PyPOTS: A Python Toolbox for Data Mining on Partially-Observed Time Series

In this refactoring version, we

1. applied SAITS loss function to the newly added imputation models (Crossformer, PatchTST, DLinear, ETSformer, FEDformer, Informer, and Autoformer) in v0.4, and add the arguments MIT_weight and ORT_weight in them for users to balance the multi-task learning;
2. modularized all neural network models and put their modules in the package pypots.nn.modules;
3. removed deprecated metric funcs (e.g. pypots.utils.metrics.cal_mae that has been replaced by pypots.utils.metrics.calc_mae);

What's Changed

• Apply SAITS loss to newly added models and update the docs by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/346
• Modularize neural network models by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/348
• Modularize NN models, remove deprecated metric funcs, and update docs by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/349
• Remove pypots.imputation.locf.modules and add assertions for BTTF by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/350
• Test building package during CI by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/353
• Avoid the import error MessagePassing not defined by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/351

Full Changelog: https://github.com/WenjieDu/PyPOTS/compare/v0.4...v0.4.1