Hugging Face's logo

Melady
/
TEMPO

Time Series Forecasting
Transformers
English
Time-series
foundation-model
forecasting
Inference Endpoints
Model card Files Community
TEMPO
File size: 3,033 Bytes 
4d5cbab
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b071939
 
 
 
d1c595b
b071939
5e03218
 
b071939
 
 
 
489d293
5e03218
b071939
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5e03218
b071939
 
 
 
 
 
 
 
 
 
5e03218
 
b071939
 
 
5e03218
b071939
 
 
 
489d293
5e03218
b071939
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
 
---
license: apache-2.0
datasets:
- ETDataset/ett
language:
- en
metrics:
- mse
- mae
library_name: transformers
pipeline_tag: time-series-forecasting
tags:
- Time-series
- foundation-model
- forecasting
---
# TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting

The official code for ICLR 2024 paper: "TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting (ICLR 2024)".

TEMPO is one of the very first open source **Time Series Foundation Models** for forecasting task v1.0 version.

![TEMPO-architecture](pics/TEMPO.png)



Please try our foundation model demo [[here]](https://4171a8a7484b3e9148.gradio.live).

![TEMPO-demo](pics/TEMPO_demo.jpg)


# Build the environment

```
conda create -n tempo python=3.8
```
```
conda activate tempo
```
```
pip install -r requirements.txt
```

# Get Data

   Download the data from [[Google Drive]](https://drive.google.com/drive/folders/13Cg1KYOlzM5C7K8gK8NfC-F3EYxkM3D2?usp=sharing) or [[Baidu Drive]](https://pan.baidu.com/s/1r3KhGd0Q9PJIUZdfEYoymg?pwd=i9iy), and place the downloaded data in the folder`./dataset`. You can also download the STL results from [[Google Drive]](https://drive.google.com/file/d/1gWliIGDDSi2itUAvYaRgACru18j753Kw/view?usp=sharing), and place the downloaded data in the folder`./stl`.

# Run TEMPO

## Training Stage
```
bash [ecl, etth1, etth2, ettm1, ettm2, traffic, weather].sh
```

## Test

After training, we can test TEMPO model under the zero-shot setting:

```
bash [ecl, etth1, etth2, ettm1, ettm2, traffic, weather]_test.sh
```
![TEMPO-results](pics/results.jpg)


# Pre-trained Models

You can download the pre-trained model from [[Google Drive]](https://drive.google.com/file/d/11Ho_seP9NGh-lQCyBkvQhAQFy_3XVwKp/view?usp=drive_link) and then run the test script for fun.

# Multi-modality dataset: TETS dataset

Here is the prompts use to generate the coresponding textual informaton of time series via [[OPENAI ChatGPT-3.5 API]](https://platform.openai.com/docs/guides/text-generation)

![TEMPO-prompt](pics/TETS_prompt.png)


The time series data are come from [[S&P 500]](https://www.spglobal.com/spdji/en/indices/equity/sp-500/#overview). Here is the EBITDA case for one company from the dataset:

![Company1_ebitda_summary](pics/Company1_ebitda_summary.png)


Example of generated contextual information for the Company marked above:

![Company1_ebitda_summary_words.jpg](pics/Company1_ebitda_summary_words.jpg)





You can download the processed data with text embedding from GPT2 from: [[TETS]](https://drive.google.com/file/d/1Hu2KFj0kp4kIIpjbss2ciLCV_KiBreoJ/view?usp=drive_link
).



## Cite
```
@inproceedings{
cao2024tempo,
title={{TEMPO}: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting},
author={Defu Cao and Furong Jia and Sercan O Arik and Tomas Pfister and Yixiang Zheng and Wen Ye and Yan Liu},
booktitle={The Twelfth International Conference on Learning Representations},
year={2024},
url={https://openreview.net/forum?id=YH5w12OUuU}
}
```