A flexible neural network framework designed with NLP in mind. It is fast on CPU, and contains a number of innovations (such as automatic operation batching) to allow for simple and efficient implementation of complicated models.
A neural machine translation toolkit designed for extensibility and easy experimentation.
A general purpose semantic parser based on abstract syntax description language.
A tool for holistic analysis of language generation systems.
You can find many other pieces of code for research projects on the NeuLab github and member pages, as well as in links from our papers.
A testbed for translation of noisy text.
A dataset for broad-coverage code generation.