search

lss-1138 / SegRNN

The official repository of the SegRNN paper: "Segment Recurrent Neural Network for Long-Term Time Series Forecasting." This work is developed by the Lab of Professor Weiwei Lin (linww@scut.edu.cn), South China University of Technology; Peng Cheng Laboratory.
Apache License 2.0
87 stars 15 forks source link

can I use this for multivariate time series classification? #4

Closed suojifeng closed 4 months ago

suojifeng commented 6 months ago

Hi author, Is it just encoder?thank you

lss-1138 commented 6 months ago

Hi, SegRNN is designed for long-term time series forecasting tasks. If you wish to use it for multivariate time series classification, some adjustments are necessary. For instance, you could replace the PMF module directly with a linear classification head.

Furthermore, I suggest referring to this framework (https://github.com/thuml/Time-Series-Library), which includes comprehensive implementations for forecasting, classification, anomaly detection, and other tasks. You might consider migrating the SegRNN to this repository to facilitate multivariate time series classification.