Forecast Anything with Transformers - Chronos and PatchTST for Time Series Forecasting

This 57-minute tutorial from Trelis Research explores advanced time series forecasting using transformer-based models, specifically Chronos and PatchTST. Learn the fundamentals of time series forecasting, understand the architectural differences between these two powerful models, and discover when to use each one for optimal results. The video provides comprehensive demonstrations on implementing forecasting with PatchTST, training PatchTST models from scratch, utilizing Chronos for predictions, and fine-tuning Chronos models for specific applications. Follow along with practical code examples and gain access to resources that will help you apply these cutting-edge techniques to forecast virtually any time series data. Perfect for data scientists and machine learning practitioners looking to enhance their forecasting capabilities with transformer technology.