PyPOTS
PyPOTS copied to clipboard
WenjieDu
A Python toolkit/library for reality-centric machine/deep learning and data mining on partially-observed time series, including SOTA neural network models for scientific analysis tasks of imputation,...
### Issue description The current templates are old ones. Now the framework is updated, hence, templates have to be renewed as well.
### 1. Feature description For multi-task methods, they should be across tasks, rather than having separate implementation for each task. ### 2. Motivation This can help simplify the code. ###...
### 1. Feature description More useful and important info should be included and saved into model files, e.g. PyPOTS version, model hyperparameters, etc. ### 2. Motivation For example, the saved...
### 1. Feature description Remind users of the best model from which epoch after model training. ### 2. Motivation Make users be aware of the best model from which epoch....
### 1. Feature description Hi! Would it be possible to add an option to visualize the final (and intermediate) self-attention maps/matrices for the SAITS model? Thank you! ### 2. Motivation...
### 1. Feature description Changes to the classify() and forward() methods to make the models compatible with TimeSHAP and other XAI methods. ### 2. Motivation I would like to be...
### 1. Feature description To enable variable 'sequence length' of the input data. ### 2. Motivation Some of the input training data are composed of multiple concatenated time series of...
### Issue description 为了方便国内中文用户的使用以及中文的检索，我们需要一个中文版本的README
### 1. Model description Pyraformer should be included in PyPOTS, and could start from the imputation task. ```bibtex @inproceedings{liu2022pyraformer, title={Pyraformer: Low-Complexity Pyramidal Attention for Long-Range Time Series Modeling and Forecasting},...