fedimpute 0.3.0

fedimpute is a benchmarking tool for federated imputation

FedImpute: a benchmarking and evaluation tool for federated imputation across various missing data scenarios.

FedImpute is a benchmarking tool for the evaluation of federated imputation algorithms over various missing data scenarios under horizontally partitioned data.

• Documentation: Documentation
• Source Code: Source Code
• Tutorial: Tutorials

Installation

Firstly, install python >= 3.10.0, we have two ways to install

Install from pip:

pip install fedimpute

Install from package repo:

git clone https://github.com/idsla/FedImpute
cd FedImpute

python -m venv .venv

# window gitbash
source .venv/Scripts/activate

# linux/unix
source .venv/bin/activate

# Install the required packages
pip install -r requirements.txt

Basic Usage

See documentation: Documentation

Supported Algorithms

Imputation Algorithms:

Method	Type	Fed Strategy	Imputer (code)	Reference
Mean	Non-NN	local, fedmean	mean	-
EM	Non-NN	local, fedem	em	EM, FedEM
MICE	Non-NN	local, fedmice	mice	FedICE
MissForest	Non-NN	local, fedtree	missforest	MissForest, Fed Randomforest
MIWAE	NN	local, fedavg, fedprox, ...	miwae	MIWAE
GAIN	NN	local, fedavg, fedprox, ...	gain	GAIN
Not-MIWAE	NN	local, fedavg, fedprox, ...	notmiwae	Not-MIWAE
GNR	NN	local, fedavg, fedprox, ...	gnr	GNR

Federated Strategies:

Method	Type	Fed_strategy(code)	Reference
Local	non-federated	local	-
FedMean	traditional	fedmean	-
FedEM	traditional	fedem	FedEM
FedMICE	traditional	fedmice	FedMICE
FedTree	traditional	fedtree	FedTree
FedAvg	global FL	fedavg	FedAvg
FedProx	global FL	fedprox	FedProx
Scaffold	global FL	scaffold	Scaffold
FedAdam	global FL	fedadam	FedAdam
FedAdagrad	global FL	fedadagrad	FedAdaGrad
FedYogi	global FL	fedyogi	FedYogi
FedAvg-FT	personalized FL	fedavg_ft	FedAvg-FT

Contact

For any questions, please contact Sitao Min