

Keras
By Keras
Keras is an open-source software library designed for deep learning, providing a Python interface for artificial neural networks. It is known for its user-friendly, modular, and extensible design, making it accessible to both beginners and experienced practitioners. Keras can run on top of multiple backends, including TensorFlow, JAX, and PyTorch, offering flexibility and performance optimization.
Keras was founded on March 27, 2015, by François Chollet, a Google engineer, as part of the research effort of project ONEIROS (Open-ended Neuro-Electronic Intelligent Robot Operating System). Initially, Keras supported multiple backends like TensorFlow, Microsoft Cognitive Toolkit, and Theano. Over time, it became integrated into the TensorFlow library and later expanded to support JAX and PyTorch.
Keras offers a high-level API for building and training deep learning models. Key features include:
The latest major release, Keras 3.0, introduced several new features:
Keras emphasizes a culture of openness, collaboration, and inclusivity. The project follows a governance model that encourages community involvement and adheres to a code of conduct to ensure a respectful and inclusive environment. The core values include simplicity, flexibility, and a focus on user experience.
Keras has a large and active user base, including researchers, developers, and organizations worldwide. It is used by major companies and institutions like Google, Waymo, YouTube, CERN, and NASA. The community is highly engaged, contributing to discussions, code development, and support through platforms like GitHub and Slack.