We're thrilled to announce that #tsai has just crossed the 3k stars milestone!🚀
Started as a spin-off project during one of @jeremyphoward's fantastic fastai courses and is gaining momentum! Join us on this exciting journey! #timeseries#deeplearning https://t.co/wej9B6WqLc
📢New ICLR23 research paper introduces PatchTST, a state-of-the-art transformer model for multivariate #timeseries#forecasting. It processes each channel independently, breaking it into patches. Access a tutorial notebook to learn more here https://t.co/tiHybZqUWy #deeplearning
Good news for time series enthusiasts!
Our deep learning open-source library, tsai, is getting a major upgrade over the next few weeks. We're enhancing our forecasting functionality to help you build more accurate predictions.
To help you take full advantage of these new features, we'll be sharing blog posts and notebooks on topics like:
✅ data preparation
✅ building a baseline
✅ forecasting with univariate or multivariate time series
✅ long-term forecasting
✅ forecasting with panel data
and more
Today the https://t.co/wej9B6WYAK repository has reached 1k commits. I want to thank all the users and contributors for making it possible, as well as @jeremyphoward and the fastai community for their work on fastai on which tsai is based.
The tsai deep learning time series library from @TimeseriesAI (https://t.co/wej9B6WqLc) has hit 1k stars this week. It is built on top of @Pytorch / @fastdotai. I am overwhelmed by the support shown by so many users 🤩
I just published a blog post that summarizes the best ideas from the top 15 teams (gold medal winners) #kaggle's time series competition (VPP)
https://t.co/erZ33mKwxE
@timeseriesAI