https://github.com/Atik-Ahamed/TimeMachine?tab=readme-ov-file

TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecasting
https://github.com/Atik-Ahamed/TimeMachine?tab=readme-ov-file

time-series-forecasting

Last synced: 3 months ago
JSON representation

TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecasting

• Host: GitHub
• URL: https://github.com/Atik-Ahamed/TimeMachine?tab=readme-ov-file
• Owner: Atik-Ahamed
• License: apache-2.0
• Created: 2024-03-19T00:24:37.000Z (10 months ago)
• Default Branch: main
• Last Pushed: 2024-07-27T03:51:24.000Z (6 months ago)
• Last Synced: 2024-08-02T04:02:55.045Z (5 months ago)
• Topics: time-series-forecasting
• Language: Python
• Homepage: https://arxiv.org/abs/2403.09898
• Size: 230 KB
• Stars: 126
• Watchers: 4
• Forks: 17
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• Awesome-state-space-models - Time Series

README

        
# TimeMachine

![Alt text](./pics/TimeMachine.PNG)

### Welcome to the official repository of: [TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecasting](https://arxiv.org/pdf/2403.09898.pdf). 

## :triangular_flag_on_post: TimeMachine is accepted to [**ECAI**](https://www.ecai2024.eu/) 

## Usage

1. Install requirements. ```pip install -r requirements.txt```

2. Navigate through our example scripts located at ```./scripts/TimeMachine```. You'll find the core of TimeMachine in ```models/TimeMachine.py```. For example, to get the multivariate forecasting results for weather dataset, just run the following command, and you can open ```./result.txt``` to see the results once the training is completed. Moreover, the results will also be available at ```csv_results```, which can be utilized to make queries in the dataframe:

```

sh ./scripts/TimeMachine/weather.sh

```

Hyper-paramters can be tuned based upon needs (e.g. different look-back windows and prediction lengths). TimeMachine is built on the popular [PatchTST](https://github.com/yuqinie98/PatchTST) framework.

## Acknowledgement

We are deeply grateful for the valuable code and efforts contributed by the following GitHub repositories. Their contributions have been immensely beneficial to our work.

- Mamba (https://github.com/state-spaces/mamba)

- PatchTST (https://github.com/yuqinie98/PatchTST)

- iTransformer (https://github.com/thuml/iTransformer)

- RevIN (https://github.com/ts-kim/RevIN)

- Reformer (https://github.com/lucidrains/reformer-pytorch)

- Informer (https://github.com/zhouhaoyi/Informer2020)

- FlashAttention (https://github.com/shreyansh26/FlashAttention-PyTorch)

- Autoformer (https://github.com/thuml/Autoformer)

- Stationary (https://github.com/thuml/Nonstationary_Transformers)

- Time-Series-Library (https://github.com/thuml/Time-Series-Library)

## Citation

If you find this repo useful in your research, please consider citing our paper as follows:

```

@article{timemachine,

  title     = {TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecasting},

  author    = {Ahamed, Md Atik and Cheng, Qiang},

  journal   = {arXiv preprint arXiv:2403.09898},

  year      = {2024}

}

```