WenjieDu / PyPOTS / 8614976540

80%

main: 80%

push

github

Apply SAITS embedding strategy to new added models (#343)

* feat: make Crossformer, PatchTST, DLinear, ETSformer, FEDformer, Informer, Autoformer take the missing mask as a part of input;

* refactor: separate n_steps of K and Q, this is necessary for Crossformer;

79 of 80 new or added lines in 10 files covered. (98.75%)

6847 of 8450 relevant lines covered (81.03%)

4.85 hits per line