We provide sample sentences from GENIA for illustration. Requirement: PyTorch (tested on 0.3.1), Python(v3)
- Put the embedding file glove.6B.100d.txt into the folder of embeddings (or directly downlaod from here)
- run the gen_data.py to generate the processed data for training, they will be defaultly placed at the ./data folder
- run train.py to start training
We already generate the processed data for our samples. So you can skip the first and second steps if you only want to test our sample sentences.
Configurations of the model and training are in config.py
Some important configurations:
- if_gpu: whether to use GPU for training
- if_C: whether we make a length restriction or not
- if_margin: whether to use softmax margin
The full GENIA data can be found here.
To reproduce the results in this paper, all the hyperparameters are listed in the supplementary paper.