| Parameter | Scope |
| Population | 20 | Dimension | 19 | Iterations | 6 | Optimiser | “Adam”, “SGD”, “Adamx”, “Adadelta”, “Adagrad”, “AdamW”, “ASGD”, “RMSprop” | Learning rate | [1e − 5, 1e − 1] | Batch size | [16, 64] | Activation function | “Relu”, “Sigmoid”, “Relu6”, “Tanh”, “LRelu”, “Softsign” | Channel number | [3, 15], [16, 30], [31, 45], [46, 60] | Kernel size | [2, 3, 3, 5, 6, 6, 7, 11] | Pooling form | “Maxpool”, “Avgpool” | Hidden unit in LSTM | [20, 80] |
|
|