Parallel Machine Learning Algorithms with MPI

This repository contains simple implementations of various machine learning algorithms parallelized using the Message Passing Interface (MPI) for distributed computing.

Algorithms Included

• Parallel Gradient Descent: Implementation of distributed linear regression with gradient averaging using MPI reduce operations.
• Parallel Model Evaluation: Distributed evaluation of machine learning models across multiple processes with communication via send/recv and reduce operations.
• Parallel K-means Clustering: Implementation of the K-means clustering algorithm in a distributed setting, parallelizing both the distance calculations and centroid updates.

Requirements

• mpi4py
• NumPy
• scikit-learn

Installation

# Install required packages
pip install mpi4py numpy scikit-learn

# Clone the repository
git clone https://github.com/lokk798/parallel-ml-algorithms-mpi.git
cd parallel-ml-algorithms-mpi

Usage

Run the algorithms using the MPI launcher:

# Run parallel gradient descent with 4 processes
mpiexec -n 4 python parallel_gradient_descent.py

# Run parallel model evaluation with 4 processes
mpiexec -n 4 python parallel_model_evaluation.py

# Run parallel K-means with 4 processes
mpiexec -n 4 python parallel_kmeans.py