AI & Machine Learning

Leverage EGI's distributed computing infrastructure for machine learning workloads and research projects with scalable resources and integrated tools.

Getting Started

Set Up Your Environment

Install our command-line interface and configure your project to start using EGI for machine learning development.

curl -sSL https://egi-community.sh/install | sh
                    

Sample Project

We provide a machine-learning template project that you can clone and immediately run with training data and models.

egi init my-machine-learning-project
cd my-machine-learning-project
egi start
                    

What's Included

Distributed Training

Train models using HPC resources from multiple locations and leverage parallel computation capabilities.

Model Sharing

Share models and data seamlessly between collaborators and across different research teams.

Experiment Tracking

Keep track of model versions, hyperparameters, and validation outputs across distributed runs.

Advanced Features

Hyperparameter Tuning

Automate hyperparameter tuning across distributed GPU and CPU resources for faster model development and validation.

egi tune --model my_model.yaml
                    

Distributed Training

Split training jobs across multiple compute nodes for better resource utilization and performance.

egi train --distributed --nodes 3
                    

Need More Help?

Explore our complete documentation to start training AI models, managing datasets, or contributing to shared open science projects using EGI.