Add-LSTM

[Erennn7]

PR: LSTM Time Series Forecasting in R

This PR introduces a fully documented implementation of a Long Short-Term Memory (LSTM) neural network in R for time series prediction and forecasting.
The implementation leverages the keras and tensorflow packages to build, train, and evaluate LSTM models capable of learning temporal dependencies in sequential data.

Overview

The provided LSTM workflow covers:

• Data Preprocessing:

  • Normalization of time series data to [0,1] range
  • Sequence creation for supervised learning (input sequences and target values)

• Model Architecture:

  • Single-layer LSTM network with configurable units and dropout
  • Dense output layer for regression tasks
  • Compilation with Adam optimizer and Mean Squared Error loss

• Training and Evaluation:

  • Training with configurable epochs and batch size
  • Validation split and monitoring of training history
  • Evaluation on test data with metrics: MSE, RMSE, MAE, and R-squared

• Prediction Capabilities:

  • Single-step and multi-step ahead predictions
  • Denormalization of predictions for interpretation
  • Visualization of actual vs. predicted series using ggplot2

• Best Practices:

  • Sequence length selection based on temporal dependencies
  • Dropout regularization to prevent overfitting
  • Early stopping, model checkpointing, and hyperparameter tuning

This LSTM implementation is suitable for:

• Synthetic or real-world time series (e.g., sine waves, stock prices, sensor data)
• Forecasting applications requiring learning of sequential patterns
• Multi-step ahead prediction and analysis of temporal trends

Complexity

• Training Complexity: O(n_samples × seq_length × lstm_units) per epoch
• Inference Complexity: O(n_samples × seq_length × lstm_units)

The approach demonstrates how LSTMs can model sequential dependencies more effectively than traditional regression or moving-average methods.

[Copilot]

Pull Request Overview

This PR introduces a comprehensive implementation of Long Short-Term Memory (LSTM) neural networks for time series forecasting in R, providing a complete workflow from data preprocessing to prediction visualization.

• Implements LSTM model architecture with configurable parameters for time series prediction
• Provides comprehensive data preprocessing including normalization and sequence generation
• Includes evaluation metrics, visualization capabilities, and multi-step ahead prediction functionality

[Copilot]

The reshape2 package is used without being explicitly loaded or checked for availability. Consider using tidyr::pivot_longer() as a more modern alternative, or add a requireNamespace check for reshape2.

[Copilot]

Random sampling for train/test split breaks the temporal order of time series data. For time series, use sequential split to maintain temporal relationships: train_indices <- 1:train_size and test_indices <- (train_size + 1):dim(X)[1].

Suggested change

	indices <- sample(1:dim(X)[1])
	train_indices <- indices[1:train_size]
	test_indices <- indices[(train_size + 1):length(indices)]
	train_indices <- 1:train_size
	test_indices <- (train_size + 1):dim(X)[1]

[Copilot]

The 'size' aesthetic is deprecated in ggplot2. Use 'linewidth' instead for line geometries.

Suggested change

	geom_line(size = 1) +
	geom_line(linewidth = 1) +