Research Article

Spatial-Temporal Feature with Dual-Attention Mechanism for Encrypted Malicious Traffic Detection

Table 7

Parameters of the TLARNN at flow layer.

LayerOutput shapeParam

Input_5(None, 30, 300)0
Time_distributed_1(None, 30, 256)377600
Bidirectional_1(None, 30, 1024)2365440
Attention_mechanism_3(None, 1024)525312
Dense_1(None, 4)4100

Total params: 3,272,452. Trainable params: 3,272,452. Nontrainable params: 0.