private mobilenetv1 — function

mobilenetv1(width_mult, config;
            activation = relu,
            inchannels = 3,
            nclasses = 1000,
            fcsize = 1024)

Create a MobileNetv1 model (reference).

Arguments

width_mult: Controls the number of output feature maps in each block (with 1.0 being the default in the paper)
configs: A “list of tuples” configuration for each layer that details:
- dw: Set true to use a depthwise separable convolution or false for regular convolution
- o: The number of output feature maps
- s: The stride of the convolutional kernel
- r: The number of time this configuration block is repeated
activate: The activation function to use throughout the network
inchannels: The number of input feature maps``
fcsize: The intermediate fully-connected size between the convolution and final layers
nclasses: The number of output classes

Tutorials