Skip to content

Latest commit

 

History

History
 
 

baselines

README.md

Flower Baselines

Note

We are changing the way we structure the Flower baselines. While we complete the transition to the new format, you can still find the existing baselines in the flwr_baselines directory. Currently, you can make use of baselines for FedAvg, FedOpt, and LEAF-FEMNIST.

Structure

Each baseline in this directory is fully self-contained in terms of source code in its own directory. In addition, each baseline uses its very own Python environment as designed by the contributors of such baseline in order to replicate the experiments in the paper. Each baseline directory contains the following structure:

baselines/<baseline-name>/
                ├── README.md
                ├── pyproject.toml
                └── <baseline-name>
                            └── *.py # several .py files

Running the baselines

Note

We are in the process of migrating all baselines to use flwr run. Those baselines that remain using the previous system (i.e. using Poetry, Hydra and start_simulation) might require you to first setup Poetry and pyenv already on your machine, please take a look at the Documentation for a guide on how to do so.

Each baseline is self-contained in its own directory. To run a baseline:

  1. Cloning the flower repository

    git clone https://github.com/adap/flower.git && cd flower

  2. Navigate inside the directory of the baseline you'd like to run.

  3. Follow the [Environment Setup] instructions in the README.md.

  4. Run the baseline as indicated in the [Running the Experiments] section in the README.md or in the [Expected Results] section to reproduce the experiments in the paper.

Contributing a new baseline

Do you have a new federated learning paper and want to add a new baseline to Flower? Or do you want to add an experiment to an existing baseline paper? Great, we really appreciate your contribution !!

Tip

A more verbose version of these steps can be found in the Flower Baselines documentation.

The steps to follow are:

  1. Create a new Python 3.10 environment and install Flower (pip install flwr)

  2. Fork the Flower repo and clone it into your machine.

  3. Navigate to the baselines/ directory, from there and with your environment activated, run:

    # Choose option "Flower Baseline" when prompted
flwr new <baseline-name>

  4. Then, go inside your baseline directory and continue with the steps detailed in the README.md.

  5. Once your code is ready, check that you have completed all the sections in the README.md and that, if a new environment is created, your baseline still runs (i.e. play the role of a person running the baseline you want to contribute).

  6. Create a Pull Request (PR). Then, the process to merge your baseline into the Flower repo will begin!

Further resources: