This repository contains the code of a Python package that implements Gaussian and Binomial distributions.

Download

Dependencies

This package requires:

• Python (>= 3.7.4)
• seaborn (>=0.11.1)
• matplotlib.pyplot (>=3.3.3)
• numpy (>=1.19.5)

Installation

To install the package use pip as follows:

$ pip3 install shay-dists

How to use

from distributions import Gaussian, Binomial

Making a Gaussian instance

The constructor receives the mean and standard deviation:

gaussian = Gaussian(25, 6)

Making a Binomial instance

The constructor receives the probability of an event ocurring and the size of the distribution:

binomial = Binomial(0.7, 10)

Adding two distributions

Due to Python magic methods, it's possible to sum two Gaussians and it outputs a new Gaussian class with the updated mean and standard deviation. The same applies to Binomial class.

gaussian_one = Gaussian(10, 3)
gaussian_two = Gaussian(32, 7)

gaussian_one + gaussian_two

Reading data from a file

You can load data from a .txt file using the method read_data_file() in both classes:

gaussian = Gaussian()
gaussian.read_data_file('numbers.txt')

And calculate the mean and standard deviation from the data using the methods calculate_mean() and calculate_stdev(), respectively.

In the Binomial class, to update the probability of an event ocurring from the dataset, use the method replace_stats_with_data().

Plotting histogram

For the Gaussian class use the method plot_histogram() and for the Binomial classe use plot_bar().

Probability density function

In the Binomial class, you can calculate the probability of k positive events ocurring in a sample of size n using the method pdf(k).