Skip to content

@fhieber fhieber released this Nov 29, 2018 · 64 commits to master since this release



  • Update to MXNet 1.3.1. Removed requirements/requirements.gpu-cu{75,91}.txt as CUDA 7.5 and 9.1 are deprecated.



  • Performance optimization to skip the softmax operation for single model greedy decoding is now only applied if no translation scores are required in the output.



  • Full training state is now returned from EarlyStoppingTrainer's fit().


  • Training state cleanup will not be performed for training runs that did not converge yet.
  • Switched to portalocker for locking files (Windows compatibility).



  • Added nbest translation, exposed as --nbest-size. Nbest translation means to not only output the most probable translation according to a model, but the top n most probable hypotheses. If --nbest-size > 1 and the option --output-type is not explicitly specified, the output type will be changed to one JSON list of nbest translations per line. --nbest-size can never be larger than --beam-size.


  • Changed sockeye.rerank CLI to be compatible with nbest translation JSON output format.
Assets 2
You can’t perform that action at this time.