Skip to content

Latest commit

 

History

History
101 lines (64 loc) · 2.19 KB

README.md

File metadata and controls

101 lines (64 loc) · 2.19 KB

ViNNPruner

Code repository for the paper 'ViNNPruner: Visual Interactive Pruning for Deep Learning' (https://arxiv.org/abs/2205.15731).
An interactive demo can be found at https://interactive-pruning.dbvis.de/.

Welcome

This is the home of ViNNPruner, an interactive pruning tool for deep neural networks.

Frontend Setup

Installation

In order to run the frontend, you need Node 12LTS or greater.

It may be possible that you might need to install vue-cli depending on your system configuration. If there are any errors during build, please also run:

npm install -g @vue/cli

Setup

Navigate to the frontend directory and run:

npm install

Afterwards, you need to build the project.

Build

npn run build

Run

After completion, run the frontend web server by entering the following command:

npm run serve

Backend Setup

PyTorch

The interactive pruning backend uses PyTorch. It was originally written for PyTorch 1.8.1. It is highly recommended using an environment.

Navigate into the backend directory and run:

pip install -r requirements.txt

Run

After all packages installed successfully, run:

flask run

Use the app

The app will be available at http://localhost:8080

Results may take a while to load on first-run as datasets are downloaded on demand.


Used Third Party Libraries & Research

License

Released under MIT License. See the LICENSE file for details.

Reference

@conference{,
 author = {Schlegel, Udo and Schiegg, Samuel and Keim, Daniel A.},
 booktitle = {Machine Learning Methods in Visualisation for Big Data 2022 @ EuroVis 2022},
 title = {ViNNPruner: Visual Interactive Pruning for Deep Learning},
 year = {2022}
}