Time-LLM is a reprogramming framework to repurpose LLMs for general time series forecasting with the backbone language models kept intact. Essentially, it uses a foundation model that is pretrained on language and appends some prompting and finetuning architectures to the input and output stages for the purpose of analyzing time series data. For NESL, we're interested in examining how different LLM backbones can be swapped in and their relative performances. Right now, I've swapped in Mamba1 in place of the GPT transformer backbone, but we'll look at converting this into other versions in the future if necessary. Currently, we may focus our priorities on full training of different architectures rather than using this method of finetuning models.
- Time-LLM comprises two key components: (1) reprogramming the input time series into text prototype representations that are more natural for the LLM, and (2) augmenting the input context with declarative prompts (e.g., domain expert knowledge and task instructions) to guide LLM reasoning.
Use python 3.11 from MiniConda
- torch==2.2.2
- accelerate==0.28.0
- einops==0.7.0
- matplotlib==3.7.0
- numpy==1.23.5
- pandas==1.5.3
- scikit_learn==1.2.2
- scipy==1.12.0
- tqdm==4.65.0
- peft==0.4.0
- transformers==4.31.0
- deepspeed==0.14.0
- sentencepiece==0.2.0
To install all dependencies:
pip install -r requirements.txt
You can access the well pre-processed datasets from [Google Drive], then place the downloaded contents under ./dataset
You can access the Google Colab jupyter notebook for this project here.
- Download datasets and place them under
./dataset - Tune the model. We provide five experiment scripts for demonstration purpose under the folder
./scripts. For example, you can evaluate on ETT datasets by:
bash ./scripts/TimeLLM_ETTh1.sh bash ./scripts/TimeLLM_ETTh2.sh bash ./scripts/TimeLLM_ETTm1.sh bash ./scripts/TimeLLM_ETTm2.shPlease refer to run_main.py, run_m4.py and run_pretrain.py for the detailed description of each hyperparameter.
1, Position Paper: What Can Large Language Models Tell Us about Time Series Analysis, in ICML 2024.
Authors: Ming Jin, Yifan Zhang, Wei Chen, Kexin Zhang, Yuxuan Liang*, Bin Yang, Jindong Wang, Shirui Pan, Qingsong Wen*
@inproceedings{jin2024position,
title={Position Paper: What Can Large Language Models Tell Us about Time Series Analysis},
author={Ming Jin and Yifan Zhang and Wei Chen and Kexin Zhang and Yuxuan Liang and Bin Yang and Jindong Wang and Shirui Pan and Qingsong Wen},
booktitle={International Conference on Machine Learning (ICML 2024)},
year={2024}
}2, Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlook, in arXiv 2023. [GitHub Repo]
Authors: Ming Jin, Qingsong Wen*, Yuxuan Liang, Chaoli Zhang, Siqiao Xue, Xue Wang, James Zhang, Yi Wang, Haifeng Chen, Xiaoli Li (IEEE Fellow), Shirui Pan*, Vincent S. Tseng (IEEE Fellow), Yu Zheng (IEEE Fellow), Lei Chen (IEEE Fellow), Hui Xiong (IEEE Fellow)
@article{jin2023lm4ts,
title={Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlook},
author={Ming Jin and Qingsong Wen and Yuxuan Liang and Chaoli Zhang and Siqiao Xue and Xue Wang and James Zhang and Yi Wang and Haifeng Chen and Xiaoli Li and Shirui Pan and Vincent S. Tseng and Yu Zheng and Lei Chen and Hui Xiong},
journal={arXiv preprint arXiv:2310.10196},
year={2023}
}3, Transformers in Time Series: A Survey, in IJCAI 2023. [GitHub Repo]
Authors: Qingsong Wen, Tian Zhou, Chaoli Zhang, Weiqi Chen, Ziqing Ma, Junchi Yan, Liang Sun
@inproceedings{wen2023transformers,
title={Transformers in time series: A survey},
author={Wen, Qingsong and Zhou, Tian and Zhang, Chaoli and Chen, Weiqi and Ma, Ziqing and Yan, Junchi and Sun, Liang},
booktitle={International Joint Conference on Artificial Intelligence(IJCAI)},
year={2023}
}4, TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting, in ICLR 2024. [GitHub Repo]
Authors: Shiyu Wang, Haixu Wu, Xiaoming Shi, Tengge Hu, Huakun Luo, Lintao Ma, James Y. Zhang, Jun Zhou
@inproceedings{wang2023timemixer,
title={TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting},
author={Wang, Shiyu and Wu, Haixu and Shi, Xiaoming and Hu, Tengge and Luo, Huakun and Ma, Lintao and Zhang, James Y and ZHOU, JUN},
booktitle={International Conference on Learning Representations (ICLR)},
year={2024}
}Our implementation adapts Time-Series-Library and OFA (GPT4TS) as the code base and have extensively modified it to our purposes. We thank the authors for sharing their implementations and related resources.