Table 2 Experimental hyperparameter configurations for model training.

From: A deep learning based framework for music-synchronized dance choreography with pose quantization and motion prediction for activity recognition

Parameters

Value/Settings

\(\:\alpha\:\)

[0.1 to 0.0001]

No of Neurons

1024

Optimizer

Differential Evolution

Training: Testing

80: 20

Drop period

6

Kernal Size

3 × 3

Patch size

16 × 16

Minimum batch size

16

Momentum

0.85

Number of Epochs

60

Option

Default

Back to article page