Use your model as comparative benchmark

[lucas0]

Hello, I'm currently downloading the elmo weights (it will take a while) and inspecting your code. I want to include it (the model for task B) in a publication about automatic rumor classification as one of the models being comparatively evaluated. We are going to be using the same data accross all compared models so I would like to know how can I run the benchmark dataset in your model? What is the format it should be in and where should the code be modified in order to run with the new data?
Best,
Lucas.