Submitty logo

If you have made changes to the configuration for an assignment and want to reprocess your previous practice submissions, or an error in the assignment configuration was detected after students submitted homeworks, you can queue these previous submissions for regrading by using the regrade.py script.

  1. This script takes as arguments, one or more paths pointing to the portion of the course directory structure to be regraded.

    • For example, to regrade student smithj’s 5th submission for hw02, run this command:

      /usr/local/submitty/bin/regrade.py /var/local/submitty/courses/f16/csci1200/submissions/hw02/smithj/5
      
    • Instead to regrade every version of student smithj’s hw02, run this command:

      /usr/local/submitty/bin/regrade.py /var/local/submitty/courses/f16/csci1200/submissions/hw02/smithj 
      
    • And similarly, to regrade all students submissions of hw02, run this command:

      /usr/local/submitty/bin/regrade.py /var/local/submitty/courses/f16/csci1200/submissions/hw02 
      
    • You can use wildcards in your pattern as well. To regrade all of student smithj’s submissions for all assignments in the course, run:

      /usr/local/submitty/bin/regrade.py /var/local/submitty/courses/f16/csci1200/submissions/\*/smithj 
      

      The command also works without the escaped wildcard – the wildcard expansion happens on the command line instead of within the python script. Note that the command line expansion may fail if the wildcard expansion exceeds the maximum argument list length.

      /usr/local/submitty/bin/regrade.py /var/local/submitty/courses/f16/csci1200/submissions/*/smithj 
      
    • You may also use a relative path. If you are currently in the /var/local/submitty/courses/f16/csci1200/ directory, you can run:

      /usr/local/submitty/bin/regrade.py submissions/hw02/smithj/5
      
  2. If the number of submissions matching the specified pattern is greater than 50, the script will require interactive confirmation [y/n] before queueing those homework submissions for reprocessing.

  3. You may also want to use the grading_done.py script to keep track of the progress in reprocessing a large set of submissions.

    /usr/local/submitty/bin/grading_done.py 
    

    This command will display the number of grading processes, the number of items waiting in the interactive queue (items recently submitted by students) and in the batch queue (items sent for reprocessing by any instructor).

    Note that currently typical instructor users cannot access the details of the interactive queue, thus only the status of the batch queue is presented.

    It is often helpful to combine these two commands, e.g.:

    /usr/local/submitty/bin/regrade.py /var/local/submitty/courses/f16/csci1200/submissions/hw02 && /usr/local/submitty/bin/grading_done.py