from Hacker News

Chronos: Pretrained (Language) Models for Probabilistic Time Series Forecasting

by stellalo on 3/17/24, 12:10 PM with 3 comments

• by huibin_shen on 3/17/24, 3:44 PM

  On a large scale of 42 time series datasets, Chronos demonstrates impressive empirical performance. In the zero-shot setting, it matches or even outperforms many baselines which are trained on the dataset.

• by sebastianpineda on 3/17/24, 6:18 PM

  this is an amazing contribution for the community!

• by cs702 on 3/17/24, 1:07 PM

  TL;DR:

  * Scale the time series data and quantize the floating point values into B bins.

  * Each bin becomes a corresponding token id in a vocabulary of B embeddings.

  * Train a small LLM to predict the next token id given a sequence of token ids.

  * At each time step, the LLM gives you a probability distribution over B bins.