dqm.diversity package

Submodules

dqm.diversity.diversity module

This module, DiversityCalculator, calculates various types of diversity in datasets. It focuses on both lexical and visual diversities, employing statistical indices for different metrics such as richness, variety, color, and shape. Useful in linguistics, image processing, and data analysis, it helps understand the diversity of elements in a dataset.

Authors:

Faouzi ADJED Anani DJATO

Dependencies:

numpy collections.Counter

Classes:

DiversityCalculator: A class that provides methods for calculating

different types of diversity in datasets.

Functions: None

Usage:

To use this module, create an instance of the DiversityCalculator class and call its compute_diversity method with appropriate arguments. Example: calculator = DiversityCalculator() diversity_score = calculator.compute_diversity(data, ‘lexical’, ‘richness’)

class dqm.diversity.diversity.DiversityCalculator

Bases: object

A class to compute various types of diversity within data.

This class offers methods to calculate lexical and visual diversities in datasets using different statistical measures. It can measure lexical diversity in terms of richness and variety, and visual diversity in terms of color and shape using indices like Shannon, Simpson, and Gini-Simpson.

compute_diversity()

Calculates diversity based on specified type and need.

compute_diversity(data, diversity_type, need)

Compute diversity of given data based on type and need.

Parameters:

• data (Iterable) – Dataset for diversity computation.

• diversity_type (str) – Type of diversity (‘lexical’ or ‘visual’).

• need (str) – Specific need for calculation (‘richness’, ‘variety’, ‘color’, ‘shape’)

Returns:

Calculated diversity value.

Return type:

diversity (float)

validate_inputs(diversity_type, need)

This method is added just to have at least two public methods in a class as required by Python coding standards.

This method validates the inputs for compute_diversity method.

Args: diversity_type (str): Type of diversity to be computed. need (str): Specific need for diversity calculation.

Return type:

None

dqm.diversity.metric module

Diversity Index Calculator

This module defines the DiversityIndexCalculator class, which offers methods to calculate various diversity indices for categorical data. These indices are useful in statistical analysis and data science to understand the distribution and diversity of categorical data.

Authors:

Faouzi ADJED Anani DJATO

Dependencies:

pandas

Classes:

DiversityIndexCalculator: Provides methods for calculating diversity indices in a dataset.

Functions: None

Usage:

from metric import DiversityIndexCalculator calculator = DiversityIndexCalculator() dataset = pandas.Series([…]) # Replace with your data simpson_index = calculator.simpson(dataset) gini_index = calculator.gini(dataset)

These methods are useful for ecological, sociological, and various other types of categorical data analysis.

class dqm.diversity.metric.DiversityIndexCalculator

Bases: object

This class provides methods to calculate various diversity indices for a given dataset.

num()

Counts the number of each category in a dataset.

simpson()

Calculates the Simpson diversity index.

prob()

Calculates the frequencies of each category in a dataset.

gini()

Calculates the Gini-Simpson index.

RD(variable)

Return type:

float

gini(variable)

Compute the Gini-Simpson index, a metric for assessing diversity that takes into consideration both the quantity of distinct categories and the uniformity of their distribution.

Parameters:

variable (Series) – The data series for which to calculate the Gini-Simpson index.

Returns:

The Gini-Simpson index.

Return type:

g (float)

num(variable)

Calculate the number of each category of a variable.

Parameters:

variable (Series) – The data series for which to count categories.

Returns:

The count of each category.

Return type:

n (Series)

prob(variable)

Calculate the frequencies of each category in a variable.

Parameters:

variable (Series) – The data series for which to calculate frequencies.

Returns:

The frequency of each category.

Return type:

p (Series)

simpson(variable)

Calculate Simpson’s index, which is a measure of diversity.

Parameters:

variable (Series) – The data series for which to calculate the Simpson index.

Returns:

The Simpson diversity index.

Return type:

s (float)

Module contents