shawnwun/RNNLG
A benchmark toolkit for training and evaluating neural network-based Natural Language Generation models in spoken dialogue system domains.

RNNLG is a research toolkit released by Cambridge Dialogue Systems Group that provides benchmark datasets and reference implementations for training deep learning models on Natural Language Generation tasks in dialogue systems. The toolkit includes four domain-specific datasets (restaurant, hotel, laptop, TV search) along with cross-domain counterfeited datasets, and implements neural network training using Theano for model development and Numpy for efficient decoding.
Frequently asked
- What is shawnwun/RNNLG?
- A benchmark toolkit for training and evaluating neural network-based Natural Language Generation models in spoken dialogue system domains.
- Is RNNLG open source?
- Yes — shawnwun/RNNLG is an open-source project tracked on heatdrop.
- What language is RNNLG written in?
- shawnwun/RNNLG is primarily written in Python.
- How popular is RNNLG?
- shawnwun/RNNLG has 490 stars on GitHub.
- Where can I find RNNLG?
- shawnwun/RNNLG is on GitHub at https://github.com/shawnwun/RNNLG.