torch.nn.Sequential

class Add(nn.Module): def __init__(self, value): super().__init__() self.value = value def forward(self, x): return x + self.value calculator = nn.Sequential( ...