google /seq2seq
A general-purpose encoder-decoder framework for Tensorflow
★ 5,619⑂ 1,281👁 1⑃ 91● 278▤ 0
+0
日增长
5,619
stars
1,281
forks
1
watchers
91
prs
278
issues
0
releases
README
---
READ THE DOCUMENTATION
CONTRIBUTING
---
A general-purpose encoder-decoder framework for Tensorflow that can be used for Machine Translation, Text Summarization, Conversational Modeling, Image Captioning, and more.
!Translation Model
---
The official code used for the Massive Exploration of Neural Machine Translation Architectures paper.
If you use this code for academic purposes, please cite it as:
@ARTICLE{Britz:2017,
author = {{Britz}, Denny and {Goldie}, Anna and {Luong}, Thang and {Le}, Quoc},
title = "{Massive Exploration of Neural Machine Translation Architectures}",
journal = {ArXiv e-prints},
archivePrefix = "arXiv",
eprinttype = {arxiv},
eprint = {1703.03906},
primaryClass = "cs.CL",
keywords = {Computer Science - Computation and Language},
year = 2017,
month = mar,
}
This is not an official Google product.Top issues
#44How to use Multiple GPUs?38 comments · 2017-03-14#216Failed to reproduce the result for NMT22 comments · 2017-05-10#285 from tensorflow.contrib.distributions.python.ops import bernoulli ImportError: cannot import name 'bernoulli'16 comments · 2017-07-26#316Error in pipeline unittest after setup13 comments · 2018-02-22#295Run python -m unittest seq2seq.test.pipeline_test on win712 comments · 2017-08-31#205trying to follow the nmt tutorial, cannot run the training step12 comments · 2017-05-04#4Serving using ExportStrategy11 comments · 2017-03-03#214BLEU during evaluation in training phase different from inference10 comments · 2017-05-09