An offical implementation of PatchTST: "A Time Series is Worth 64 Words: Long-term Forecasting with Transformers." (ICLR 2023) https://arxiv.org/abs/2211.14730
Hi, thanks for the work. After digging into the code, I found that res_attention parameter is set to False in PatchTST_self_supervised setting while True in PatchTST_supervised setting. Could you please share some insights behind this? Thanks!
Hi, thanks for the work. After digging into the code, I found that![image](https://github.com/yuqinie98/PatchTST/assets/38466901/c30b2a51-4c97-40a5-8559-25c1bd4b12f3)
res_attention
parameter is set toFalse
inPatchTST_self_supervised
setting whileTrue
inPatchTST_supervised
setting. Could you please share some insights behind this? Thanks!