#timeseries

Ekambaram, Vijay, Arindam Jati, Pankaj Dayama, Sumanta Mukherjee, Nam H. Nguyen, Wesley M. Gifford, Chandra Reddy, and Jayant Kalagnanam. 2024. “Tiny Time Mixers (TTMs): Fast Pre-Trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series.” arXiv [Cs.LG]. arXiv. http://arxiv.org/abs/2401.03955.

arXiv.org

Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced...

Large pre-trained models excel in zero/few-shot learning for language and vision tasks but face challenges in multivariate time series (TS) forecasting due to diverse data characteristics....

#timeseries

Another work on pre-trained large models for time series.

https://github.com/DAMO-DI-ML/NeurIPS2023-One-Fits-All

GitHub

GitHub - DAMO-DI-ML/NeurIPS2023-One-Fits-All: The official code for "One Fits All: Power General Time Series Analysis by Pretrained…

The official code for "One Fits All: Power General Time Series Analysis by Pretrained LM (NeurIPS 2023 Spotlight)" - DAMO-DI-ML/NeurIPS2023-One-Fits-All

#timeseries #ai

https://lu.ma/timegpt

And demo:

https://github.com/Nixtla/nixtla

Luma

TimeGPT Launch Announcement · Luma

Introducing TimeGPT Beta
We recognize that the essence of knowledge lies in uncertainty reduction - the continuous pursuit to anticipate and plan for what's…

#timeseries

Finding a suitable forecasting metric to evaluate the forecasting models is often the key to a forecasting project. Right? We use metrics when developing models, we also use metrics to monitor models.

There are a bunch of metrics people choose from or adapt from. To be faster when choosing and adapting metrics, I created a page on the properties of different metrics for time series forecasting problems. For reproducibility, I also included all the code used to write this page.

https://dl.leima.is/time-series/timeseries-metrics.forecasting/

dl.leima.is

Time Series Forecasting Metrics - Time Series with Deep Learning Quick Bite