Skip to main content

Autograders in Python Basics with Tracy

Learn more about Tracy Turtle autograder limitations and alternatives for using autograders in related CodeHS courses

Billy Gotsch avatar
Written by Billy Gotsch
Updated yesterday

Tracy programs are difficult to grade with an autograder accurately. This is because CodeHS autograders can only check the code a student wrote and are not able to verify the output of a studentโ€™s program in the canvas. For instance, if we ask a student to draw a circle in each corner of the canvas, the autograder can verify if the student includes a for loop and a circle() command, however, it cannot evaluate if the student code results in circles which are positioned as expected.

Historically, Tracy autograder challenges have caused a fair amount of frustration among teachers and students. As students learn more commands there are multiple correct ways that code could be written to achieve the goal. The latest version of our Python Tracy course removes autograders because it is impossible to check all of the ways that code could be written that will lead to the correct solution. Students have greater flexibility in creating a solution.

Because creating autograders to account for every variation is not possible, the course includes the Lesson called Testing Tracy Programs, which teaches students how to test their code and evaluate their results using the rubrics provided with each exercise.

Banner: This is a free feature

If your Sections require autograded assignments, we recommend considering the following options:

  • The Introduction to Programming with Turtle Graphics Module in Introduction to Computer Science in Python 3 is an older version of the Tracy and Python console content. This Module includes autograders. However, the autograder limitations discussed above apply to these assignments. Student code may sometimes pass the autograder when it is not correct or student code may not pass the autograder even though it produces the correct result.

  • The โ€‹Mix and Matchโ€‹ course has 6 out of 12 modules using autograders. The modules that don't include autograders are: Exploring Computing, Exploring the Internet, Exploring Data and Spreadsheets, Exploring Web Design 2, and Exploring Digital Citizenship.

  • The Video Game Design in JavaScriptโ€‹ course includes autograders for all coding assignments, except for the final project which is open-ended. This course uses JavaScript rather than Python, so the change in language may need to be evaluated to fit your class needs.

Check out this article for more information on adding existing CodeHS content to your current Sections: Adding Courses, Modules, and Assignments to Your Section ๐Ÿ“–

Banner: This requires a Pro license

For teachers with Pro licenses, these activities can be graded using the AI Grading.


Still have questions? Contact our team at hello@codehs.com to learn more!

Did this answer your question?