Table 6 Environment setting analysis.

From: Explainable dual LSTM-autoencoders with exogenous features for anomaly detection and supply chain forecasting

Layer

Requirement

OS / Driver

Linux (Ubuntu 22.04), CUDA 12.1+, cuDNN 8.9+

Language

Python 3.10

DL framework

PyTorch 2.2+ (optional: PyTorch-Lightning 2.3 for training loops)

Numerics data

numpy 1.26, pandas 2.2, scipy 1.11

Feature eng.

PyWavelets 1.5 (wavelets), statsmodels 0.14 (seasonals/ARIMA baselines), prophet 1.1 (baseline)

ML utils

scikit-learn 1.4 (PCA, MI, scalers), ruptures 1.1 (change-points)

XAI

shap 0.45, lime 0.2

Viz

matplotlib 3.8

Back to article page