def forward()

in projects/home/recap/model/mlp.py [0:0]

• 7 lines of code
• 4 McCabe index (conditional complexity)

  def forward(self, x: torch.Tensor) -> torch.Tensor:
    net = x
    for i, layer in enumerate(self.layers):
      net = layer(net)
      if i == 1:  # Share the first (widest?) set of activations for other applications.
        shared_layer = net
    return {"output": net, "shared_layer": shared_layer}