Week1: Conv Model application- ZeroPadding2D error

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

Hi Everyone,
I am getting this error while compiling the code for Week 1 Programming Assignment 2 -

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-133-f33284fd82fe> in <module>
----> 1 happy_model = happyModel()
      2 # Print a summary for each layer
      3 for layer in summary(happy_model):
      4     print(layer)
      5 

<ipython-input-132-93e688c7667d> in happyModel()
     32 
     33             # YOUR CODE STARTS HERE
---> 34             tfl.add([tfl.ZeroPadding2D(padding=3)(input_shape=tf.keras.Input(shape= (64,64,3))),
     35                        tfl.Conv2D(filters=32, Kernel_size=(7,7), strides=1),
     36                        tfl.BatchNormalization(axis=3),

/opt/conda/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer.py in __call__(self, *args, **kwargs)
    914     #   not to any other argument.
    915     # - setting the SavedModel saving spec.
--> 916     inputs, args, kwargs = self._split_out_first_arg(args, kwargs)
    917     input_list = nest.flatten(inputs)
    918 

/opt/conda/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer.py in _split_out_first_arg(self, args, kwargs)
   2978     else:
   2979       raise ValueError(
-> 2980           'The first argument to `Layer.call` must always be passed.')
   2981     return inputs, args, kwargs
   2982 

ValueError: The first argument to `Layer.call` must always be passed.

Could you please guide me where i am going wrong with this?

Thank you.

2

When you’re using the .add() method, you don’t need the square brackets.

3

I tried to remove that, but still gives me the same error. Not sure how to add layers in the sequential model API within square brackets. :frowning_face:

4

Do i need to use the add() function or is this something that will be automatically taken care of in the sequential model function?

5

I removed the .add() function and now i get this error -

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-224-f33284fd82fe> in <module>
----> 1 happy_model = happyModel()
      2 # Print a summary for each layer
      3 for layer in summary(happy_model):
      4     print(layer)
      5 

<ipython-input-223-c66c2a94f9a1> in happyModel()
     41                        tfl.MaxPool2D(pool_size=(2, 2)),
     42                        tfl.Flatten(),
---> 43                        tfl.Dense(1, activation='sigmoid')
     44             # YOUR CODE ENDS HERE
     45         ])

/opt/conda/lib/python3.7/site-packages/tensorflow/python/training/tracking/base.py in _method_wrapper(self, *args, **kwargs)
    455     self._self_setattr_tracking = False  # pylint: disable=protected-access
    456     try:
--> 457       result = method(self, *args, **kwargs)
    458     finally:
    459       self._self_setattr_tracking = previous_value  # pylint: disable=protected-access

/opt/conda/lib/python3.7/site-packages/tensorflow/python/keras/engine/sequential.py in __init__(self, layers, name)
    140         layers = [layers]
    141       for layer in layers:
--> 142         self.add(layer)
    143 
    144   @property

/opt/conda/lib/python3.7/site-packages/tensorflow/python/training/tracking/base.py in _method_wrapper(self, *args, **kwargs)
    455     self._self_setattr_tracking = False  # pylint: disable=protected-access
    456     try:
--> 457       result = method(self, *args, **kwargs)
    458     finally:
    459       self._self_setattr_tracking = previous_value  # pylint: disable=protected-access

/opt/conda/lib/python3.7/site-packages/tensorflow/python/keras/engine/sequential.py in add(self, layer)
    180       raise TypeError('The added layer must be '
    181                       'an instance of class Layer. '
--> 182                       'Found: ' + str(layer))
    183 
    184     tf_utils.assert_no_legacy_layers([layer])

**TypeError: The added layer must be an instance of class Layer. Found: tf.Tensor(**
[[[[0 0 0]
   [0 0 0]
   [0 0 0]
   ...
   [0 0 0]
   [0 0 0]
   [0 0 0]]

  [[0 0 0]
   [0 0 0]
   [0 0 0]
   ...
   [0 0 0]
   [0 0 0]
   [0 0 0]]]], shape=(600, 70, 70, 3), dtype=int64)
6

There are two ways to implement the Sequential model.

  • One is by creating a Sequential layer, and passing it a list (within square brackets) of the layers you want to use, separated by commas.

  • The other is by creating a Sequential layer, and then using the .add method to add the other layers. This method does not use a list, so you don’t need square brackets or commas.

Given the error you are seeing, I suspect there’s a problem (a syntax error) in how you’re defining the list of layers.

7

I am doing it by the 1st method you suggested as in the assignment, square brackets are already placed in within the sequential model function → 

model = tf.keras.Sequential([ .. ])

I am passing in the list of layers comma separated.

8

would it be okay if I can message you my code for evaluation? Not sure, where the code is wrong

1 Like
10

Thank you. (mentor edited the reply)

11

For someone who might be facing the same problem as me :
Tip - Pass InputLayer as your first parameter to the sequential keras model and everything else will fall into place. :slight_smile:

1 Like
12

UnboundLocalError Traceback (most recent call last)
in
----> 1 happy_model = happyModel()
2 # Print a summary for each layer
3 for layer in summary(happy_model):
4 print(layer)
5

in happyModel()
21
22 ## ZeroPadding2D with padding 3, input shape of 64 x 64 x 3
—> 23 model.add(tfl.ZeroPadding2D(padding=(3,3))),
24
25 ## Conv2D with 32 7x7 filters and stride of 1

UnboundLocalError: local variable ‘model’ referenced before assignment

13

Hi, I believe you are using sequential keras model for assignment -

In that case, .add() function is not necessary. You can directly pass the parameters comma separated.
Hope this helps. :slight_smile: