zepingyu0512/srnn
Sliced Recurrent Neural Network (SRNN) implementation in Keras for faster sequence processing.
Not currently ranked — collecting fresh signals.
star history
SRNN is a recurrent neural network architecture that accelerates standard RNN by slicing sequences into subsequences processed in parallel. Implemented in Keras with TensorFlow backend, this research project demonstrates the SRNN(8,2) configuration on the Yelp sentiment analysis dataset. The work was published at COLING 2018 and also supports text generation tasks similar to language modeling.
Frequently asked
- What is zepingyu0512/srnn?
- Sliced Recurrent Neural Network (SRNN) implementation in Keras for faster sequence processing.
- Is srnn open source?
- Yes — zepingyu0512/srnn is an open-source project tracked on heatdrop.
- What language is srnn written in?
- zepingyu0512/srnn is primarily written in Python.
- How popular is srnn?
- zepingyu0512/srnn has 468 stars on GitHub.
- Where can I find srnn?
- zepingyu0512/srnn is on GitHub at https://github.com/zepingyu0512/srnn.