Int OnInit

Jul 11, 2024 | mql5.com | Int OnInit |shashank rai

I first encountered an algorithm called PatchTST when I started to dig into the AI advancements associated with time series predictions on Huggingface.co. As anyone who has worked with large language models (LLMs) would know, the invention of transformers has been a game changer for developing tools for natural language, image, and video processing. But what about time series? Is it something that's just left behind? Or is most of the research simply behind closed doors?

Jun 28, 2024 | mql5.com | Int OnInit |shashank rai |Check ForOpen |return spreadPips

TimeTokenizer (Custom Transformer): This class handles the time tokenization. It extracts the time column from the input tensor, converts it into a fractional representation of the day, and then combines it back with the other price data. DailyRollingNormalizer (Custom Transformer): This class performs the daily rolling normalization. It iterates through the price data, keeping track of the rolling maximum and minimum for each day. The prices are then normalized using these dynamic values.