Keras is an open-source high-level neural network API written in Python, designed for fast experimentation. It runs on top of backends such as TensorFlow and is integrated as TensorFlow’s official high-level interface.
Semantic Classification
Content
- Keras provides a user-friendly API for defining, training and evaluating neural networks using composable layers, models and training loops. Originally a standalone library supporting multiple backends, it became the official high-level interface of TensorFlow and later regained multi-backend support.
- Its design emphasises readability and rapid prototyping, hiding lower-level tensor operations behind concise model definitions. Keras is widely used in education and research as an accessible entry point to deep learning.