`aim2dat.fct`¶

Modules to analyze and compare functions (e.g. DOS or spectra).

Submodules¶

Package Contents¶

Classes¶

`DiscretizedAxis`	Class to create an axis to discretize a 1d function i.e. a 2d plot
`DiscretizedGrid`	Class to create a grid to discretize a 1d function i.e. a 2d plot.
`FunctionAnalysis`	Class to analyze and compare functions.
`FunctionDiscretizationFingerprint`	Fingerprint for functions based on the DOS-Fingerprint presented in

class aim2dat.fct.DiscretizedAxis(axis_type, **kwargs)[source]¶

Class to create an axis to discretize a 1d function i.e. a 2d plot in a grid. Different methods for the discretization are available.

Overview

Properties¶
`T`	x –> y or y –> x.
`axis`	Axis array. Contains the discrete values.
`axis_type`	Specify whether this axis should be used as x or y axis in a grid.
`discretization_method`	Discretize the specified range.
`is_empty`	Check whether the axis is empty.
`shape`	Tuple specifying the dimensions of the axis (like numpy).

Methods¶
`discretize_axis`(**kwargs)	Perform the discretization of the specified range.
`transpose`()	Change the axis_type: x –> y or y –> x.

property T : DiscretizedAxis¶

x –> y or y –> x.

Returns:¶: DiscretizedAxis – Transposed axis.
Type:¶: Change the axis_type

property axis : numpy.array¶

Axis array. Contains the discrete values.

Returns:¶: np.array – The discretized range.

property axis_type : str¶

Specify whether this axis should be used as x or y axis in a grid.

Returns:¶: str – indicating whether this axis should be used as x or y axis in case of a merge into a grid.
Raises:¶: ValueError – The axis_type attribute needs to be specified

property discretization_method : collections.abc.Callable¶

Discretize the specified range. Can be chosen via a string, accepting the methods specified in _available_discretization_methods or by passing a callable function.

Returns:¶: Callable – Method to discretize the axis.

property is_empty : bool¶

Check whether the axis is empty.

Returns:¶: bool – Whether the axis is empty.

property shape : tuple¶

Tuple specifying the dimensions of the axis (like numpy).

Returns:¶: tuple – The shape of the axis.

discretize_axis(**kwargs) → DiscretizedAxis[source]¶

Perform the discretization of the specified range.

Returns:¶: DiscretizedAxis

transpose() → DiscretizedAxis[source]¶

Change the axis_type: x –> y or y –> x.

Returns:¶: DiscretizedAxis – Transposed axis.

class aim2dat.fct.DiscretizedGrid(**kwargs)[source]¶

Class to create a grid to discretize a 1d function i.e. a 2d plot.

Use the plot_grid method to visualize the created grid.

Overview

Properties¶
`grid`	Return the internal grid as a list of lists.
`is_empty`	Check whether the axis is empty.

Methods¶
`create_grid`()	Create the internal grid which is based on a list of lists.
`plot_grid`()	Plot the grid.

property grid : list¶

Return the internal grid as a list of lists.

Returns:¶: type – list: List of lists representing the x-values and discretized y-values.

property is_empty : bool¶

Check whether the axis is empty.

Returns:¶: bool – Whether the axis is empty.

create_grid() → DiscretizedGrid[source]¶

Create the internal grid which is based on a list of lists. Each list contains the energy-value (x) as the first argument and the DOS-values (y) as a list in the second argument.

Returns:¶: DiscretizedGrid – Discretized grid.

plot_grid()[source]¶: Plot the grid.

class aim2dat.fct.FunctionAnalysis(**kwargs)[source]¶

Class to analyze and compare functions.

Overview

Properties¶
`allowed_distance_methods`	Return allowed distance methods.

Methods¶
`calculate_correlation`(label1, label2)	Calculate the pearson-correlation between the values of two functions.
`calculate_discrete_fingerprint`(label, grid)	Calculate a discretized fingerprint of a function (doi:10.1038/s41597-022-01754-z).
`calculate_distance`(label1, label2, method)	Calculate the distance between the values of two functions.
`compare_areas`(label1, label2)	Compare the enclosed areas of two functions.
`compare_functions_by_discrete_fingerprint`(label1, label2, grid)	Compare two functions based on a discretized fingerprint
`import_data`(data_label, x_values, y_values)	Import data into the internal memory.

property allowed_distance_methods : list¶: Return allowed distance methods.

calculate_correlation(label1: str, label2: str) → float[source]¶

Calculate the pearson-correlation between the values of two functions.

Parameters:¶

label_1 (str) – Label of the first dataset.
label_2 (str) – Label of the second dataset.

Returns:¶

float – The correlation.

calculate_discrete_fingerprint(label: str, grid: aim2dat.fct.discretization.DiscretizedGrid) → numpy.array[source]¶

Calculate a discretized fingerprint of a function (doi:10.1038/s41597-022-01754-z).

Parameters:¶

label (str) – Label of the dataset.
grid (DiscretizedGrid) – The grid to discretize the function.

Returns:¶

np.array – Discretized fingerprint.

calculate_distance(label1: str, label2: str, method: str = 'euclidian') → float[source]¶

Calculate the distance between the values of two functions.

Parameters:¶

label1 (str) – Label of the first dataset.
label2 (str) – Label of the second dataset.
method (str) – The metric to calculate the distance. Defaults to “euclidian”.

Returns:¶

float – Abs. error

compare_areas(label1: str, label2: str) → float[source]¶

Compare the enclosed areas of two functions.

Parameters:¶

label1 (str) – Label of the first dataset.
label2 (str) – Label of the second dataset.

Returns:¶

float – The ratio of the areas. (A1 / A2)

compare_functions_by_discrete_fingerprint(label1: str, label2: str, grid: aim2dat.fct.discretization.DiscretizedGrid) → float[source]¶

Compare two functions based on a discretized fingerprint (doi:10.1038/s41597-022-01754-z).

Parameters:¶

label1 (str) – Label of the first dataset.
label2 (str) – Label of the second dataset.
grid (DiscretizedGrid) – The grid to discretize the function.

Returns:¶

float – Similarity.

import_data(data_label: str, x_values: numpy.array, y_values: numpy.array)[source]¶

Import data into the internal memory.

Parameters:¶

data_label (str) – Label for the internal memory.
x_values (np.array) – x-values of the function.
y_values (np.array) – y-values of the function.

class aim2dat.fct.FunctionDiscretizationFingerprint(grid, **kwargs)[source]¶

Fingerprint for functions based on the DOS-Fingerprint presented in doi:10.1038/s41597-022-01754-z.

Overview

Methods¶
`calculate_fingerprint`(x_values, y_values, label)	Calculate the fingerprint.
`compare_fingerprints`(label_1, label_2)	Compare two fingerprints that are stored in the internal memory.
`plot_fingerprint`(x_values, y_values)	Plot the discretized function and the corresponding grid.

calculate_fingerprint(x_values: numpy.array, y_values: numpy.array, label: str = None) → numpy.array[source]¶

Calculate the fingerprint.

Parameters:¶

x_values (np.array) – x-values of the function.
y_values (np.array) – y-values of the function. In case it’s a 2D-array, each row will be interpreted as a dataset and the fingerprint is calculated by concatenating the individual fingerprints.
label (str) – Label for the internal memory. Defaults to None.

Returns:¶

np.array – The discretized fingerprint.

compare_fingerprints(label_1: str, label_2: str) → float[source]¶

Compare two fingerprints that are stored in the internal memory.

Parameters:¶

label_1 (str) – Label of the first fingerprint.
label_2 (str) – Label of the second fingerprint.

Returns:¶

float – Similarity measure.

plot_fingerprint(x_values: numpy.array, y_values: numpy.array) → matplotlib.pyplot.Figure[source]¶

Plot the discretized function and the corresponding grid.

Parameters:¶

x_values (np.array) – x-values of the function.
y_values (np.array) – y-values of the function.

Returns:¶

plt.Figure – Plot of the discretized function.

aim2dat.fct¶

Submodules¶

Package Contents¶

Classes¶

`aim2dat.fct`¶