dependency_parsing_tf
Tensorflow implementation of "A Fast and Accurate Dependency Parser using Neural Networks"
https://cs.stanford.edu/~danqi/papers/emnlp2014.pdf
Tensorboard
tensorboard --logdir=path of model variables' folder
example: tensorboard --logdir=/dependency_parsing_tf/data/params_2017-09-18
Recent changes
- transition to tf 1.2
- added cube activation function (ref: paper)
- trainable word embeddings - initialized with 50d word2vec
- l2 loss for regularization (ref: paper)
- tensorboard visualization
- Dev UAS: 90.03 Test UAS: 90.42
- No functionality for LAS currently. it can be done with few changes in feature_extraction.py. I will try to add it.
training (exisiting dataset)
python parser_model.py
For new dataset
- Build new vocabulary & embedding matrices -> set "load_existing_dump=False" in parser_model.py. This will overwrite existing "data/dump" directory content
- python parser_model.py
training dataset
CONLL format