|
Course
Highlights
This one-day course package is especially designed for intermediate users to provide hands-on experience using the features in the Matlab language to write efficient, robust, and well organized code. These concepts form the foundation for writing full applications, developing algorithms, and extending product capabilities. Details of performance optimization are covered throughout the course, as well as tools for writing, debugging, and profiling code.
Course
Objectives
The aim of the course is to provide structured data and code for Matlab programmers to achieve performance optimization of the written code with respect to storage requirement and execution time.
Who Must Attend
This hands-on course is designed for intermediate users. It is especially useful for those who want to gain deeper understanding on Matlab programming techniques. Engineers, researchers, scientists, and managers working with systems level design will be shown techniques on code performance improvement.
Course Benefits
Upon the completion of the course, the participants will gain skills and knowledge on how to create applications which are able to handle unexpected input and error conditions as well as exploring different function types and the tradeoffs associated with these choices.
Prerequisites
Attended "Comprehensive MATLAB" course or equivalent experience using Matlab and familiarity with basic programming concepts is required
Course
Outline
Programming for Correctness
Objective: Creating robust applications that withstand unexpected input and produce meaningful errors is the goal of many application developers. Through the use of built- in MATLAB functions and programming constructs, this section covers standard techniques for handling error conditions. Tools for debugging code, diagnosing problems, and measuring performance are an integral part of this section.
- Warnings and errors
- Handling errors
- The try-catch construct
- The MException object
- M-Lint code check
- Directory reports
- Assessing performance
- MATLAB Profiler
Structuring Data
Objective: This section examines choices for storing data within a MATLAB application. The choice of data type plays an important role in an application's storage requirements and execution time. Highlighted in this section are common techniques for improving performance when storing, accessing, and processing data.
- Cell arrays
- Structure arrays
- Other data containers
- Preallocation
- Vectorization
- Memory management
- Passing data between functions
Structuring Code
Objective: Modularizing code into readable and maintainable functions is often an important design consideration. This section explores the different function types available in the MATLAB language, and discusses some of the tradeoffs associated with these choices. The effect of the function type on performance, as well as reliability of the code, is an important theme throughout the chapter.
- Private functions
- Function handles
- Subfunctions
- Nested functions
- Precedence rules
- Variable scope and visibility
- Comparison of function types
Exercises
• Practice exercises
• Application-specific exercises
• Case studies
|
