This project explores the interaction between natural language and trading card games, focusing on Magic: The Gathering as a case study. The goal is to analyze how in these contexts it becomes important to have a well-structured and predictable natural language. The main contribution concerns the application of a Large Language Model to generate scripts for cards. Through this approach, the potential and limitations of advanced language models in the context of script generation using a domain-specific language were evaluated. Several experiments were performed and the best quantitative and qualitative results were obtained with a Small Language Model. As a result of these experiments, it will be possible to formalize the language in its entirety, and the Small Language Model will be made available for card script generation.
Make sure you have unzipped the oracle-cards-20231201220150.json file in the scripts folder. Make sure you have unzipped the cardsfolder.zip file in the forge-scripts folder.
unzip scripts/oracle-cards-20231201220150.zip
unzip forge-scripts/cardsfolder.zip
python /scripts/prepareForgeScryfallTrainingData.py "scripts/oracle-cards-20231201220150.json" "forge-scripts" "compiled-data/compiled_cards_data"