Tutorial 3 Matrix Op

Octave Tutorial 3: Vector and Matrix Operations
with extracts from Introduction to Octave, by P.J.G. Long
In this tutorial you will learn how to distinguish between componentwise operations on matrices and matrix operations that are not componentwise. perform basic operations on vectors and matrices using the operators + - * ^ and . perform specic operations on vectors using built-in functions. Storing a list of numbers in one vector allows Octave to use some of its more powerful features to perform calculations. If you are familiar with C or C++, you know that if you wanted, say, to multiply each element of an array by 2, you would have to use a for loop to step through each element. This can also be done in Octave, but it is much better (and faster) to make use of Octaves vector operators. Matrix operations in Octave follow the rules of linear algebra. Lets start with addition and scalar multiplication of vectors. Recall that these are componentwise operations. This means that if we add two vectors X and Y we simply add each component of X to the corresponding component of Y . Matrix addition is dened the same way. Notice that in order to add two matrices, their dimensions must be identical. Try it! Dene octave#:#> X= [1 2 3]; octave#:#> Y= [2 2 2]; Now add the two vectors by typing octave#:#> X+Y Scalar multiplication and division are also componentwise operations. For example, to multiply all elements of the previously dened vector X by 3, simply type octave#:#> X*3 What happens if you try to multiply two vectors? Try typing octave#:#> X*Y Octave complains about this because here the single asterisk * denotes matrix multiplication. As you know, matrix multiplication is not a componentwise operation, instead it is dened only if the dimensions of the matrices satisfy certain conditions. In order to perform the multiplication X*Y, vector Y would have to be a 3 by 1 matrix (i.e. a 3 row column vector). This means that the command octave#:#> X*Y where we used the transpose operator , would successfully show the result of the matrix 2 multiplication [1 2 3] 2 . 2 If we insist on wanting to do a componentwise multiplication of matrices, we need to use the operator .* (a dot followed by a single asterisk). Try octave#:#> X.*Y and check that each component of the result is the product of the corresponding components of X and Y. 1
The dot in front of operators such as * / and ^ means its is an element-by-element operation. Of course, you can always perform operations on individual elements of a vector or matrix. For example, octave#:#> X(2)/Y(3) will divide the second element of vector X by the third element of Y. Now lets move on to more sophisticated operations. What happens if you pass a vector to a built-in function, e.g. log which is the function that computes the natural logarithm ? Try it! Lets redene X octave#:#> X= [1 2 3]; Now type octave#:#> log(X) ans = 0.00000 0.69315 1.09861 Octave will return a vector of the same size in which each entry is found by performing the specied operation on the corresponding entry of the original vector. Another example, the sine function: octave#:#> sin(X) ans = 0.84147 0.90930 0.14112 For a list of predened functions that can be used in Octave, see Tutorial 1. The ability to work with these vector functions is one of the advantages of using Octave. Now complex operations can be dened that can be done quickly and easily. For example, image youd like to know how a complicated function behaves on a certain interval, e.g. x sin x y= for x > 0. The simplest way would be to plot its graph over a suciently large 1 + cos x positive interval of values of x. In Octave this can be done with three simple commands. Start by dening a vector x, for example octave#:#> x=[0:0.1:100]; Then build a vector y containing the values of the function y = f (x) at values of x specied by the elements of x. This sounds complicated but in fact its just one command octave#:#> y= x.*sin(x)./(1+cos(x)); Note that to perform the multiplication between x and sin x and the division between the numerator and denominator we need to use componentwise operators, so for the multiplication it is not simply * but .* and so on. Now plot the values of y against the corresponding values of x using the plot command, octave#:#> plot(x,y) The plot command will open a new window showing the graph of the function over the interval (0, 100) at steps of 0.1. Easy, isnt? You can then tweak the graph as you like. If you want to plot the graph on a dierent interval of values of x, change the length of x; if you want a more rene grid to plot your graph, change the increment step in the denition of x.
Practice Problems 1. Given X=[1 4 6] and Y=[-1 3 10], use Octave to nd the elements of the vector obtained by adding X to twice Y. 2. Given X = [1, 4] and A = matrix 2XA. 3. Given a vector X, explain the output of the command X.^ 2. Test your answer on a few vectors. 4. Given the vectors X and Y dened in Exercise 1, what is the command required to divide each element of X by the corresponding element of Y? 5. The Octave function sum(X) returns the sum of the elements in vector X. Use this function to compute the sum over the columns of the matrix A=[2 4 1; 6 7 2; 3 5 9]. 6. Let x = [2 9 1 16]. Use an Octave command to compute the square of each element of x. Then use the function sqrt to compute the square root of each element of x in the most ecient way. 7. Let x = [2 5 1 6] and y = [4 1 3 5]. Use an Octave command to divide each element of y by the corresponding element in x. Perform this calculation in the most ecient way. 8. Plot the sine function y = sin x over its period. Then plot the function y = sin the same interval. 9. Plot the function y = cos(x2 ) over the interval (0, 2).
x 2
3 1 6 , use Octave to compute the elements of the 5 2 7
on

Tutorial 3 Matrix Op

Uploaded by

Copyright:

Available Formats

Tutorial 3 Matrix Op

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tutorial 3 Matrix Op

Uploaded by

Copyright:

Available Formats

Octave Tutorial 3: Vector and Matrix Operations

with extracts from Introduction to Octave, by P.J.G. Long

3 1 6 , use Octave to compute the elements of the 5 2 7

You might also like