nlp-models
A collection of NLP models in PyTorch. If PyTorch implementation is not publicly available, we would like to include it here.
Frameworks
- allennlp: An open-source NLP research library, built on PyTorch.
- ParlAI: A framework for training and evaluating AI models on a variety of openly available dialog datasets.
- OpenNMT-py: Pytorch port of OpenNMT, an open-source (MIT) neural machine translation system.
Tutorials
Sequence Models
- QRNN
- publication: Quasi-Recurrent Neural Networks
- code
- sru
- publication: Training RNNs as Fast as CNNs
- code
Sequence-to-Sequence Learning
- AuCson's fast seq2seq
- batched tensor manipulation of attention
- CopyNet
- publication: Incorporating Copying Mechanism in Sequence-to-Sequence Learning
- code: by Minje Choi
- MemNN
- code: MemNN: Facebook Torch implementation for various memory networks
- Key-variable memory
- publication: Key-Value Memory Networks for Directly Reading Documents
- code (in progress)
- ConvS2S
- publication: Convolutional Sequence to Sequence Learning A Convolutional Encoder Model for Neural Machine Translation
- code: fairseq-py: Facebook AI Research official implementation
- T2T Transformer
- publication: Attention is All You Need
- code: attention-is-all-you-need-pytorch
Machine Comprehension
- BiDAF
- website
- PyTorch implementation wanted
- DrQA
- publication: Reading Wikipedia to Answer Open-Domain Questions
- code: DrQA: Official implementation slightly different from ParlAI's version
Program Synthesis
- NL2Code
- publication: A Syntactic Neural Model for General-Purpose Code Generation
- code: NL2Code: Theano implementation by the authors
- pytorch implementation (in progress)
Program Induction
- Neural Symbolic Machine
- publication: Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision
- code (in progress)
- N-Gram Machine
- publication: Learning to Organize Knowledge with N-Gram Machines
- code (in progress)
Generative Models
- Texygen: Text GANs model zoo in Tensorflow