A PyTorch implementation of bidirectional LSTM-CRF for sequence tagging, adapted from the PyTorch tutorial. It is a character-based model and supports CUDA and mini-batch training.
Training data should be formatted as below:
word/tag word/tag word/tag ...
word/tag word/tag word/tag ...
...
To prepare data:
python prepare.py training_data
To train:
python train.py model word_to_idx tag_to_idx training_data.csv num_epoch
To predict:
python predict.py model.epochN word_to_idx tag_to_idx test_data