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

1. apply the SAITS embedding strategy to models Crossformer, PatchTST, DLinear, ETSformer, FEDformer, Informer, and Autoformer to make them applicable to POTS data as imputation methods;
2. fix a bug in USGAN loss function;
3. gather several Transformer embedding methods into the packagepypots.nn.modules.transformer.embedding;
4. add the attribute best_epoch for NN models to record the best epoch num and log it after model training;
5. make the self-attention operator replaceable in the class MultiHeadAttention for Transformer models;
6. rename the argument d_inner of all models in previous versions into d_ffn. This is for unified argument naming and easier understanding;
7. remove deprecated functions save_model() and load_model() in all NN model classes, which are now replaced by save()and load();

What's Changed

• Removing deprecated functions by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/318
• Add Autoformer as an imputation model by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/320
• Removing deprecated save_model and load_model, adding the imputation model Autoformer by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/321
• Simplify MultiHeadAttention by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/322
• Add PatchTST as an imputation model by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/323
• Renaming d_inner into d_ffn by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/325
• Adding PatchTST, renaming d_innner into d_ffn, and refactoring Autofomer by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/326
• Add DLinear as an imputation model by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/327
• Add ETSformer as an imputation model by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/328
• Add Crossformer as an imputation model by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/329
• Add FEDformer as an imputation model by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/330
• Add Crossformer, Autoformer, PatchTST, DLinear, ETSformer, FEDformer as imputation models by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/331
• Refactor embedding package, remove the unused part in Autoformer, and update the docs by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/332
• Make the self-attention operator replaceable in Transformer by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/334
• Add informer as an imputation model by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/335
• Speed up testing procedure by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/336
• Add Informer, speed up CI testing, and make self-attention operator replaceable by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/337
• debug USGAN by @AugustJW in https://github.com/WenjieDu/PyPOTS/pull/339
• Fix USGAN loss function, and update the docs by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/340
• Add the attribute best_epoch to record the best epoch num by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/342
• Apply SAITS embedding strategy to new added models by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/343
• Release v0.4, apply SAITS embedding strategy to the newly added models, and update README by @WenjieDu in https://github.com/WenjieDu/PyPOTS/pull/344

Full Changelog: https://github.com/WenjieDu/PyPOTS/compare/v0.3.2...v0.4