This repository contains Python code to perform Monte Carlo simulations of scenarios in which a set of n elements is randomly drawn from a set of m elements, with replacement.
The code is written in such a way to enable the description of any particular scenarios of interest within the context of the problem. Implementations of the following scenarios are provided:
- Drawing all unique elements.
- Drawing no pairs, but allowing for the drawing of the same element multiple times.
Environment management is with poetry. Once poetry is installed, clone the repository and run poetry install to install the dependencies.
The simulation may be run with exact_pairs_probability.py by executing exact_pairs_probability.py with the desired options. For example:
poetry run python exact_pairs_probability.py -n 5 -m 26 -k 1000000 --experiment_type all_uniqueThis will run the simulation for the scenario in which n=5 elements are drawn from a set of m=26 elements, with k=1000000 iterations, and the probability that all elements are unique is calculated.
Run poetry run python exact_pairs_probability.py --help for a list of available options.
Results are plotted during execution of exact_pairs_probability.py. By passing an argument to the --save_filename option, the data may be saved to a file for later plotting. The plot_results.py script may be used to plot the results.
Set up the development environment using poetry install then launch a shell with poetry shell.
Pre-commit hooks are available to enable automated code formatting and linting, and can be installed with pre-commit install.