hi ,i have a problem with my music_inference_model() function especially the task 2.E,the task is mentionning using RepeatVector(1) but when i tried it,the tester is not working saying Outputs = 50
Single output shape = (None, 90)
len(pred) = 50
pred[0].shape = (1, 90)
Test failed at index 7
Expected value
[‘Lambda’, (None, 1, 90), 0]
does not match the input value:
[‘RepeatVector’, (None, 1, 90), 0, 1],it’s mentionning that it requires lambda not RepeatVecto(), i also tried to use lambda but it still mentionning the RepeatVector problem,can you help me please?
i used RepeatVector but didn’t work,recarding the other steps the tests passed successfully so logically there are no problems
Can i send code?
Also the test output mentionning that the expected value is lambda :Outputs = 50
Single output shape = (None, 90)
len(pred) = 50
pred[0].shape = (1, 90)
Test failed at index 7
Expected value
[‘Lambda’, (None, 1, 90), 0]
does not match the input value:
[‘RepeatVector’, (None, 1, 90), 0, 1] so i’m confuses
Sorry, but I’m confused: the error message seems to say that Lambda is the expected value. I checked my output and there is no Lambda and I pass the comparator test. So where does that error message come from? Is Nizar using an old or different version of the notebook?
Or maybe there’s a bug in how the test is written that generates that message?
def comparator(learner, instructor):
if len(learner) != len(instructor):
raise AssertionError("Error in test. The lists contain a different number of elements")
for index, a in enumerate(instructor):
b = learner[index]
if tuple(a) != tuple(b):
print(colored(f"Test failed at index {index}", attrs=['bold']),
"\n Expected value \n\n", colored(f"{b}", "green"),
"\n\n does not match the input value: \n\n",
colored(f"{a}", "red"))
raise AssertionError("Error in test")
print(colored("All tests passed!", "green"))
It reverses the role of “expected” and “actual”. I will file a bug.
Yes, Prof Ng is a genuinely great teacher. There used to be a specialization called Machine Learning Engineering for Production (MLOps), but I think it is being revised and is no longer available. I have not taken it in the past, but I believe that it is not taught by Prof Ng.
There is also a specialization about Data Engineering, but I am also not familiar with that one. Please have a look at the list of courses and see if that sounds relevant to you.
But as Tom suggests, there are lots of possible interpretations of your question. If I’m missing your point, please let us know in more detail what kind of material you are seeking.
I just searched on Coursera and found the DLAI MLOps course, so I guess I was wrong and it still is available.
As I mentioned before, I have not taken it, so cannot offer any more information than the above.
When I searched for “machine learning engineering for operations” on Coursera, I found offerings on the subject from a number of other providers, but I also have no further information about any of them. So you would need to check their curricula and look for reviews and the like.
An abbreviated version of the MLOPs course is still available. But it’s a lot shorter than it once was. The material had become dated and was no longer supportable.
…