About

This repo contains the implementation of the 'Method of Four Russians' for the 'Computation complexity' course.

The code is structured as follows:

• main.py contains 2 implementations of matrix multiplication: default (matmul) and the four russians algorithm (matmul4r) with all the necessary supplementary functions;
• tests.py contains the tests;
• comparison.ipynb contains the speed comparison between standard and four russians implementations on matrices up to size (100, 100)

Usage

Code:

import numpy as np
from main import matmul, matmul4r

size = 5
# All matrices must by of type np.array!
mat_a = np.random.randint(0, 2, (size, size))
mat_b = np.random.randint(0, 2, (size, size))

res_default = matmul(mat_a=mat_a, mat_b=mat_b)  # default algo, general case
res_default_binary = matmul(mat_a=mat_a, mat_b=mat_b, binary=True)  # default algo, binary case

res_4r = matmul4r(mat_a=mat_a, mat_b=mat_b)  # four russians algo, only binary

Tests:

In the repository root (with installed pytest in current environment) run

pytest tests.py

All the tests run on every commit with the help of GitHub Actions.

Useful links

• louridas.github.io/ - no implementation, only description
• neerc.ifmo.ru/wiki/ - description in russian