A repository with the code for the paper with the same title. The experiments are based on the more general-purpose graph neural network library OpenGNN. You can install it by following it's README.md.
Experiments are based around the train_and_eval.py script. Besides the main experiments, this repo also contains the following folders:
Parsers: A collection of scripts to parse and process various datasets to the format used by the experiments
Data: A collection of scripts to utility functions to handle and analyse the formated data
Models: Some bash script wrapppers around the main script with some model/hyperparameter combination for diferent experiments
Getting Started
As an example, we will show how run a sequenced-graph to sequence model with attention on the CNN/DailyMail dataset. This assumed the process data is located in /data/naturallanguage/cnn_dailymail/split/{train,valid,test}/{inputs,targets}.jsonl.gz.
For instruction on how to process see the corresponding subfolder.
Start by build vocabularies for the node and edge labels in the input side and tokens in the output side by running