A wrapper around tensor2tensor to flexibly train, interact, and generate data for neural chatbots.

2018-JDDC大赛第4名的解决方案

An Implementation of 'Attention is all you need' with Chinese Corpus

Datasets I have created for scientific summarization, and a trained BertSum model

An open-source neural machine translation system developed by Natural Language Processing Group, Nanjing University.

Data Augmentation by Backtranslation (DAB) ヽ( •_-)ᕗ

Kaggle新赛(baseline)-基于BERT的fine-tuning方案+基于tensor2tensor的Transformer Encoder方案

Tensor2tensor experiment with SpecAugment

NMT for chinese-english using tensor2tensor

Tesseract and Transformer

Grammar Error Correction Based on Tensor2Tensor

Unofficial Implementation of Universal Transformer https://arxiv.org/abs/1807.03819

Vietnamese Diacritic Restoration using Transformer Sequence-to-Sequence Model

Transfer learning of NLP with transformer, based on tensor2tensor

Text Sentiment Classification (Computational Intelligence Lab, ETH Zurich, 2018)

Trained Transformer model for Speech Recognition