Neural_machine_translation_with_attention Ex2 Error in test. The lists contain a different number of elements

Course Q&A Deep Learning Specialization Sequence Models
1

Hi all. I cannot understand why this is happening

expected_summary = [[‘InputLayer’, [(None, 30, 37)], 0],
[‘InputLayer’, [(None, 64)], 0],
[‘Bidirectional’, (None, 30, 64), 17920],
[‘RepeatVector’, (None, 30, 64), 0, 30],
[‘Concatenate’, (None, 30, 128), 0],
[‘Dense’, (None, 30, 10), 1290, ‘tanh’],
[‘Dense’, (None, 30, 1), 11, ‘relu’],
[‘Activation’, (None, 30, 1), 0],
[‘Dot’, (None, 1, 64), 0],
[‘InputLayer’, [(None, 64)], 0],
[‘LSTM’,[(None, 64), (None, 64), (None, 64)], 33024,[(None, 1, 64), (None, 64), (None, 64)],‘tanh’],
[‘Dense’, (None, 11), 715, ‘softmax’]]

summary of model [[‘InputLayer’, [(None, 30, 37)], 0], [‘InputLayer’, [(None, 64)], 0], [‘Bidirectional’, (None, 30, 64), 17920], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘Dense’, (None, 30, 10), 1290, ‘tanh’], [‘Dense’, (None, 30, 1), 11, ‘relu’], [‘Activation’, (None, 30, 1), 0], [‘Dot’, (None, 1, 64), 0], [‘InputLayer’, [(None, 64)], 0], [‘LSTM’, [(None, 64), (None, 64), (None, 64)], 33024, [(None, 1, 64), (None, 64), (None, 64)], ‘tanh’], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘RepeatVector’, (None, 30, 64), 0, 30], [‘Concatenate’, (None, 30, 128), 0], [‘Dense’, (None, 11), 715, ‘softmax’]]

AssertionError Traceback (most recent call last)
in
34
35
—> 36 modelf_test(modelf)

in modelf_test(target)
31 assert len(model.outputs) == 10, f"Wrong output shape. Expected 10 != {len(model.outputs)}"
32
—> 33 comparator(summary(model), expected_summary)
34
35

~/work/W3A1/test_utils.py in comparator(learner, instructor)
16 def comparator(learner, instructor):
17 if len(learner) != len(instructor):
—> 18 raise AssertionError(“Error in test. The lists contain a different number of elements”)
19 for index, a in enumerate(instructor):
20 b = learner[index]

AssertionError: Error in test. The lists contain a different number of elements

2

I hope you sorted this out, but if you haven’t.

Check all of your previously defined functions ‘Carefully’ …
and make sure you double check the provided functions, and use if directed.

I had the exact same issue.

3

I have carefully checked all the functions and double checked the provided functions for an hour, yet I still got this exact error except I got 10 extra [‘Dense’, (None, 11), 715, ‘softmax’] at the end.

4

My issue was related to the one_step_attention() function and not using the provided functions directly above it.
But my issue was the same as the original post you made. with Repeat Vector and Concatenate repeated several times.

2 Likes
5

Were you able to sort this out?
I am having the same issue

1 Like
6

Did you sort it out? I’m have the same issue too

7

Hi, yes. I think I was re-writing the dense layer manually instead of using the “output_layer” from before in step 2C

8

Thanks. I found the issue was caused by a mistake I did in the previous exercise’s one_step_attention function, specifically in the ‘energies’ variable. Even though the tests were successful for exercise 1, that mistake followed me to exercise 2.

9

Hi all,

My final dense layer is repeating 10 times. Anyone know why this might be happening please?

10

Welcome to the community !

Please see a picture in that assignment.

image
image1768×1564 238 KB

There are multiple output units that correspond to the number of output characters, i.e, time steps. In our assignment, it is T_y. And it is set to 10 to produce “YYYY-MM-DD”, 10 characters.
Each unit consists of multiple layers like attention, post-attention LSTM, Softmax, …

If you look at modelf(), there is a “loop”, (for t in range(Ty)), to create a set of "one_step_attention, post_activation_LSTM, and Dense layer with Softmax activation. As you see, T_y = 10. So, you will see 10 Dense layers in model.summary()

11

Hi, thanks for reply. Perhaps I should have been clearer, my problem is that they are unusually stacked at the end:

neural_translation
neural_translation1005×408 31.4 KB

12

@anon57530071 I have solved it now. I was using Keras directly but when replaced with your functions (concatenator, etc…) everything worked fine.

13

Great ! Thank you for letting us know.

14

I was getting 10 != 1
Turns out that the “outputs” for the model should equal to “outputs” and not “out”. I thought it would be out, since out is defined as the output layer. But I guess you have to provide ALL the out’s that you appended in the outputs.
Hope this helps someone :smiley: