Matlab for Engineers Gari Clifford 2010 2014 Centre

Matlab for Engineers Gari Clifford © 2010 - 2014 Centre for Doctoral Training in Healthcare Innovation Institute of Biomedical Engineering Department of Engineering Science University of Oxford Supported by the RCUK Digital Economy Programme grant number EP/G 036861/1

Overview of course – caveats and introductions • • First – this is the 1 st year we have run it, bug-ironing to do together. There may be licensing issues (which is one reason for this course). There will be timing changes as the day progresses. We will not cover all the material today – there will be homework! Course & Instructors: 1. Mauro Villarroel (RA working with Prof Tarassenko) • Coding structure, commenting, readability and testing • Debugging & profiling, 2. Alistair Johnson (3 rd year DPhil working with me) • Vectorisation, Parallelisation, Compilation, Mex, • … & me 1. MCC, The Linux environment & Cluster Computing

Overview of course – Why this course? Learning objectives: • To make your code readable for others to use and for you to re-use later (and hence save you and your research group a lot of time)! • To speed up your code – you can often reduce the execution time from days to hours – meaning you will finish your DPhil in 3 years! • To learn how to deal with finite resources like licenses, computational power, memory, execution time • Learn how to deal with data and data structures properly • Learn how to debug, profile & optimise your code (before we let you on the cluster) • Find out how to run parallel jobs on your own machine • Find out how to start using a remote cluster and run some basic scripts/programs

Overview of course – timetable & structure (AM) • Each of us will take about ¼ of the day each • You will work through examples on your computer as you go. • We will circulate to check you are doing OK Part 1: The basics [9. 20 -10. 30 am] • Coding styles – Making your code readable for you and your friends • Data Types – 15 different data types and how to convert them • Importing and Exporting Data – shortcuts and complex tasks Part 2: Debugging and optimising code [11 am-12. 30 pm] • Common errors • The Matlab debugger • The Matlab profiler • Preallocation & Vectorisation Lunch: [12. 30 -1. 25 pm]

Overview of course – timetable & structure (PM) Part 3: Improving your code further Functions vs scripts [1. 30 pm-2. 30 pm] • Commenting your code • Function types • Function Input/Output • Input parsing • Help files Parallel computing [2. 30 pm - 3. 15 pm] • Simple parallelisation - parfor details • Parallelization exercise • Parallel crashes, profiling, tips, and a remote demo • Licensing • Graphics Processing Unit Computation • MEXing – compiling external code to run in Matlab Part 4: Compiling and deploying code [3. 45 pm-5 pm] • Linux tools – shells, virtual shells, scripting, and batch jobs • MCC – the Matlab compiler • Cluster computing at the IBME