The skeleton of a project intended to create a large language model that will create motions/topics for debates. It includes a web scraper, which collects data by scraping websites to collect debate motions. It also includes an implementation of the transformer architecture as well as a training script.

In order to run the training program download the packages for the training program use pip -r install requirements. Resources used: Datacamp and FreeCodeCamp

Current status: The training script is being linked with the architecture and data is being collected to be used for training.