Update README.md

Signed-off-by: David Rotermund <54365609+davrot@users.noreply.github.com>

pytorch/own_layer/README.md

@@ -21,7 +21,7 @@ reset_parameters(): I copied it from the original code.
 
 forward(): We get an input tensor and need to produce an output tensor. Please remember that dimension 0 contains the batch. Here we just multiply the input with the weights and add the bias to it (if available).
 
-extra_repr(): Plots the information about the layer in a nice fashion. 
+extra_repr(): Prints the information about the layer in a nice fashion. 
 
 
 ```python
@@ -81,4 +81,84 @@ class MyOwnLayer(torch.nn.Module):
         return f"in_features={self.in_features}, out_features={self.out_features}, bias={self.bias is not None}"
 ```
 
+I just add it to the networs as any other layer:
+
+```python
+network = torch.nn.Sequential(
+    torch.nn.Conv2d(
+        in_channels=input_number_of_channel,
+        out_channels=number_of_output_channels_conv1,
+        kernel_size=kernel_size_conv1,
+        stride=stride_conv1,
+        padding=padding_conv1,
+    ),
+    torch.nn.ReLU(),
+    torch.nn.MaxPool2d(
+        kernel_size=kernel_size_pool1, stride=stride_pool1, padding=padding_pool1
+    ),
+    torch.nn.Conv2d(
+        in_channels=number_of_output_channels_conv1,
+        out_channels=number_of_output_channels_conv2,
+        kernel_size=kernel_size_conv2,
+        stride=stride_conv2,
+        padding=padding_conv2,
+    ),
+    torch.nn.ReLU(),
+    torch.nn.MaxPool2d(
+        kernel_size=kernel_size_pool2, stride=stride_pool2, padding=padding_pool2
+    ),
+    torch.nn.Flatten(
+        start_dim=1,
+    ),
+    MyOwnLayer(
+        in_features=number_of_output_channels_flatten1,
+        out_features=number_of_output_channels_full1,
+        bias=True,
+    ),
+    torch.nn.ReLU(),
+    torch.nn.Linear(
+        in_features=number_of_output_channels_full1,
+        out_features=number_of_output_channels_output,
+        bias=True,
+    ),
+    torch.nn.Softmax(dim=1),
+).to(device=device_gpu)
+```
+
+We can print information as usual:
+
+```python
+print(network[-4])
+```
+
+Output: 
+
+```python
+MyOwnLayer(in_features=576, out_features=1024, bias=True)
+```
+
+We can check the stored parameters:
+
+```python
+for parameter in network[-4].parameters():
+    print(type(parameter), parameter.shape)
+
+print()
+
+for name, parameter in network[-4].named_parameters():
+    print(name, type(parameter), parameter.shape)
+```
+
+Output:
+
+```python
+<class 'torch.nn.parameter.Parameter'> torch.Size([1024, 576])
+<class 'torch.nn.parameter.Parameter'> torch.Size([1024])
+
+weight <class 'torch.nn.parameter.Parameter'> torch.Size([1024, 576])
+bias <class 'torch.nn.parameter.Parameter'> torch.Size([1024])
+```
+
+And train the network as usual:
+
 ![Figure_1.png](Figure_1.png)