Grading Error in NLP course 1 Week 4 assignment

Course Q&A Natural Language Processing NLP with Classification and Vector Spaces
1

Creates Grading error in course 1 week 4 assignment

Grader Error: Grader feedback not found

Kindly help me to resolve the issue

2

Most likely that means one of two things:

  1. A problem on the grading servers. Please try submitting again and see if the problem persists. If so then the next theory is:

  2. The metadata or other structural information in your notebook has been corrupted. To remedy this, you need to get a clean copy of the notebook and then carefully “copy/paste” over just your solution code from your damaged copy. There is a topic on the DLS FAQ Thread about how to get a clean notebook and those instructions work for NLP as well. Let me know if you have trouble following that link.

3

I didn’t experience it on this assignment, but I did run into it on one of the other NLP exercises. Turns out due to cut and paste a global context variable had ‘leaked’ into one of my graded functions. A little fragment of provided code had something like this:

temp_weights = test_computing_weights()

and later in the notebook I reused that variable, temp_weights where I should have used the real_weights variable passed into my function as a parameter. The grader had all kinds of fits and it took me most of the day yesterday to track it down. As soon as I made the change, my grade went from 0% to 100%. Hope this helps.

4

Thanks Mr.Paulin.I have got clean copy of the notebook after following the steps given by you and got my assignments graded and submitted .Thanks Paulin

1 Like
5

Screenshot (453)
Screenshot (453)1920×1080 240 KB

Screenshot (454)
Screenshot (454)1920×1080 237 KB

Screenshot (455)
Screenshot (455)1920×1080 235 KB

Screenshot (457)
Screenshot (457)1920×1080 250 KB

These issues in the screenshot are preventing me from grading my scores, i need help

6

Sorry, I have not done that assignment, so I can’t help with the specific details. But you are getting very similar messages in two different test cases. I suggest that you look at the test functions to see what they are doing and perhaps that will give some clues about the nature of your problem. Just on general principles, it sounds like you have unintentionally modified the definitions of some of the functions under test. The function definitions are typically given as part of the template code and don’t need to be modified. You might want to get a clean copy of the notebook and compare to see if you can find the pattern for what you modified. There is a link earlier on this thread about how to get a clean copy.

1 Like
7

Without seeing the OP’s code, I think that is a really good intuition. Here is some additional context based on my experience with this exercise:

  • Each of the functions throwing errors for the OP has many parameters in a comma delimited list in their function definition in the notebook.

  • The names and values of the parameters to pass during unit testing are defined in a compound data structure named input in w4_unittest.py

  • The parameters causing the runtime errors are all legitimate parameter names in the notebook’s original function definitions. That is, approximate_knn() expects a parameter named hash_tables, align_embeddings() expects a parameter named verbose etc.

So it seems something has happened to those function definition lines of code in the notebook.

8

please i need a clean copy of this notebook, kindly share with me. I am still stuck here. Thank you

9

The link was given earlier on this thread. You can do it for yourself. Even though that thread is from DLS, the method is generic and will work here in NLP as well. Please see the very first topic on that FAQ thread.

10

Thank you sir, it worked . :smiley: .