Is there a specific way to implement the graded functions?

@Mubsi
Is there a fixed way to implement the graded functions?
output

For eg: In the Week-2 Assignments, I have not used the df.isnull() function as a part of my code. I used NumPy and other methods to complete it. Even though the output is correct, I scored a 5/6.

I observed something similar for Week-1 as well! There also, I implemented the standard deviation in one of the functions using NumPy, but the grader showed incorrect output!

It would be great if you could shed some light on this!

Thanks in advance
Abhik Biswas

Hi @Abhik_Biswas,

Yes, the AI4M are not that robust, and sometimes they require specific solutions. Can you share your lab ID with me so I could figure out where you didn’t get the score ?

Thanks,
Mubsi

@Mubsi here is my Lab ID:

trulkhhp

Thanks
Abhik

Hi @Abhik_Biswas,

Yes, I believe for Ex 1, the expected functions by the autograder are that of Pandas, but you used NumPy. In order to get full credit, I’d recommend to use the Pandas function.

Best,
Mubsi

1 Like

Noted @Mubsi

Just a small suggestion: Please let us know the functions that are to be used, in the description cells above. That way, we can directly write the correct function.

There are numerous ways to write solutions to a problem, and I was really disheartened :disappointed: when I lost marks even after it showed “All tests passed!”

Thanks
Abhik

Hi @Abhik_Biswas,

In the “Hints” for Ex 1, it was clearly stated which function to use.

Best,
Mubsi

Hi @Mubsi ,
I recently finished the ML Specialization and the hints there had solutions to the problem itself. That’s what I thought for this course also. Will most certainly keep this in mind in the future.

Thanks
Abhik