Anyone else getting randomized seeds that will push over the number of responses correct that is even allowed? Whenever I run it myself, I will get everything the way it should be. Even when I run it through the grader, if the seed does a response under 5 then it will work fine but mine will go up to 20 at times. Any idea how to fix this?
Oh it is actually storing the number of correct responses in the prior tests and adding them onto the next tests. Whenever I run it myself, it just resets but I guess the autograder is storing it for some reason?
Figured it out. Just had to move a variable initialization around
Thanks for posting @phoa - helped me - never thought about needing to adjust code to account for GitHub autograder testing numerous times