Skip to content

Latest commit

 

History

History
19 lines (10 loc) · 393 Bytes

File metadata and controls

19 lines (10 loc) · 393 Bytes

dynamic-segmentation

Dynamic convolutional LSTM approach for character-level word segmentation

The models and architecture heavily relies on https://github.com/juditacs/morph-segmentation

Libraries used

  • tensorflow 1.3 (and master) custom version merged together
  • tensorflow fold master
  • PyTorch

Setup

pip install -r requirements.txt
python setup.py install