Skip to content

QiSima/TimeSeriesForecasting-RD

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Reinforced Decoder: Towards Training Recurrent Neural Networks for Time Series Forecasting

Pytorch implementation of the Reinforced Decoder. Here, a novel training and decoding approach that introduces auxiliary models and leverages reinforcement learning to guide the decoder processing, is proposed for S2S models in time series forecasting.

Getting Started

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

  • Python 3.8.8
  • numpy == 1.20.1
  • pandas == 1.2.4
  • scikit_learn == 0.24.1
  • torch == 2.2.1

Dataset

All the datasets needed for Reinforced Decoder can be obtained from the following public links, which cover both simulated and real-world time series and have different time granularities and are either univariate or multivariate.

Examples

You can easily implement Reinforced Decoder by running the provided python file. For instance, to training the LSTM-based S2S model by Reinforced Decoder, execute the following command:

python run_model.py

Acknowledgement

This library is constructed based on the following repos:

The S2S model with Informer and ESLSTM structures is based on the following repos:

Contact

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages