Encoder error - Week 4

Course Q&A Deep Learning Specialization Sequence Models
1

All,

I’m running into a problem with my Encoder and am stuck. Any tips?

ERROR:

positions 5
d 4
angle_rads [[0. 0. 0. 0. ]
[1. 1. 0.01 0.01]
[2. 2. 0.02 0.02]
[3. 3. 0.03 0.03]
[4. 4. 0.04 0.04]]
angle rad shape (5, 4)

TypeError Traceback (most recent call last)
in
1 # UNIT TEST
----> 2 Encoder_test(Encoder)

~/work/W4A1/public_tests.py in Encoder_test(target)
114 x = np.array([[2, 1, 3], [1, 2, 0]])
115
→ 116 encoderq_output = encoderq(x, True, None)
117
118 assert tf.is_tensor(encoderq_output), “Wrong type. Output must be a tensor”

/opt/conda/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer.py in call(self, *args, **kwargs)
1010 with autocast_variable.enable_auto_cast_variables(
1011 self._compute_dtype_object):
→ 1012 outputs = call_fn(inputs, *args, **kwargs)
1013
1014 if self._activity_regularizer:

in call(self, x, training, mask)
56 # Pass the output through the stack of encoding layers
57 for i in range(self.num_layers):
—> 58 x = self.enc_layers(x)
59 # END CODE HERE
60

TypeError: ‘ListWrapper’ object is not callable

2

When your Encoder calls “x = self.enc_layers(…)”, you need to index the layer with [i], and you need to pass all of the parameters x, training, and mask. Not just ‘x’.

6 Likes
3

Got it. Thank Tom!

JB

4

For any other “less than expert” Python coders out there, remember that depending on context, one of the following may be appropriate while the other is not:

thing = method[something](something else)
thing = method(something)[something else]

This observation got me unstuck after (admittedly) many attempts.

1 Like
5

Thanks you for this very useful tip !

1 Like