Federated learning package
Project description
iFLearner - A Powerful and Lightweight Federated Learning Framework
iFLearner is a federated learning framework, which provides a secure computing framework based on data privacy security protection, mainly for federated modeling in deep learning scenarios. Its security bottom layer supports various encryption technologies such as homomorphic encryption, secret sharing, and differential privacy. The algorithm layer supports various deep learning network models, and supports mainstream frameworks such as Tensorflow, Mxnet, and Pytorch.
Architecture
The design of iFLearner is based on a few guiding principles:
-
Event-driven mechanism: Use an event-driven programming paradigm to build federated learning, that is, to regard federated learning as the process of sending and receiving messages between participants, and describe the federated learning process by defining message types and the behavior of processing messages.
-
Training framework abstraction: Abstract deep learning backend, compatible with support for multiple types of framework backends such as Tensorflow and Pytorch.
-
High scalability: modular design, users can customize aggregation strategies, encryption modules, and support algorithms in various scenarios.
-
Lightweight and simple: The framework is Lib level, light enough, and users can simply transform their deep learning algorithms into federated learning algorithms.
Documentation
Contributor
FAQ
License
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for iflearner-0.1.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | b85ebe100be312d8ac2180d2811b07024ccca8076a1a11f713df790e12f35f5d |
|
MD5 | ec317ed0bc5994f86769201fa0c9b4e2 |
|
BLAKE2b-256 | f748b08d22b60b62dafa0fd7d8a969a11978a42a76b38a990ff59e9ad6686b39 |