Package 'WaveletLSTM'

Title: Wavelet Based LSTM Model
Description: A wavelet-based LSTM model is a type of neural network architecture that uses wavelet technique to pre-process the input data before passing it through a Long Short-Term Memory (LSTM) network. The wavelet-based LSTM model is a powerful approach that combines the benefits of wavelet analysis and LSTM networks to improve the accuracy of predictions in various applications. This package has been developed using the algorithm of Anjoy and Paul (2017) and Paul and Garai (2021) <DOI:10.1007/s00521-017-3289-9> <doi:10.1007/s00500-021-06087-4>.
Authors: Dr. Ranjit Kumar Paul [aut], Dr. Md Yeasin [aut, cre]
Maintainer: Dr. Md Yeasin <[email protected]>
License: GPL-3
Version: 0.1.0
Built: 2025-02-26 04:02:13 UTC
Source: https://github.com/cran/WaveletLSTM

Help Index

Wavelet Based LSTM Model

Description

Wavelet Based LSTM Model

Usage

WaveletLSTM(
  ts,
  MLag = 12,
  split_ratio = 0.8,
  wlevels = 3,
  epochs = 25,
  LSTM_unit = 20
)

Arguments

ts

Time Series Data
MLag

Maximum Lags
split_ratio

Training and Testing Split
wlevels

Wavelet Levels
epochs

Number of epochs
LSTM_unit

Number of unit in LSTM layer

Value

  • Train_actual: Actual train series

  • Test_actual: Actual test series

  • Train_fitted: Fitted train series

  • Test_predicted: Predicted test series

References

Paul, R.K. and Garai, S. (2021). Performance comparison of wavelets-based machine learning technique for forecasting agricultural commodity prices, Soft Computing, 25(20), 12857-12873

Examples

y<-rnorm(100,mean=100,sd=50)
WTSLSTM<-WaveletLSTM(ts=y)