C1M1_Assignment Notebook Issue

Course Q&A Multi-Agent Systems with CrewAI
,
1

In the graded assignment notebook the agent cells all pass unittests but all of the task cells have failed. This is the response I get from running unittests.test_analyze_code_quality_task(analyze_code_quality). Any help or direction would be appreciated. Thank you.

test_analyze_code_quality_task
test_analyze_code_quality_task1348×530 47.8 KB

2

it’s pointing you have used wrong agent to the task!!! perhaps check while recalling code quality analysis the task has the correct agent recall.

I am not mentor for this course, but if you want me to check your code, you can send me screenshot of the codes you wrote by personal DM

3

Hey @pmj7671! As @Deepti_Prasad mentioned I think you are assigning the wrong agent to the task. I will change the output of the “got” part of the message as it isn’t very intuitive to read. Make sure you are assigning the correct agent to each task and that you didn’t change the role of the agents, otherwise this test will fail.

4

Thanks, Deepi_Prasad. I have tried to set the agent in the task multiple ways but no success. I will continue to focus on this and report back. Thank you!

5

ok let me know if you are still stuck up, you can DM me the codes for review if you want.

Try to look each step before you encountered this error from beginning, different how agents are being recalled, I feel instead of recall the variable argument, you probably have recall the global variable for agent recall in the task step.

Good luck debugging!!!

6

Thanks so much Deepti_Prasad for your assistance and support. It turns out that I was not setting the agent’s name correctly in the agent Role. Everything passed now!

1 Like
7

Thanks you Magdalena for your assistance. It was very helpful. It turns out that I was not setting the agent’s name correctly in the agent Role. Everything passed now! On to the next module.

1 Like
8

Hi. I still have an issue with the grading of my coding assignment notebook for Module 1. The output of the code cell executing the unittest for Exercise 4 shows ‘All tests passed!’ but the Grader shows Exercise 4 having been failed. I had a look at the unittest.py file and cannot see why there should be an issue when grading for Exercise 4 compared to Exercise 2. Both unit tests test code blocks that follow the same pattern (the definitions of the Tech Lead agent in Ex. 4 and Senior Developer agent in Ex. 2 have the same structure without the need to include tools). Thank you for checking and thanks for any feedback!

9

Hey @mjgrav2001! Unittests and autograders aren’t exactly the same, so it could be that you are passing all unittests but fail some of the grader’s, however there are a lot of tests in commen, so you shouldn’t be getting 0 points for an exercise you are passing all unittests for.
Can you please share the output of the grader? You should see a message helping you debug the issue. Also, did you try re-submitting the grader?