Open gorold opened 3 years ago
@gorold Hello, Gerald Woo! I am very interested in your work. When will you open the complete code of ICLR 2022 Paper COST: CONTRASTIVE LEARNING OF DISENTANGLED SEASONAL-TREND REPRESENTATIONS FOR TIME SERIES FORECASTING
?
Hi @943fansi, thanks for your interest, perhaps you can drop me an email and I will get back to you from there, since this is not related to the TNC repo or this issue.
https://github.com/sanatonek/TNC_representation_learning/blob/7f5ac2dc99c412cb5cb9266793ec3f20809a8911/evaluations/classification_test.py#L160-L172
Hi Sana, would like to inform you of a potential issue in the testing code. The number of points being used from the test set is limited to the length of the training set. This could lead to some missing class labels in certain datasets.