https://github.com/mmasoud1/meshdist_nvflare

MeshNet distributed learning by using the NVFlare framework
https://github.com/mmasoud1/meshdist_nvflare

Last synced: about 2 months ago
JSON representation

MeshNet distributed learning by using the NVFlare framework

• Host: GitHub
• URL: https://github.com/mmasoud1/meshdist_nvflare
• Owner: Mmasoud1
• Created: 2024-09-05T14:58:50.000Z (4 months ago)
• Default Branch: main
• Last Pushed: 2024-10-27T20:36:04.000Z (2 months ago)
• Last Synced: 2024-10-28T00:54:53.131Z (2 months ago)
• Language: Python
• Size: 170 KB
• Stars: 0
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# MeshDist_nvflare

MeshNet distributed learning by using the NVFlare framework.  

This work is in progress.. 

## Quick Setup Guide

### 1. Clone the Repository

First, clone the repository from GitHub or the relevant location to your machine:

git clone https://github.com/Mmasoud1/MeshDist_nvflare.git

cd MeshDist_NVFlare

### 2. Check for PyTorch with CUDA Support 

In the terminal, check if CUDA and torch is installed :

nvcc --version

python -c "import torch; print(torch.cuda.is_available())"

This should return true. 

### 3. Install nvflare

pip install nvflare==2.4.0

Recommended: 

pip install numpy==1.22.0

pip install pandas==2.0.3

### 4. Set Required Environment Variables

export PYTHONPATH=$PYTHONPATH:[path to this dir]/app/code/

export NVFLARE_POC_WORKSPACE=[path to this dir]/poc-workspace/

### 5. Run nvflare simulator

nvflare simulator -c site1,site2 ./jobs/job