[ICML2024] Unified Training of Universal Time Series Forecasting Transformers
612
stars
49
forks
source link
Bug fix: MultiOutSizeLinear masks outputs to 0 when dimension of parameter >1 and dim * patch_size is not present in out_feat_size. #46
Closed
gorold closed 1 month ago
This PR fixes a major bug for the
MultiOutSizeLinear
class.When
dim > 1
in the following code snippet, https://github.com/SalesforceAIResearch/uni2ts/blob/c4d75a3053c14e27ac4d875d7b31987abc1090a1/src/uni2ts/distribution/_base.py#L74-L85 it causesfor feat_size in self.out_features_ls
bedim * patch_size
, whereasout_feat_size
contains the patch sizes. This means that certain patch sizes could be ignored. https://github.com/SalesforceAIResearch/uni2ts/blob/c4d75a3053c14e27ac4d875d7b31987abc1090a1/src/uni2ts/module/ts_embed.py#L172-L178For the current configuration, this means that outputs with patch sizes 8 and 16 are always masked to 0.
Closes #44