https://github.com/zimmerrol/neural-network-gui

A lightweight GUI to create neural network architectures.
https://github.com/zimmerrol/neural-network-gui

Last synced: about 1 year ago
JSON representation

A lightweight GUI to create neural network architectures.

• Host: GitHub
• URL: https://github.com/zimmerrol/neural-network-gui
• Owner: zimmerrol
• Created: 2017-08-30T13:37:17.000Z (almost 9 years ago)
• Default Branch: master
• Last Pushed: 2017-09-22T23:21:12.000Z (over 8 years ago)
• Last Synced: 2025-05-12T14:28:14.297Z (about 1 year ago)
• Language: C++
• Size: 9.73 MB
• Stars: 6
• Watchers: 2
• Forks: 1
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# neural-network-gui (aka NNCreator)

A lightweight GUI to create neural network architectures which can be used in popular deep learning frameworks.

# Structure

This project consits out of two important parts

  * The GUI written in C# with WPF (**NNGui**)

  * The network parser written in C++ (**NNCreator**)

  

## NNGui

This WPF application, written in C#, offers a easy to understand user interface allowing (unexperienced) users to create network architectures easily. Nevertheless, there are almost no restrictions regarding the network architecture, so that is just as well possible to generate more complex models with NNGui.

After the network architecture has been defined in NNGui one can export it as an human-readable .xml file. This .xml file can then later on be parsed by the NNCreator to create a real network model.

#### Short Demo of the UI ####

![](http://i.imgur.com/H6MoHbw.gif)

## NNCreator

This program/library, written in pure C++, is meant to import the previously .xml files describing a network's architecture. Afterwards it will create a graph to represent this architecture in one of the popular deep learning frameworks. At the beginning only [Microsoft's CNTK](https://github.com/Microsoft/CNTK) is supported, but it is planned to support different frameworks like [Caffe](http://caffe.berkeleyvision.org/), [Caffe2](https://github.com/caffe2/caffe2) and [tensorflow](https://www.tensorflow.org/) later, too.

#### Short Demo of the command line tool ####

![](http://i.imgur.com/V1S4hFq.gif)