Power Transform Revisited: Numerically Stable, and Federated

Installation

conda create -n powertf python=3.11
conda activate powertf

git clone https://github.com/xuefeng-xu/powertf.git && cd powertf

pip install -e .

For training experiments, see train/README.md.

For numerical experiments, see numerical/README.md.

For federated experiments, see federated/README.md.

Plots are saved in ./img/*.pdf

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
federated		federated
numerical		numerical
train		train
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
dataloader.py		dataloader.py
pyproject.toml		pyproject.toml