The aim of this project is to make accurate predictions of the eventual retweet cascade size using NLP transfer learning approaches.
Tweet data for model training is not published yet, but AWS S3 storage bucket will be opened for public access shortly.
This project is optimized for running on the Paperspace Gradient platform,
but scripts are generally able to run in any environment containing the required
packages. See the accompanying Dockerfile for details about environment setup.
You might need to adjust paths for model and data storage.
This repository is structured as follows:
- The root folder contains configuration files (Dockerfile, Makefile) and scripts to reproduce experiments
- The
retpredfolder contains the corresponding module which implements basic utilities for data processing and model training