Parametric Coding Assignments

Pang, Bei

Permalink

https://hdl.handle.net/2142/97836 Copy

Description

Title

Parametric Coding Assignments

Author(s)

Pang, Bei

Contributor(s)

Lumetta, Steven S.

Issue Date

2016-12

Keyword(s)

• coding assignment
• automatic feedback
• concolic testing

Abstract

This thesis introduces a set of parametric coding assignments for ECE 120, an entry-level course of the Department of Electrical and Computer Engineering. These coding assignments are called “parametric” because the function implemented in these assignments varies according to different input values. Another significant feature of the parametric coding assignments is that the assignments of different students are different from each other in terms of parameters and implementation requirements. But the difficulties and skills required to complete an assignment for all students are identical. These coding assignments use an automated feedback framework to grade students’ codes with KLEE, a symbolic testing tool, and to provide feedback (grading results and error messages) on students’ code submitted via Subversion, version control software used to distribute and collect coding assignments. The automated feedback framework was designed and implemented by Jianxiong Gao, a Ph.D. student of Prof. Steve S. Lumetta, and this framework was successfully applied to ECE 220 in previous semesters. In this semester, Fall 2016, around one-quarter of ECE 120 students signed up voluntarily to try these assignments. However, because these coding assignments are optional exercises for the students, we received only a small number of submissions. Discussion of the limited but enlightening submissions is in Chapter 4.

Type of Resource

text

Language

en

Permalink

http://hdl.handle.net/2142/97836

Owning Collections