Empowering the League of Legends community with high-quality, open-source datasets, foundational tools and models for data science and research.
We believe in the power of data to unlock deeper insights into League of Legends gameplay.
We're committed to delivering:
- High-Quality Datasets: Curate, normalize, and enrich raw LoL data into formats specifically designed for efficient data analysis and machine learning workloads. All datasets will be open-code, most datasets will be FREE, a few will be monetized to help finance GPTilt.
- Foundational Tools for Common Problems: Create utility libraries and tools that simplify common needs when working with LoL data.
- Community Collaboration: Enable data scientists, analysts, and enthusiasts to easily access and utilize LoL data for their projects.
✅ Basic dataset of 10K challenger matches
✅ Ultimate datasets, of 10M match events and 15 minute snapshots
🏗️ GPTilt PoC - Proof of Concept of a Generative Pre-Trained Transformer-based model for modelling LoL event sequences
📊 Datasets
We welcome community involvement! If you find issues, have suggestions for datasets or tools, or want to contribute, please check the relevant repository's contribution guidelines or open an issue there.
GPTilt isn't endorsed by Riot Games and doesn't reflect the views or opinions of Riot Games or anyone officially involved in producing or managing Riot Games properties. Riot Games, and all associated properties are trademarks or registered trademarks of Riot Games, Inc.