philipperemy/keras-attention
A Keras/TensorFlow library providing attention layer implementations with Luong and Bahdanau score functions for deep learning models.
Not currently ranked — collecting fresh signals.
star history
This repository implements attention mechanisms as reusable Keras layers, supporting both multiplicative (Luong) and additive (Bahdanau) scoring methods. It integrates with TensorFlow 2.x and provides utilities for visualizing attention weights in neural network predictions. The library serves as a building block for sequence-to-sequence models and transformer architectures.
Frequently asked
- What is philipperemy/keras-attention?
- A Keras/TensorFlow library providing attention layer implementations with Luong and Bahdanau score functions for deep learning models.
- Is keras-attention open source?
- Yes — philipperemy/keras-attention is open source, released under the Apache-2.0 license.
- What language is keras-attention written in?
- philipperemy/keras-attention is primarily written in Python.
- How popular is keras-attention?
- philipperemy/keras-attention has 2.8k stars on GitHub.
- Where can I find keras-attention?
- philipperemy/keras-attention is on GitHub at https://github.com/philipperemy/keras-attention.