Desktop appSoftware development

ml_forge

Build, train, run inference, and export image-classification models with a visual node editor.

Stars
417
Forks
50
License
MIT
Updated
Updated Jul 11, 2026

Checking live repository facts…

Project overview

ML Forge is a visual PyTorch pipeline editor for building image-classification models without writing the training workflow by hand. Users drag data-preparation, model, and training nodes onto a canvas, connect them, and let the application generate and run the code. It shows live loss curves, saves checkpoints, supports inference with the trained model, and exports projects as readable PyTorch. Python 3.10 or newer is required, and PyTorch plus torchvision must be installed separately before training.

Repository facts

Primary language
Python
License
MIT
Repository updated
Jul 11, 2026
Default branch
main

Resource types

Desktop app

Use cases

Software development

Runtime

Desktop

Audience

Developers

Related projects

Browse more similar projects