Research Article
An LSTM-Autoencoder Architecture for Anomaly Detection Applied on Compressors Audio Data
Table 13
Performance results on train dataset (created by authors).
| | Model architecture | Loss function | Activation
Function | Optimizer | Accuracy | Precision | Recall | F1 |
| | GRU | Mean square error | SoftMax | Adam | 89% | 88% | 55% | 67% | | GRU-1 | Sparse categorical cross entropy | Sigmoid | Adam | 99% | 100% | 99% | 99% | | GRU-2 | Sparse categorical cross entropy | SoftMax | Adam | 99% | 99% | 99% | 99% | | Stacked GRU | Sparse categorical cross entropy | SoftMax | Adam | 99% | 99% | 99% | 99% | | LSTM | Sparse categorical cross entropy | Sigmoid | Adam | 99% | 100% | 97% | 98% | | LSTM-1 | Sparse categorical cross entropy | SoftMax | Adam | 99% | 100% | 99% | 99% | | Stacked LSTM | Sparse categorical cross entropy | SoftMax | Adam | 99% | 100% | 99% | 99% | | LSTM- Autoencoder | Sparse categorical cross entropy | SoftMax | Adam | 100% | 100% | 100% | 100% |
|
|
