Federated_learning codes used for the the paper "Evaluation of Federated Learning Aggregation Algorithms" and "A Federated Learning Aggregation Algorithm for Pervasive Computing: Evaluation and Comparison"

Overview

Federated Distance (FedDist)

This is the code accompanying the Percom2021 paper "A Federated Learning Aggregation Algorithm for Pervasive Computing: Evaluation and Comparison" and the code of federated learning experiments by Sannara Ek during his master thesis.

Overview


This experiments compares 3 federated learning algorithms along with a new one, FedDist. The FedDist algorithm incorporates a pair-wise distance scheme for identifying outlier-like neurons/filters. These outlier-like neurons/filter may be in fact features learned from sparse data and so it is directly added to the server model for the next round of training.

Core Dependencies (tested and stable)


  • Tensorflow 2.2.2
  • PyTorch 1.1
  • scikit-learn 0.22.1

All the working scripts are presented in a Jupiter notebook file format.

There is an array of 3rd party packages that is necessary for the entirety of the scripts to run. It is recommended to run command "pip3 install -r requirements.txt" in your virtual environment and working directory to replicate the environments used in this experiment.

!Note! Visual Studio is required to solve dependency problems when working on a Windows Machine

Data Preparation


"DATA_UCI.ipynb" and "DATA_REALWORLD_SPLITSUB.ipynb" are respectively used to prepare the UCI and REALWORLD dataset for training. Simply run all cells in a Jupyter notebook. The formatted dataset will be placed in a new directory "datasetStand"

FL script implementations


The FedAvg and FedPer implementations are found in the file "FedAvg_FedPer.ipynb". You must specify which algorithm you which to run in the third cell of the notebook by changing the "algorithm" variable to either "FEDAVG" or "FEDPER"

FedDist is found in the "FedDist.ipynb" file.

FedMA is found in the "FedMA.ipynb" file.

For all the federated algorithms, the third cell gives a variety of options and testing environment to choose from. We recommend leaving the configuration in default other than changing the "algorithm" variable and specifying the GPU/CPU to use. Simply run all cells to start training.

If preferred to run as a python script, convert the files to a .py format VIA Jupiter notebook (FILES -> Download as -> Python (.py)).

Additionally with the command below from a console achieves the same result:

jupyter nbconvert --to script '[ScriptName].ipynb'

Simply specify the wanted parameters in the third cell beforehand.

Results Interpretability


All results of each experiments shall generate the "savedModels" folder. Within this folder will contain subfolders with the name of the chosen configuration and model architecture of the experiment. Additionally, within each model architecture folder will contain the another subfolder with the name of the dataset used for the experiment. E.g a directory should appear like:

./savedModels/FED_5C_10LE_50CR_400D_100D_BALANCED/UCI

Now within this folder:

The final server model is saved in a .h5 format. The recorded training statistics foreach communication round, such as the accuracy and loss of the clients model and server model, are stored in the trainingStats folder. The results regarding the Global accuracy and the detail of the server model can be found on the generated Server-Measure.csv file. Results for the Personalization accuracy can be found in the indivualClients Measure.csv file and finally the Generalization accuracy can be found at the AllClientsMeasure.csv file.

Sample script sequence:


An example of execution would be to first download and format the dataset (UCI and REALWORLD) then execute one of the FL algorithms (requires several days on CPU).

1.DATA_UCI.ipynb
2.DATA_REALWORLD_SPLITSUB.ipynb
3.FedAvg_FedPer.ipynb/FedDist.ipynb/FedMA.ipynb

Citing this work:


@INPROCEEDINGS{Lala2103:Federated,
AUTHOR="Sannara Ek and François Portet and Philippe Lalanda and German Vega",
TITLE="A Federated Learning Aggregation Algorithm for Pervasive Computing:
Evaluation and Comparison",
BOOKTITLE="2021 IEEE International Conference on Pervasive Computing and
Communications (PerCom) (PerCom 2021)",
ADDRESS="Kassel, Germany",
DAYS=21,
MONTH=mar,
YEAR=2021,
KEYWORDS="Federated Learning; Edge Computing; Human activity recognition"
}

Contact:


Please contact the authors by [firstname].[lastname]@univ-grenoble-alpes.fr if you have issues with the code.

To contact Sannara Ek, Please use [firstname].[lastname]@gmail.com

Owner
GETALP
Study Group for Machine Translation and Automated Processing of Languages and Speech
GETALP
Chainer Implementation of Fully Convolutional Networks. (Training code to reproduce the original result is available.)

fcn - Fully Convolutional Networks Chainer implementation of Fully Convolutional Networks. Installation pip install fcn Inference Inference is done as

Kentaro Wada 218 Oct 27, 2022
Anti-UAV base on PaddleDetection

Paddle-Anti-UAV Anti-UAV base on PaddleDetection Background UAVs are very popular and we can see them in many public spaces, such as parks and playgro

Qingzhong Wang 2 Apr 20, 2022
Provide baselines and evaluation metrics of the task: traffic flow prediction

Note: This repo is adpoted from https://github.com/UNIMIBInside/Smart-Mobility-Prediction. Due to technical reasons, I did not fork their code. Introd

Zhangzhi Peng 11 Nov 02, 2022
OpenVisionAPI server

🚀 Quick start An instance of ova-server is free and publicly available here: https://api.openvisionapi.com Checkout ova-client for a quick demo. Inst

Open Vision API 93 Nov 24, 2022
Source code release of the paper: Knowledge-Guided Deep Fractal Neural Networks for Human Pose Estimation.

GNet-pose Project Page: http://guanghan.info/projects/guided-fractal/ UPDATE 9/27/2018: Prototxts and model that achieved 93.9Pck on LSP dataset. http

Guanghan Ning 83 Nov 21, 2022
Test-Time Personalization with a Transformer for Human Pose Estimation, NeurIPS 2021

Transforming Self-Supervision in Test Time for Personalizing Human Pose Estimation This is an official implementation of the NeurIPS 2021 paper: Trans

41 Nov 28, 2022
Exporter for Storage Area Network (SAN)

SAN Exporter Prometheus exporter for Storage Area Network (SAN). We all know that each SAN Storage vendor has their own glossary of terms, health/perf

vCloud 32 Dec 16, 2022
Code for unmixing audio signals in four different stems "drums, bass, vocals, others". The code is adapted from "Jukebox: A Generative Model for Music"

Status: Archive (code is provided as-is, no updates expected) Disclaimer This code is a based on "Jukebox: A Generative Model for Music" Paper We adju

Wadhah Zai El Amri 24 Dec 29, 2022
An evaluation toolkit for voice conversion models.

Voice-conversion-evaluation An evaluation toolkit for voice conversion models. Sample test pair Generate the metadata for evaluating models. The direc

30 Aug 29, 2022
Code for "Learning the Best Pooling Strategy for Visual Semantic Embedding", CVPR 2021

Learning the Best Pooling Strategy for Visual Semantic Embedding Official PyTorch implementation of the paper Learning the Best Pooling Strategy for V

Jiacheng Chen 106 Jan 06, 2023
PConv-Keras - Unofficial implementation of "Image Inpainting for Irregular Holes Using Partial Convolutions". Try at: www.fixmyphoto.ai

Partial Convolutions for Image Inpainting using Keras Keras implementation of "Image Inpainting for Irregular Holes Using Partial Convolutions", https

Mathias Gruber 871 Jan 05, 2023
U-Net Implementation: Convolutional Networks for Biomedical Image Segmentation" using the Carvana Image Masking Dataset in PyTorch

U-Net Implementation By Christopher Ley This is my interpretation and implementation of the famous paper "U-Net: Convolutional Networks for Biomedical

Christopher Ley 1 Jan 06, 2022
[ICCV 2021] A Simple Baseline for Semi-supervised Semantic Segmentation with Strong Data Augmentation

[ICCV 2021] A Simple Baseline for Semi-supervised Semantic Segmentation with Strong Data Augmentation

CodingMan 45 Dec 12, 2022
Alex Pashevich 62 Dec 24, 2022
PyTorch implementation of "Learn to Dance with AIST++: Music Conditioned 3D Dance Generation."

Learn to Dance with AIST++: Music Conditioned 3D Dance Generation. Installation pip install -r requirements.txt Prepare Dataset bash data/scripts/pre

Zj Li 8 Sep 07, 2021
Compare neural networks by their feature similarity

PyTorch Model Compare A tiny package to compare two neural networks in PyTorch. There are many ways to compare two neural networks, but one robust and

Anand Krishnamoorthy 181 Jan 04, 2023
MediaPipeのPythonパッケージのサンプルです。2020/12/11時点でPython実装のある4機能(Hands、Pose、Face Mesh、Holistic)について用意しています。

mediapipe-python-sample MediaPipeのPythonパッケージのサンプルです。 2020/12/11時点でPython実装のある以下4機能について用意しています。 Hands Pose Face Mesh Holistic Requirement mediapipe 0.

KazuhitoTakahashi 217 Dec 12, 2022
The Power of Scale for Parameter-Efficient Prompt Tuning

The Power of Scale for Parameter-Efficient Prompt Tuning Implementation of soft embeddings from https://arxiv.org/abs/2104.08691v1 using Pytorch and H

Kip Parker 208 Dec 30, 2022
Real-world Anomaly Detection in Surveillance Videos- pytorch Re-implementation

Real world Anomaly Detection in Surveillance Videos : Pytorch RE-Implementation This repository is a re-implementation of "Real-world Anomaly Detectio

seominseok 62 Dec 08, 2022
Practical and Real-world applications of ML based on the homework of Hung-yi Lee Machine Learning Course 2021

Machine Learning Theory and Application Overview This repository is inspired by the Hung-yi Lee Machine Learning Course 2021. In that course, professo

SilenceJiang 35 Nov 22, 2022