def forward()

in qlearn/commun/norm_flows.py [0:0]

• 14 lines of code
• 5 McCabe index (conditional complexity)

    def forward(self, z, kl=True):
        if kl:
            if z.dim() == 1:
                logdets = 0
            else:
                logdets = Variable(torch.zeros_like(z[:, 0]))
            for flow in self.flow_list:
                z, logdet = flow(z, kl=True)
                logdets += logdet
            return z, logdets
        else:
            for flow in self.flow_list:
                z = flow(z, kl=False)
            return z