gradescope / autograder_samples

Examples of autograders for running on Gradescope
https://gradescope-autograders.readthedocs.io/en/latest/
82 stars 164 forks source link
autograder autograding computer-science course-management gradescope

Documentation Status

Overview

Gradescope provides a language-agnostic platform for running your autograders on our infrastructure. By running in Docker containers, we give you full flexibility in setting up whatever language, compilers, libraries, or other dependencies you need for your programming assignments. You provide us with a setup script and an autograder script, along with whatever supporting code you need, and we manage accepting student submissions, running your autograder at scale, and distributing the results back to students and to you.

!!! note "Updates" Our autograder platform is under active development! Check out the Updates page to see what we've changed recently.

How it works

As an instructor, you create a new assignment on Gradescope, and upload your autograder zip file following our specifications. Your code produces output in the format we request. Students submit to Gradescope and have their work evaluated on demand. They can submit as many times as they want. At the end of the process, you can download their code and their results.

Follow our instructions to get started with autograding.

Examples

If you want to jump right in, we have built the following examples for autograders in different languages. You can see how they are built, and use them as a starting point for your own autograders. If you have any further questions, feel free to contact us at help@gradescope.com

Getting Help

You can also ask questions on the GitHub discussions forum for this project, especially if there is no private information in your question.

If you need any help getting set up with the autograder platform, please contact us at help@gradescope.com. Please mention Autograder in the subject line, it'll help us categorize the type of support request.

Issues regarding setup or installation of packages are often not Gradescope specific. In general, information about installing packages on our base operating system (currently Ubuntu 22.04 by default) will be relevant to Gradescope. If you get errors during your autograder setup phase it may help to search for those errors on Google or Stack Overflow.

Given the multitude of software packages that are in use on the Gradescope autograder platform, the Gradescope team is not intimately familiar with each package's setup and configuration. When trying to install or use packages, it is a good idea to check the package's installation instructions or source code repository for any tips on installing on Ubuntu 22.04, or instructions on configuration and usage. It is also helpful to contact the authors of such packages when possible because they will be more familiar with their own code.

Pricing

The code autograder platform is available with an Institutional license or Institutional Trial. If you have any questions regarding pricing, please contact us at sales@gradescope.com.