Skip to content

OpenNMT-tf 1.19.0

Compare
Choose a tag to compare
@guillaumekln guillaumekln released this 08 Feb 16:50
· 1142 commits to master since this release

OpenNMT-tf 1.19.0

New features

  • Experimental Horovod support
  • Experimental opennmt.v2 namespace that will expose modules compatible with TensorFlow 2.0
  • sacreBLEU external evaluator (requires Python 3)
  • Simplify configuration of non structural arguments: tokenization and pretrained embedding can now be configured in the YAML file directly

Fixes and improvements

  • In distributed training, only the master should save checkpoints
  • Clarify logging of training throughput, use source/target terminology instead of features/labels
  • Do not save the learning rate and words per second counters in the checkpoints