Source codes and corpora of paper "Iterated Dilated Convolutions for Chinese Word Segmentation"
Source codes and corpora of paper "Iterated Dilated Convolutions for Chinese Word Segmentation" published in NNW journal.
It implements the following
4models for CWS:
Both CPU and GPU are supported. GPU training is
Run following script to convert corpus to TensorFlow dataset.
$ ./scripts/run.sh $dataset $model
$ ./scripts/run.sh pku cnn
It will train a
pkudataset, then evaluate performance on test set.
To enable CRF layer, simply append
--viterbito your command, e.g.
$ ./scripts/run.sh pku cnn --viterbi