Graphical FrontEnd¶

Graphical frontend for fitting photo-current spectra.

class pypec.Analyse_PEC.Analyse_PEC(master=None)[source]¶

Construct a frame widget with the parent MASTER.

Valid resource names: background, bd, bg, borderwidth, class, colormap, container, cursor, height, highlightbackground, highlightcolor, highlightthickness, relief, takefocus, visual, width.

Methods

update_nb_runs()

Update the number of cpu and run per process on the graphical interface.

AddFiles_cb
Fit_cb
ask_quit
autoscale
create_fit_lines
get_progress
on_Run_Fit
on_hv_limits
on_start_workers
on_stop_button
plot_Graph
plot_Re_Im
plot_fit_lines
plot_ligne_V
prm_binary
process_queue
remove_fit_folder
remove_fit_lines
start
update_figure
update_legend
update_nb_fit_in_run

AddFiles_cb()[source]¶

Fit_cb()[source]¶

ask_quit()[source]¶

autoscale(chart)[source]¶

create_fit_lines()[source]¶

get_progress(run=1, fit=0)[source]¶

on_Run_Fit()[source]¶

on_hv_limits(*args)[source]¶

on_start_workers()[source]¶

on_stop_button()[source]¶

plot_Graph()[source]¶

plot_Re_Im()[source]¶

plot_fit_lines(*args)[source]¶

plot_ligne_V()[source]¶

prm_binary()[source]¶

process_queue()[source]¶

remove_fit_folder()[source]¶

remove_fit_lines()[source]¶

start()[source]¶

update_figure()[source]¶

update_legend(axes, run=1, fit=0, location='upper left')[source]¶

update_nb_fit_in_run()[source]¶

update_nb_runs()[source]¶: Update the number of cpu and run per process on the graphical interface.

class pypec.Analyse_PEC.ParameterTable(master, prm, last_prm_folder, **kwargs)[source]¶

Construct a frame widget with the parent MASTER.

Valid resource names: background, bd, bg, borderwidth, class, colormap, container, cursor, height, highlightbackground, highlightcolor, highlightthickness, relief, takefocus, visual, width.

class pypec.Analyse_PEC.ParameterWindow(master, prm_init, last_prm_folder)[source]¶

Construct a toplevel widget with the parent MASTER.

Valid resource names: background, bd, bg, borderwidth, class, colormap, container, cursor, height, highlightbackground, highlightcolor, highlightthickness, menu, relief, screen, takefocus, use, visual, width.

Methods

get_paths
get_prm

get_paths()[source]¶

get_prm()[source]¶

class pypec.Analyse_PEC.ScrolledFrame(master, **kwargs)[source]¶

Construct a frame widget with the parent MASTER.

Valid resource names: background, bd, bg, borderwidth, class, colormap, container, cursor, height, highlightbackground, highlightcolor, highlightthickness, relief, takefocus, visual, width.

Iph Functions¶

Modules - Iph Functions

This module contains functions requiered for computing and optimizing the photo-current values from input parameters i.e. triplets $(K_i, \theta_i, Eg_i)$ and experimental data for each semi-conductive phase [1].

pypec.iph_functions.get_Iph_calc(hv, prm_array, phi_N)[source]¶

Compute the complex values of $Iph$ based on the values and states of the triplets $(K_i, \theta_i, Eg_i)$ [1].

$Iph^{\ast} = \frac{Iph}{\Phi_{N}}$

Parameters:

hv: 1d array: Vector of energies for which the complex $Iph$ has to be computed.
prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .
phi_N: 1d array: Represents the values of the normalized photon flux to the maximum value. If nphf is a unity vector, the true photo-current is returned otherwise the as-measured photo-current is returned.

Returns:

iph_calc_complex: 1d array: Vector of the computed complex values of $Iph$ .

pypec.iph_functions.get_LCC(iph_exp_complex, iph_calc_complex)[source]¶

pypec.iph_functions.get_distance(iph_exp_complex, iph_calc_complex)[source]¶

Compute the distance $D$ between $Iph_{exp}$ and $Iph_{calc}$ . The distance is computed by multiplying the distances on real and imaginary parts of $Iph$ :

$\Delta Re & = Re \, Iph_{exp} - Re \, Iph_{calc} \\ \Delta Im & = Im \, Iph_{exp} - Im \, Iph_{calc} \\ D_{Re} & = \sqrt{\sum{\Delta Re ^2}}\\ D_{Im} & = \sqrt{\sum{\Delta Im ^2}}\\ D & = D_{Re} \cdot D_{Im}$

Parameters:

iph_exp_complex: 1d numpy array: Contains the complex values of the $Iph_{exp}$ .
iph_calc_complex: 1d numpy array: Contains the complex values of the $Iph_{calc}$ .

Returns:

D: float: The computed distance on real and imaginary parts of $Iph$ :.

pypec.iph_functions.get_exp_data(filepath)[source]¶

Get the data array of data files according to their extension.

Supported files are .dot files recorded by PECLab software and .data files were the first three columns represent $h\nu$ , $\left | Iph^{\ast} \right |$ , $\theta$ .

Parameters:

filepath: string: Path to the data file.

Returns:

data_array: 2d array: Experimental data.

pypec.iph_functions.get_header_footer_dot_file(filepath)[source]¶

Find the number of lines in header and footer in .dot files.

Parameters:

filepath: path to the dot file

Returns:

skip_header: int: number of lines in header
skip_footer: int: number of lines in footer
nbpoints: int: number of data lines

pypec.iph_functions.get_random_prm_values(prm_array, K_bound=(1e-12, 0.1), theta_bound=(-180.0, 180.0), Eg_bound=(0.1, 6.2), phase_flag=True)[source]¶

Generates random values for the triplets $(K_i, \theta_i, Eg_i)$ to be fitted based on the states given by the prm_array.

By default, the limits are:

$K_i$ : $[10^{-12},10^{-1}]$
$\theta _i$ : $[-\pi,+\pi]$
$Eg_i$ : $[0.1, 6.0]$

Parameters:

prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .
K_bound:tuple: Contains the lower and upper limits for the $K_i$ values.
theta_bound: tuple: Contains the lower and upper limits for the $\theta _i$ values.
Eg_bound:tuple: Contains the lower and upper limits for the $Eg_i$ values.
phase_flag: bool: Indicates if the values of $\theta _i$ have to be randomized.

Returns:

random_prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .

pypec.iph_functions.get_results_array(hv, iph_exp_complex, iph_calc_complex)[source]¶

Build the data array of the experimental and calculated data: $h\nu$ , $\left| Iph_{exp} \right|$ , $\theta _{exp}$ , $\left| Iph_{calc} \right|$ and $\theta _{calc}$

Parameters:

hv: 1d numpy array: Contains the energy vector.
iph_exp_complex: 1d array: Contains the complex values of $Iph_{exp}$ .
iph_calc_complex: 1d array: Contains the complex values of $Iph_{calc}$ .

Returns:

data_array: 2d array: Array containing the .

pypec.iph_functions.get_summary(fit_folder)[source]¶

List result files for the triplets $(K_i, \theta_i, Eg_i)$ at the end and the minimum of each run.

Compute the distance, the LCCs for the energy interval that was used for minimizing the the triplets $(K_i, \theta_i, Eg_i)$ .

The results are saved in 4 files: .SumEnd, .SumEndEg, *.SumMin, *.SumMinEg.

Parameters:

fit_folder: string: Path of the fit folder.

pypec.iph_functions.import_prm_file(filepath)[source]¶

Import the triplets $(K_i, \theta_i, Eg_i)$ from text file where each line represents a contributing semi-conductive phase.

Parameters:

filepath: string: Absolute or relative file path to the text file.

Returns:

prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .

pypec.iph_functions.minimize(hv, iph_exp_complex, phi_N, weights, prm_array, Ki_log_flag=True, maxiter=None, maxfun=None, xtol=1e-11, ftol=1e-23, full_output=True, retall=False, disp=False, callback=None)[source]¶

Execute the Nelder-Mead algorithm based on parameter values given by prm_array and energy vector $h\nu$ .

First, the prm_array is flattened and the parameters $(K_i, \theta_i, Eg_i)$ to be fitted are extracted and sent to the target_function() through the Nelder-Mead algorithm.

Once the parameters were computed by the Nelder-Mead algorithm, the prm_array is updated with the new values.

Parameters:

hv: 1d numpy array: Contains the energy vector.
iph_exp_complex: 1d numpy array: Contains the complex values of the experimental photo-current.
phi_N: 1d array: Contains the normalized photon spectrum.
weights: 1d array: Contains the weights of the data.
prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .
Ki_log_flag: bool: Indicates if the $K_i$ values are in logarithmic space.
maxiterint, optional: Maximum number of iterations to perform.
maxfunnumber, optional: Maximum number of function evaluations to make.
xtolfloat, optional: Relative error in xopt acceptable for convergence.
ftolnumber, optional: Relative error in func(xopt) acceptable for convergence.
full_outputbool, optional: Set to True if fopt and warnflag outputs are desired.
retallbool, optional: Set to True to return list of solutions at each iteration.
dispbool, optional: Set to True to print convergence messages.
callbackcallable, optional: Called after each iteration, as callback(xk), where xk is the current parameter vector.

Returns:

prm_array: 2d array: Represents the updated values and states of the triplets $(K_i, \theta_i, Eg_i)$ .
foptfloat: Value of function at minimum: fopt = func(xopt).

pypec.iph_functions.plot_summary(fit_folder)[source]¶

Plot the result files that were created by the get_summary() for he triplets $(K_i, \theta_i, Eg_i)$ at the end and the minimum of each run.

The results are saved in 2 files: -0-End.pdf, -0-Min.pdf.

Parameters:

fit_folder: string: Path of the fit folder.

pypec.iph_functions.save_pdf(filepath, hv, iph_exp_complex, iph_calc_complex, mask, all_results)[source]¶

pypec.iph_functions.save_results(run, process_id, fit_folder, datafilepath, suffix, hv, mask, iph_exp_complex, phi_N, prm_min_run, prm_end_run, distance_min_run, distance_end_run, minimization_results, header_minimization_results)[source]¶

pypec.iph_functions.scatter_logpolar(ax, theta, r_, ticks=5, bullseye=0.0, **kwargs)[source]¶

pypec.iph_functions.shift_phase(prm_array, theta_bound=(-180.0, 180.0))[source]¶

Compute the modulo of $\theta _i$ values with $2\pi$ and then shift the values of $\theta _i$ by the amplitude of the boundaries in order to be in between the boundaries.

By default, the boundaries for $\theta _i$ are set to $[-\pi,+\pi]$ .

Parameters:

prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .
theta_bound: tuple: Contains the lower and upper limits for the :math`theta _i` values.

Returns:

prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ where the $\theta _i$ values were shifted.

pypec.iph_functions.sort_prm_Eg(prm_array)[source]¶

Sort the prm_array based on values of $Eg_i$ .

Parameters:

prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .

Returns:

prm_array: 2d array: Represents the sorted values and states of the triplets $(K_i, \theta_i, Eg_i)$ .

pypec.iph_functions.target_func(p, hv, prm_array, iph_exp_complex, phi_N, weights, Ki_log_flag=True)[source]¶

Update the triplets $(K_i, \theta_i, Eg_i)$ from the flattened parameter vector p sent by the optimization algorithm. The prm_array will be flattened and the indexes of the parameters to be fitted will be updated with p vector.

The calculated complex values of $Iph$ will be sent along the experimental values to the get_distance() function. The value of the distance between the experimental and calculated data will sent back to the optimization algorithm.

Parameters:

p: 1d array: Parameter vector sent by the optimization algorithm which is always. flattened.
hv: 1d array: Vector of energies for which the complex values of $Iph$ have to be calculated.
prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .
iph_exp_complex: 1d numpy array: Contains the complex values of the experimental $Iph$ .
phi_N: 1d array: Represents the values of the normalized photon flux to the maximum value.
weights: 1d array: Contains the values of the data weights.
Ki_log_flag: bool: Indicates if the $K_i$ values are in logarithmic space.

Returns:

distance: float: Calculated distance between experimental and calculated data values. See the get_distance() function.

pypec.iph_functions.validate_prm(prm_array, K_bound=(1e-12, 0.1), Eg_bound=(0.1, 6.2))[source]¶

Check if the values of $K_i$ and $Eg_i$ are within the boundaries.

Parameters:

prm_array: 2d array: Represents the values and states of the triplets $(K_i, \theta_i, Eg_i)$ .
K_bound:tuple: Contains the lower and upper limits for the $K_i$ values.
Eg_bound:tuple: Contains the lower and upper limits for the $Eg_i$ values.

Returns:

valid: bool: Set to True if value of $K_i$ or $Eg_i$ is out of the boundaries.

Parallel Processes¶

Module for controlling the parallel processes running during the fitting procedure

class pypec.Parallel_Process.MinimizationProcess(output_queue, name, prm_init, nb_run, nb_SC, init_type, random_loops, hv, iph_exp_complex, iph_exp_complex_CI, phi_N, phi_N_CI, weights, hv_limits, nb_fit_in_run, fit_folder, filepath, suffix, NelderMead_options=(1e-11, 1e-23, 200, 200), ParameterConstraints=((1e-12, 0.1), (-180, 180), (0.1, 6.2), True), update_every=5)[source]¶

Methods

run()

Method to be run in sub-process; can be overridden in sub-class

shutdown

run()[source]¶: Method to be run in sub-process; can be overridden in sub-class

shutdown()[source]¶

class pypec.Parallel_Process.PlotSummaryProcess(fit_folder)[source]¶

Parameters:

fit_folder: str: Path to the folder where the summary files will be saved.

Attributes:

fit_folder: str: Path to the folder where the summary files will be saved.

Methods

run()

Method to be run in sub-process; can be overridden in sub-class

run()[source]¶: Method to be run in sub-process; can be overridden in sub-class

pypec.Parallel_Process.get_cpu_number()[source]¶: DocString

pypec.Parallel_Process.get_queue()[source]¶: DocString

pypec.Parallel_Process.initialize_processes(n, daemon=True, **kwargs)[source]¶

Create N processes as daemons and return them in a list.

Parameters:

n: int: Number of processes to be created.
daemon: bool: Flag for creating daemon process.
kwargs: dict: See MinimizationProcess

pypec.Parallel_Process.start_processes(workers)[source]¶

Parameters:

workers: subprocess workers

Returns:

0: return integer

Graphical FrontEnd¶

Iph Functions¶

Parallel Processes¶

Table of Contents

Previous topic

Next topic

This Page