Hendrik/xgboost

Rong Ou 14ef38b834

Initial support for federated learning (#7831 )

Federated learning plugin for xgboost:
* A gRPC server to aggregate MPI-style requests (allgather, allreduce, broadcast) from federated workers.
* A Rabit engine for the federated environment.
* Integration test to simulate federated learning.

Additional followups are needed to address GPU support, better security, and privacy, etc.

2022-05-05 21:49:22 +08:00

834 B

Raw Blame History

XGBoost Plugin for Federated Learning

This folder contains the plugin for federated learning. Follow these steps to build and test it.

Install gRPC

sudo apt-get install build-essential autoconf libtool pkg-config cmake ninja-build
git clone -b v1.45.2 https://github.com/grpc/grpc
cd grpc
git submodule update --init
cmake -S . -B build -GNinja -DABSL_PROPAGATE_CXX_STD=ON
cmake --build build --target install

Build the Plugin

# Under xgboost source tree.
mkdir build
cd build
cmake .. -GNinja -DPLUGIN_FEDERATED=ON
ninja
cd ../python-package
pip install -e .  # or equivalently python setup.py develop

Test Federated XGBoost

# Under xgboost source tree.
cd tests/distributed
./runtests-federated.sh