Pooling

class dragon.vm.caffe.layers.Pooling(layer_param)[source]

Apply the n-dimension pooling.

The spatial output dimension is computed as:

\[\text{Dim}_{out} = (\text{Dim}_{in} + 2 * pad - \text{K}_{size}) / stride + 1 \]

Examples:

layer {
    type: "Pooling"
    bottom: "conv2"
    top: "pool2"
    pooling_param {
        kernel_size: 3
        stride: 2
        pool: AVG
    }
}