FilterNet (NeurIPS 2024)

The repo is the official implementation for the paper: "FilterNet: Harnessing Frequency Filters for Time Series Forecasting".

Getting Started

1、Environment Requirements

To get started, ensure you have Conda installed on your system and follow these steps to set up the environment:

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

2、Download Data

All the datasets needed for FilterNet can be obtained from the Google Drive provided in Autoformer.

3、Training Example

For datasets with a small number of variables, such as ETTh, ETTm, and Exchange, we recommend using PaiFilter as follows:

bash ./scripts/PaiFilter/ETTm1.sh
bash ./scripts/PaiFilter/ETTm2.sh
bash ./scripts/PaiFilter/ETTh2.sh

For datasets with a large number of variables such as ECL, Traffic, and weather, it is recommended to use TexFilter as follows:

bash ./scripts/PaiFilter/ECL.sh
bash ./scripts/PaiFilter/Traffic.sh
bash ./scripts/PaiFilter/Weather.sh

Our Other Work about Learning in the Frequency Domain for Time Series Analysis

🚩 [IJCAI 2024]: Deep Frequency Derivative Learning for Non-stationary Time Series Forecasting

🚩 [NeurIPS 2023]: Frequency-domain MLPs are more effective learners in time series forecasting

🚩 [NeurIPS 2023]: FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective

🚩 [arXiv]: A Survey on Deep Learning based Time Series Analysis with Frequency Transformation

Acknowledgement

We appreciate the following GitHub repositories for providing valuable code bases and datasets:

https://github.com/wanghq21/MICN

https://github.com/thuml/TimesNet

https://github.com/aikunyi/FreTS

https://github.com/VEWOXIC/FITS

https://github.com/plumprc/RTSF

https://github.com/cure-lab/LTSF-Linear

https://github.com/zhouhaoyi/Informer2020

https://github.com/thuml/Autoformer

https://github.com/ant-research/Pyraformer

https://github.com/MAZiqing/FEDformer

https://github.com/yuqinie98/PatchTST

https://github.com/thuml/iTransformer

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