There are five two-hour lab sessions. These occur on week
A Tuesdays from 3-5 in LF31 (Sun Lab).
There are three lab exercises for this course.
Exercise 1 is on Bayesian classification; this lasts for two lab
sessions. Exercise 2 is on multi-layer perceptron neural networks; it lasts for two
lab sessions. Exercise 3 is on genetic algorithms; it lasts for a
single session. The labs use the Matlab software on Sun workstations - a tutorial for this package
will be handed out during lectures.
The labs will be marked in the lab sessions, that is you will show me
or a demonstrator your results. In addition, send the following to the
- Exercise 1:
- evaluate.m -- the Matlab code which computes the feature
which is used to classify the images
- Exercise 2:
- script.m -- a matlab script containing the
procedure which gave you the best generalization performance.
- Exercise 3:
- fitness.m -- the fitness function of the genetic
algorithm as a matlab function.