Skip to content
forked from daochenzha/ltsm

Understanding Different Design Choices in Training Large Time Series Models

Notifications You must be signed in to change notification settings

zxr-creator/ltsm

 
 

Repository files navigation

Understanding Different Design Choices in Training Large Time Series Models

This work investigates the transition from traditional Time Series Forecasting (TSF) to Large Time Series Models (LTSMs), leveraging universal transformer-based models. Training LTSMs on diverse time series data introduces challenges due to varying frequencies, dimensions, and patterns. We explore various design choices for LTSMs, including pre-processing, model configurations, and dataset setups. We introduce Time Series Prompt, a statistical prompting strategy, and $\texttt{LTSM-bundle}$, which encapsulates the most effective design practices identified. $\texttt{LTSM-bundle}$ is developed by Data Lab at Rice University.

Resources

🌟 Please star our repo to follow the latest updates on LTSM-bundle!

📣 We have released our paper and source code of LTSM-bundle-v1.0!

📚 Follow our latest English Tutorial or 中文教程 to costomize your LTSM!

🌎 For more information, please visit:

Why LTSM-bundle?

The LTSM-bundle package leverages the HuggingFace transformers toolkit, offering flexibility to switch between different advanced language models as the backbone. It is easy to tailor the general LTSMs to their specific time series forecasting needs by selecting the most suitable language model from a wide array of options. The flexibility enhances the adaptability of the package across different industries and data types, ensuring optimal performance in diverse scenarios.

Installation

conda create -n ltsm python=3.8.0
conda activate ltsm
git clone git@github.com:daochenzha/ltsm.git
cd ltsm
pip3 install -e .
pip3 install -r requirements.txt

Quick Exploration on LTSM-bundle

Training on [Time Series Prompt] and [Linear Tokenization]

bash scripts/train_ltsm_csv.sh

Training on [Text Prompt] and [Linear Tokenization]

bash scripts/train_ltsm_textprompt_csv.sh

Training on [Time Series Prompt] and [Time Series Tokenization]

bash scripts/train_ltsm_tokenizer_csv.sh

Datasets and Time Series Prompts

Download the datasets

cd datasets
download: https://drive.google.com/drive/folders/1hLFbz0FRxdiDCzgFYtKCOPJYSBVvwW9P

Download the time series prompts

cd prompt_bank/propmt_data_csv
download: https://drive.google.com/drive/folders/1hLFbz0FRxdiDCzgFYtKCOPJYSBVvwW9P

Cite This Work

If you find this work useful, you may cite this work:

@article{ltsm-bundle,
  title={Understanding Different Design Choices in Training Large Time Series Models},
  author={Chuang*, Yu-Neng and Li*, Songchen and Yuan*, Jiayi and Wang*, Guanchu and Lai*, Kwei-Herng and Yu, Leisheng and Ding, Sirui and Chang, Chia-Yuan and Tan, Qiaoyu and Zha, Daochen and Hu, Xia},
  journal={arXiv preprint arXiv:2406.14045},
  year={2024}
}

About

Understanding Different Design Choices in Training Large Time Series Models

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 97.2%
  • Shell 2.8%