OpenNMT Forum

Feature Requests


Topic Replies Activity
Support other forms of dropout/regularization 9 May 3, 2017
Resume training alternatives 15 April 6, 2017
Minimum Risk Training for NMT 1 April 3, 2017
Ensemble translation 3 March 30, 2017
A w2v-coupled training procedure 1 March 12, 2017
Model compression via weight tying 2 March 6, 2017
Support Crayon Logging 7 March 6, 2017
Alternate method of requesting vocab_size during preprocessing 3 March 2, 2017
Pre-trained/fixed embeddings for features 1 March 2, 2017
Support n_best option for CTranslate 5 February 22, 2017
It will be better if there is a Coverage Attention Module 3 February 21, 2017
Pre-processing steps to split? 2 February 20, 2017
Same source/target embeddings for intralingual task 3 February 9, 2017
Creating phrase tables out of alignment output for translating UNKs 3 January 16, 2017
Support of Translation History 1 January 22, 2017
External reward signal for human-in-the-loop training 3 January 20, 2017
Use CuDNN RNN for faster training 2 January 17, 2017
Larger scale monolingual resource for target language 2 January 17, 2017
Variable Dropout setting 4 January 17, 2017
BytePairEncoding 5 January 16, 2017
Add multi-threading in tokenization 3 January 13, 2017
Variable size features 3 January 12, 2017
Seq2seq pretraining 1 January 8, 2017
Large Vocabulary model 2 January 6, 2017
Parallel Training 1 December 28, 2016
`Preprocess.lua` length options 3 December 28, 2016
Differentiable memory 2 December 26, 2016