feat: 🎸 add six LTSF baselines (thanks to @superarthurlx)

Author: zezhishao

README.md

@@ -40,11 +40,12 @@
 
 $\text{BasicTS}^{+}$ (**Basic** **T**ime **S**eries) is a benchmark library and toolkit designed for time series forecasting. It now supports a wide range of tasks and datasets, including spatial-temporal forecasting and long-term time series forecasting. It covers various types of algorithms such as statistical models, machine learning models, and deep learning models, making it an ideal tool for developing and evaluating time series forecasting models. You can find detailed tutorials in [Getting Started](./tutorial/getting_started.md).
 
+🎉 **Update (June 2025):** Adds six LTSF baselines: CARD, TimeXer, Bi-Mamba, ModernTCN, S-D-Mamba, and S4.
+
 🎉 **Update (May 2025):** BasicTS now supports training universal forecasting models—such as **TimeMoE** and **ChronosBolt**—with the [BLAST](https://arxiv.org/abs/2505.17871) corpus. BLAST enables **faster convergence**, **notable reductions in computational cost**, and superior performance even with limited resources. See [here](./tutorial/training_with_BLAST.md).  
 
 If you find this project helpful, please don't forget to give it a ⭐ Star to show your support. Thank you!
 
-
 > [!IMPORTANT]
 > If you find this repository helpful for your work, please consider citing the following benchmarking paper:
 >
@@ -148,6 +149,11 @@ MOIRAI (inference) | Unified Training of Universal Time Series Forecasting Trans
 
 | 📊Baseline    | 📝Title                                                                                                  | 📄Paper                                                | 💻Code                                                                        | 🏛Venue    | 🎯Task |
 | :------------ | :------------------------------------------------------------------------------------------------------- | :----------------------------------------------------- | :---------------------------------------------------------------------------- | :--------- | :----- |
+| S-D-Mamba         | Is Mamba Effective for Time Series Forecasting?                           | [Link](https://arxiv.org/abs/2403.11144v3)                  | [Link](https://github.com/wzhwzhwzh0921/S-D-Mamba)                                        | NeuroComputing'24 | LTSF   |
+| Bi-Mamba         | Bi-Mamba+: Bidirectional Mamba for Time Series Forecasting                           | [Link](https://arxiv.org/abs/2404.15772)                  | [Link](https://github.com/Leopold2333/Bi-Mamba4TS)                                        | arXiv'24 | LTSF   |
+| ModernTCN         | ModernTCN: A Modern Pure Convolution Structure for General Time Series Analysis                           | [Link](https://openreview.net/forum?id=vpJMJerXHU)                  | [Link](https://github.com/luodhhh/ModernTCN)                                        | ICLR'24 | LTSF   |
+| TimeXer         | TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variables                           | [Link](https://arxiv.org/abs/2402.19072)                  | [Link](https://github.com/thuml/TimeXer)                                        | NeurIPS'24 | LTSF   |
+| CARD         | CARD: Channel Aligned Robust Blend Transformer for Time Series Forecasting                           | [Link](https://arxiv.org/abs/2305.12095)                  | [Link](https://github.com/wxie9/CARD)                                        | ICLR'24 | LTSF   |
 | SOFTS         | SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core Fusion                           | [Link](https://arxiv.org/pdf/2404.14197)                  | [Link](https://github.com/Secilia-Cxy/SOFTS)                                        | NeurIPS'24 | LTSF   |
 | CATS          | Are Self-Attentions Effective for Time Series Forecasting?                                               | [Link](https://arxiv.org/pdf/2405.16877)                  | [Link](https://github.com/dongbeank/CATS)                                        | NeurIPS'24 | LTSF   |
 | Sumba         | Structured Matrix Basis for Multivariate Time Series Forecasting with Interpretable Dynamics             | [Link](https://xiucheng.org/assets/pdfs/nips24-sumba.pdf) | [Link](https://github.com/chenxiaodanhit/Sumba/)                                 | NeurIPS'24 | LTSF   |
@@ -172,6 +178,7 @@ MOIRAI (inference) | Unified Training of Universal Time Series Forecasting Trans
 | NHiTS         | Neural Hierarchical Interpolation for Time Series Forecasting                                            | [Link](https://arxiv.org/abs/2201.12886)                  | [Link](https://github.com/cchallu/n-hits)                                        | AAAI'23    | LTSF   |
 | PatchTST      | A Time Series is Worth 64 Words: Long-term Forecasting with Transformers                                 | [Link](https://arxiv.org/abs/2211.14730)                  | [Link](https://github.com/yuqinie98/PatchTST)                                    | ICLR'23    | LTSF   |
 | TiDE          | Long-term Forecasting with TiDE: Time-series Dense Encoder                                               | [Link](https://arxiv.org/abs/2304.08424)                  | [Link](https://github.com/lich99/TiDE)                                           | TMLR'23    | LTSF   |
+| S4         | Efficiently Modeling Long Sequences with Structured State Spaces                           | [Link](https://openreview.net/pdf?id=uYLFoz1vlAC)                  | [Link](https://github.com/state-spaces/s4)                                        | ICLR'22 | LTSF   |
 | TimesNet      | Temporal 2D-Variation Modeling for General Time Series Analysis                                          | [Link](https://openreview.net/pdf?id=ju_Uqw384Oq)         | [Link](https://github.com/thuml/TimesNet)                                        | ICLR'23    | LTSF   |
 | Triformer     | Triangular, Variable-Specific Attentions for Long Sequence Multivariate Time Series Forecasting          | [Link](https://arxiv.org/abs/2204.13767)                  | [Link](https://github.com/razvanc92/triformer)                                   | IJCAI'22   | LTSF   |
 | NSformer      | Exploring the Stationarity in Time Series Forecasting                                                    | [Link](https://arxiv.org/abs/2205.14415)                  | [Link](https://github.com/thuml/Nonstationary_Transformers)                      | NeurIPS'22 | LTSF   |
@@ -289,9 +296,9 @@ Thanks goes to these wonderful people ([emoji key](https://allcontributors.org/d
 
 This project follows the [all-contributors](https://github.com/all-contributors/all-contributors) specification. Contributions of any kind welcome!
 
-## ⭐ Star History
+<!-- ## ⭐ Star History
 
-[![Star History Chart](https://api.star-history.com/svg?repos=GestaltCogTeam/BasicTS&type=Date)](https://star-history.com/#GestaltCogTeam/BasicTS&Date)
+[![Star History Chart](https://api.star-history.com/svg?repos=GestaltCogTeam/BasicTS&type=Date)](https://star-history.com/#GestaltCogTeam/BasicTS&Date) -->
 
 ## 🔗 Acknowledgement
 

README_CN.md

@@ -41,6 +41,8 @@
 
 $\text{BasicTS}^{+}$ (**Basic** **T**ime **S**eries) 是一个面向时间序列预测的基准库和工具箱，现已支持时空预测、长序列预测等多种任务与数据集，涵盖统计模型、机器学习模型、深度学习模型等多类算法，为开发和评估时间序列预测模型提供了理想的工具。你可以在[快速上手](./tutorial/getting_started_cn.md)找到详细的教程。
 
+🎉 **更新（2025年6月）：** 添加了6个长序列预测基线：CARD、TimeXer、Bi-Mamba、ModernTCN、S-D-Mamba、S4。
+
 🎉 **更新（2025年5月）：** BasicTS 现已支持使用 [BLAST](https://arxiv.org/abs/2505.17871) 语料库训练通用预测模型（例如 **TimeMoE** 和 **ChronosBolt**）。BLAST 能够实现 **更快的收敛速度**、**显著降低计算成本**，并且即使在资源有限的情况下也能获得卓越性能。[查看](./tutorial/training_with_BLAST_cn.md)。
 
 如果你觉得这个项目对你有帮助，别忘了给个⭐Star支持一下，非常感谢！
@@ -149,6 +151,11 @@ MOIRAI (inference) | Unified Training of Universal Time Series Forecasting Trans
 
 | 📊Baseline    | 📝Title                                                                                                  | 📄Paper                                                | 💻Code                                                                        | 🏛Venue    | 🎯Task |
 | :------------ | :------------------------------------------------------------------------------------------------------- | :----------------------------------------------------- | :---------------------------------------------------------------------------- | :--------- | :----- |
+| S-D-Mamba         | Is Mamba Effective for Time Series Forecasting?                           | [Link](https://arxiv.org/abs/2403.11144v3)                  | [Link](https://github.com/wzhwzhwzh0921/S-D-Mamba)                                        | NeuroComputing'24 | LTSF   |
+| Bi-Mamba         | Bi-Mamba+: Bidirectional Mamba for Time Series Forecasting                           | [Link](https://arxiv.org/abs/2404.15772)                  | [Link](https://github.com/Leopold2333/Bi-Mamba4TS)                                        | arXiv'24 | LTSF   |
+| ModernTCN         | ModernTCN: A Modern Pure Convolution Structure for General Time Series Analysis                           | [Link](https://openreview.net/forum?id=vpJMJerXHU)                  | [Link](https://github.com/luodhhh/ModernTCN)                                        | ICLR'24 | LTSF   |
+| TimeXer         | TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variables                           | [Link](https://arxiv.org/abs/2402.19072)                  | [Link](https://github.com/thuml/TimeXer)                                        | NeurIPS'24 | LTSF   |
+| CARD         | CARD: Channel Aligned Robust Blend Transformer for Time Series Forecasting                           | [Link](https://arxiv.org/abs/2305.12095)                  | [Link](https://github.com/wxie9/CARD)                                        | ICLR'24 | LTSF   |
 | SOFTS         | SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core Fusion                           | [Link](https://arxiv.org/pdf/2404.14197)                  | [Link](https://github.com/Secilia-Cxy/SOFTS)                                        | NeurIPS'24 | LTSF   |
 | CATS          | Are Self-Attentions Effective for Time Series Forecasting?                                               | [Link](https://arxiv.org/pdf/2405.16877)                  | [Link](https://github.com/dongbeank/CATS)                                        | NeurIPS'24 | LTSF   |
 | Sumba         | Structured Matrix Basis for Multivariate Time Series Forecasting with Interpretable Dynamics             | [Link](https://xiucheng.org/assets/pdfs/nips24-sumba.pdf) | [Link](https://github.com/chenxiaodanhit/Sumba/)                                 | NeurIPS'24 | LTSF   |
@@ -173,6 +180,7 @@ MOIRAI (inference) | Unified Training of Universal Time Series Forecasting Trans
 | NHiTS         | Neural Hierarchical Interpolation for Time Series Forecasting                                            | [Link](https://arxiv.org/abs/2201.12886)                  | [Link](https://github.com/cchallu/n-hits)                                        | AAAI'23    | LTSF   |
 | PatchTST      | A Time Series is Worth 64 Words: Long-term Forecasting with Transformers                                 | [Link](https://arxiv.org/abs/2211.14730)                  | [Link](https://github.com/yuqinie98/PatchTST)                                    | ICLR'23    | LTSF   |
 | TiDE          | Long-term Forecasting with TiDE: Time-series Dense Encoder                                               | [Link](https://arxiv.org/abs/2304.08424)                  | [Link](https://github.com/lich99/TiDE)                                           | TMLR'23    | LTSF   |
+| S4         | Efficiently Modeling Long Sequences with Structured State Spaces                           | [Link](https://openreview.net/pdf?id=uYLFoz1vlAC)                  | [Link](https://github.com/state-spaces/s4)                                        | ICLR'22 | LTSF   |
 | TimesNet      | Temporal 2D-Variation Modeling for General Time Series Analysis                                          | [Link](https://openreview.net/pdf?id=ju_Uqw384Oq)         | [Link](https://github.com/thuml/TimesNet)                                        | ICLR'23    | LTSF   |
 | Triformer     | Triangular, Variable-Specific Attentions for Long Sequence Multivariate Time Series Forecasting          | [Link](https://arxiv.org/abs/2204.13767)                  | [Link](https://github.com/razvanc92/triformer)                                   | IJCAI'22   | LTSF   |
 | NSformer      | Exploring the Stationarity in Time Series Forecasting                                                    | [Link](https://arxiv.org/abs/2205.14415)                  | [Link](https://github.com/thuml/Nonstationary_Transformers)                      | NeurIPS'22 | LTSF   |
@@ -289,9 +297,9 @@ BasicTS 支持多种类型的数据集，涵盖时空预测、长序列预测及
 
 此项目遵循 [all-contributors](https://github.com/all-contributors/all-contributors) 规范。欢迎任何形式的贡献！
 
-## ⭐ Star History
+<!-- ## ⭐ Star History
 
-[![Star History Chart](https://api.star-history.com/svg?repos=GestaltCogTeam/BasicTS&type=Date)](https://star-history.com/#GestaltCogTeam/BasicTS&Date)
+[![Star History Chart](https://api.star-history.com/svg?repos=GestaltCogTeam/BasicTS&type=Date)](https://star-history.com/#GestaltCogTeam/BasicTS&Date) -->
 
 ## 🔗 致谢
 

baselines/BiMamba/README.md

@@ -1,4 +1 @@
-BiMamba additionally requires the following dependency:
-```bash
-pip install mamba_ssm
-````
+BiMamba requires `mamba_ssm` package. Please refer to the [mamba_ssm documentation](https://github.com/state-spaces/mamba).

baselines/S_Mamba/ETTm2.py

@@ -149,5 +149,4 @@
 CFG.EVAL = EasyDict()
 
 # Evaluation parameters
-CFG.EVAL.HORIZONS = [12, 24, 48, 96] 
 CFG.EVAL.USE_GPU = True # Whether to use GPU for evaluation. Default: True

baselines/S_Mamba/Electricity.py

@@ -149,5 +149,4 @@
 CFG.EVAL = EasyDict()
 
 # Evaluation parameters
-CFG.EVAL.HORIZONS = [12, 24, 48, 96] 
 CFG.EVAL.USE_GPU = True # Whether to use GPU for evaluation. Default: True

baselines/S_Mamba/README.md

@@ -0,0 +1 @@
+S-D-Mamba requires `mamba_ssm` package. Please refer to the [mamba_ssm documentation](https://github.com/state-spaces/mamba).

baselines/S_Mamba/Weather.py

@@ -149,5 +149,4 @@
 CFG.EVAL = EasyDict()
 
 # Evaluation parameters
-CFG.EVAL.HORIZONS = [12, 24, 48, 96] 
 CFG.EVAL.USE_GPU = True # Whether to use GPU for evaluation. Default: True

baselines/TimeXer/ETTm2.py

@@ -147,5 +147,4 @@
 CFG.EVAL = EasyDict()
 
 # Evaluation parameters
-CFG.EVAL.HORIZONS = [12, 24, 48, 96]
 CFG.EVAL.USE_GPU = True # Whether to use GPU for evaluation. Default: True

baselines/TimeXer/Electricity.py

@@ -147,5 +147,4 @@
 CFG.EVAL = EasyDict()
 
 # Evaluation parameters
-CFG.EVAL.HORIZONS = [12, 24, 48, 96]
 CFG.EVAL.USE_GPU = True # Whether to use GPU for evaluation. Default: True

baselines/TimeXer/Weather.py

@@ -147,5 +147,4 @@
 CFG.EVAL = EasyDict()
 
 # Evaluation parameters
-CFG.EVAL.HORIZONS = [12, 24, 48, 96]
 CFG.EVAL.USE_GPU = True # Whether to use GPU for evaluation. Default: True