Implementing a sub class in parent class in pytorch

i want to implement a pytroch nn.class with a subclass in it for a neural network but i am getting
this error

raise NotImplementedError
NotImplementedError

pleae some one can figure out

This (ConvBNR) is the sub class which i want to implemet in the main class(DPclass) of neural netwrok

class ConvBNR(nn.Module)
      def __init__(self,in_channels, out_channels):
             #def __init__():
            super(ConvBNR,self).__init__()
            self.net = nn.Sequential(
            nn.Conv2d(32,32,3, 1, 1),
            nn.BatchNorm2d(32),
            nn.ReLU()
        )

       def call(self, x, training=None):
        x = self.net(x, training=training)

        return x



 This is my main class of NN

class (DPclass) (nn.Module):
    def __init__(self):
        #super(DPclass)()
    super().__init__()
         self.start = nn.Sequential(
            nn.Conv2d(in_channels=3, out_channels=32, kernel_size=3, stride=1, padding=1),
            nn.BatchNorm2d(32),
            nn.ELU()
            #nn.Dropout(0.25)
        )
       
        #self.start2 = ConvBNRelu(32,64,3,1,1)
        self.start2 = ConvBNRelu(32,64)
        some other layers......... 

   def forward(self, x):
         A = self.start(x)#here1
        print("A.shape1",A.shape)
        A = self.start2(x)#here1
       .........

sorry i cant properly ident the code here but i hope you will figure out

The ConvBNR modules implements the call method while you should provide a forward method.

oh right thank you sir