximation into the optical near field.

Parameters:

im : array_like - Incident 2D wave field

l : number - Wave length

delta_z : number - Propagation distance

dx, dy : int - Pixel widths in horizontal and vertical direction

Returns:

array_like - Propagated wave field

_solve()

Runs the warmup-algorithm if configured and starts the actual solving process to solve the given ptychography problem

show()

This procedure generates several figures visualizing the result.

class proxtoolbox.Problems.Ptychography_NTT_01_26210(config={'Nx': 192, 'Ny': 192, 'Nz': 1, 'algorithm': <class 'proxtoolbox.Problems.ptychography.PALM'>, 'beta0': 1, 'beta_max': 1, 'beta_switch': 30, 'between_blocks_scheme': 'averaged', 'block_cols': 2, 'block_maxiter': 1, 'block_rows': 2, 'block_tol': 1.0, 'blocking_scheme': 'single_view', 'blocking_switch': False, 'bs_factor': 1, 'bs_mask': None, 'ignore_error': False, 'maxiter': 2, 'noise': None, 'nx': 26, 'ny': 26, 'object_guess_type': 'constant', 'overrelax': 1, 'probe_guess_type': 'robin_initial', 'probe_mask_gamma': 1, 'ptychography_prox': 'Rodenburg', 'rmsfraction': 0.5, 'rodenburg_inner_it': 1, 'sample_area_center': (0.5, 0.5), 'scan_stepsize': 3.5e-07, 'scan_type': 'raster', 'switch_object_support_constraint': False, 'switch_probemask': False, 'tol': -1, 'warmup': True, 'warmup_alg': <class 'proxtoolbox.Problems.ptychography.PALM'>, 'warmup_maxiter': 0, 'within_blocks_scheme': 'none'})

Bases: proxtoolbox.Problems.ptychography.Ptychography

Ptychography NTT 01 26210

Initialization based on Ptychography

_get_ptychography_data()

Overloads the corresponding method of Ptychography to create other experiment data

class proxtoolbox.Problems.Sudoku(new_config={})

Bases: proxtoolbox.Problems.problems.Problem

Sudoku Problem

The goal of a standard Sudoku puzzle is to fill a 9x9 array with the numbers from 1 to 9. Every row, every column and each of the 9 3x3 subblocks should contain each number only once.

Starting point is a grid that already contains several numbers. Usually, there exists a unique solution.

The initialization of a Sudoku instance takes the default configuration and updates the parameters with the arguments in new_config.

Parameters:

new_config : dict, optional - Parameters to initialize the problem. If unspecified, the default config is used.

_postsolve()

Processes the solution and generates the output

_presolve()

Prepares argument for actual solving routine

_solve()

Runs the algorithm to solve the given sudoku problem

show()

Generates graphical output from the solution