
Chronos: Learning the Language of Time Series with Abdul Fatir Ansari - #685
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Tokenizing Time Series Data
This chapter explores a straightforward tokenization scheme that applies natural language processing methodologies to time series forecasting. It discusses normalization, quantization, and the selection of the T5 model, highlighting the effectiveness of using uniformly spaced bins over more complex methods.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.