Are Transformers Effective for Time Series Forecasting?| Pit filling

The paper : Are Transformers Effective for Time Series Forecasting?

Code ：https://github.com/cure-lab/DLinear

---

        As eye-catching as this title ,Transformer Is it really effective in time series prediction ？ This paper is in Abstract I doubt it Transformer-based Of TSF The relatively high accuracy of the solution is not the ability of its structure to extract temporal features , On the contrary, I think so DMS(Direct-multi-step) The credit of prediction strategy , Then a simple model is proposed Dlinear（Decomposition Linear Model）, To be honest, I see all kinds of former Deformation of , It's really good to publish such an article , Borrow a netizen's comment , A heavy sword has no edge , It's a coincidence , The work of this paper is really good , Several students from the Chinese University of Hong Kong .

---

Transformer Is it really effective for time series prediction tasks ？ - You know // Chinese translation has its own views , The comment area is also very exciting

Are Transformers Effective for Time Series Forecasting?_ Bili, Bili _bilibili// This friend from Shenzhen University spoke very well , More basic

Explanation of academic terms in some papers  

【 Deep learning 】 About Permutation Invariant The explanation of _Yo3ngLau The blog of -CSDN Blog