A3: music_inference_model()

Course Q&A Deep Learning Specialization Sequence Models
1

Step 2.D:

    # Select the next value according to "out",
    # Set "x" to be the one-hot representation of the selected value
    # See instructions above.

Isn’t the input of the current time step the output of the previous step? If so, x is set to out? Then when calling tf.one_hot(), indices(first argument) must be x and the second argmax of out? I get an error:
TypeError: Value passed to parameter ‘indices’ has DataType float32 not in list of allowed values: uint8, int32, int64

Any clues are appreciated.

2

Your first argument is correct but the second argument is a depth of your one-hot vector and it is a scaler.

Hint for a second argument (from notebook):

  • Convert the index into its n_values-one-hot encoding using tf.one_hot.
3

depth of one-hot vector? is it n_values?

4

Yes.

5

Thanks. Then I get this error when I run music_inference_model()

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

6

Your code might be wrong. Please revisit the instructions and compare your code with it. If still facing any issue, please share full error.

7

Here is the full error:

AssertionError Traceback (most recent call last)
in
3 # UNIT TEST
4 inference_summary = summary(inference_model)
----> 5 comparator(inference_summary, music_inference_model_out)

~/work/W1A3/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

8

I think I figured it out. thanks for the help tho.