Skip to content

Code for CVPR2023 DaFKD : Domain-aware Federated Knowledge Distillation

License

Notifications You must be signed in to change notification settings

eternal-f1ame/DaFKD2023

 
 

Repository files navigation

DA-FKD: Domain Aware Federated Knowledge Distillation

Advance Machine Learning Paper - 2

Aaditya Baranwal (B20EE001)

System Requirements

  • OS: Ubuntu 20.04/22.04 / Windows 10/11
  • Python: >=3.6
  • conda: >=23.0.0
  • Follow the steps below to install the required packages
conda create -n <ENV-NAME> python=3.9
conda activate <ENV-NAME>
pip install -r requirements.txt

Clone the repository

  • Navigate to the directory where you want to clone the repository: < DIR >
git clone <REPO URL>
  • Navigate to the cloned repository
cd <DIR>/DA-FKD

Download the Dataset

  • URL:

    https://github.com/zalandoresearch/fashion-mnist/tree/master/data

Train the Teacher Model

  • Do this once to make th sh file executable

    chmod +x run_fed_standalone.sh
  • Run the following command to train the teacher model

    sh run_fed_standalone.sh

Releases

No releases published

Packages

No packages published

Languages

  • Python 99.2%
  • Shell 0.8%