huffb

huffb / HierFL

Public

A Flask web app for visualizing Hierarchical Federated Learning (HierFL) with MNIST/FMNIST image recognition capabilities, featuring real-time training progress tracking, model download, recognition result visualization, and result export.

24
5
100% credibility
Found Mar 28, 2026 at 24 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A research tool for training image recognition models in a distributed way across clients, edges, and cloud, with a web demo for MNIST digit recognition.

How It Works

1
👀 Discover the digit trainer

You find this handy tool that teaches computers to recognize handwritten numbers without sharing your private drawings.

2
🌐 Open the web page

Launch the simple, friendly web interface right from your computer.

3
⚙️ Choose training options

Pick easy settings like how many learners join and how long to practice.

4
🚀 Start the learning session

Hit the button to begin training and watch the progress bar fill up as your AI gets smarter.

5
📈 Check results and charts

See colorful graphs showing how well it's learning accuracy over time.

6
🖊️ Test with your drawings

Upload a picture of handwritten digits to see if it recognizes them correctly.

🎉 Celebrate smart recognition

Your AI nails the numbers perfectly, and you can download the trained model to use anywhere.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 24 to 24 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is HierFL?

HierFL is a Python Flask web app that lets you run and visualize hierarchical federated learning experiments on MNIST, Fashion-MNIST, or CIFAR-10 datasets. It simulates client-edge-cloud training flows with real-time progress tracking via a clean web UI, plus image upload for recognition, model downloads, and result exports to Excel. Perfect flask github projects example for devs wanting a ready-to-run flask web app demo of privacy-focused ML.

Why is it gaining traction?

It stands out with browser-based controls for tweaking params like client count, edge aggregations, and DP noise, showing live accuracy/loss charts without CLI hassle. The built-in digit recognizer on trained models adds instant demo value, and Flask's lightweight setup makes it a solid flask github template for custom FL UIs. Hooks devs with websocket-like progress updates in a flask websocket example style.

Who should use this?

ML researchers prototyping hierarchical FL setups for edge devices, students demoing client-edge-cloud training in classrooms, or privacy engineers testing DP noise on image classification. Ideal for flask web ui fans building hireflow apps or flask github tutorial extensions.

Verdict

Grab it for quick HierFL viz in a flask webserver—24 stars and 1.0% credibility score signal early-stage niche tool, but docs and README cover basics for fast spins. Fork and harden for production; skip if you need battle-tested scale.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.