This repository contains the code for replicating results from
- Coreference Resolution with Entity Equalization
- In ACL 2019
- The baseline model is from the paper Higher-order Coreference Resolution with Coarse-to-fine Inference
- Code for baseline model: https://github.com/kentonl/e2e-coref
- Install python (either 2 or 3) requirements:
pip install -r requirements.txt
- Download pretrained models at https://drive.google.com/file/d/1fkifqZzdzsOEo0DXMzCFjiNXqsKG_cHi
- Move the downloaded file to the root of the repo and extract:
tar -xzvf e2e-coref.tgz
- Move the downloaded file to the root of the repo and extract:
- Download GloVe embeddings and build custom kernels by running
setup_all.sh
.- There are 3 platform-dependent ways to build custom TensorFlow kernels. Please comment/uncomment the appropriate lines in the script.
- Extract BERT features by running
extract_bert_features.sh
- To train your own models, run
setup_training.sh
- This assumes access to OntoNotes 5.0. Please edit the
ontonotes_path
variable.
- This assumes access to OntoNotes 5.0. Please edit the
- Experiment configurations are found in
experiments.conf
- Choose an experiment that you would like to run, e.g.
best
- Training:
python train.py <experiment>
- Results are stored in the
logs
directory and can be viewed via TensorBoard. - Evaluation:
python evaluate.py <experiment>
- Command-line demo:
python demo.py final
- To run the demo with other experiments, replace
final
with your configuration name.
- Create a file where each line is in the following json format (make sure to strip the newlines so each line is well-formed json):
{
"clusters": [],
"doc_key": "nw",
"sentences": [["This", "is", "the", "first", "sentence", "."], ["This", "is", "the", "second", "."]],
"speakers": [["spk1", "spk1", "spk1", "spk1", "spk1", "spk1"], ["spk2", "spk2", "spk2", "spk2", "spk2"]]
}
clusters
should be left empty and is only used for evaluation purposes.doc_key
indicates the genre, which can be one of the following:"bc", "bn", "mz", "nw", "pt", "tc", "wb"
speakers
indicates the speaker of each word. These can be all empty strings if there is only one known speaker.- Run
python predict.py <experiment> <input_file> <output_file>
, which outputs the input jsonlines with predicted clusters.
- It does not use GPUs by default. Instead, it looks for the
GPU
environment variable, which the code treats as shorthand forCUDA_VISIBLE_DEVICES
. - The training runs indefinitely and needs to be terminated manually. The model generally converges at about 400k steps.