CyberZHG/keras-self-attention
A Keras library providing self-attention layers for processing sequential data in neural networks.
Not currently ranked — collecting fresh signals.
star history
This repository implements self-attention mechanisms as Keras layers for processing sequential data. It provides SeqSelfAttention layers that can be added to neural network architectures to consider context across all timestamps when computing relevance. The library integrates with standard Keras components like Embedding, LSTM, and Bidirectional layers, enabling attention-enhanced sequence models for tasks like text classification and sequence labeling.
Frequently asked
- What is CyberZHG/keras-self-attention?
- A Keras library providing self-attention layers for processing sequential data in neural networks.
- Is keras-self-attention open source?
- Yes — CyberZHG/keras-self-attention is open source, released under the MIT license.
- What language is keras-self-attention written in?
- CyberZHG/keras-self-attention is primarily written in Python.
- How popular is keras-self-attention?
- CyberZHG/keras-self-attention has 657 stars on GitHub.
- Where can I find keras-self-attention?
- CyberZHG/keras-self-attention is on GitHub at https://github.com/CyberZHG/keras-self-attention.