MATLAB Notes For Professionals

MATLAB
MATLAB
Notes for Professionals
®
Notes for Professionals
100+ pages
of professional hints and tricks
Disclaimer
GoalKicker.com This is an unocial free book created for educational purposes and is
not aliated with ocial MATLAB® group(s) or company(s).
Free Programming Books All trademarks and registered trademarks are
the property of their respective owners
Contents
About ................................................................................................................................................................................... 1
Chapter 1: Getting started with MATLAB Language ................................................................................... 2
Section 1.1: Indexing matrices and arrays ................................................................................................................... 2
Section 1.2: Anonymous functions and function handles .......................................................................................... 8
Section 1.3: Matrices and Arrays ................................................................................................................................ 11
Section 1.4: Cell arrays ................................................................................................................................................ 12
Section 1.5: Hello World ............................................................................................................................................... 14
Section 1.6: Scripts and Functions .............................................................................................................................. 14
Section 1.7: Helping yourself ....................................................................................................................................... 15
Section 1.8: Data Types ............................................................................................................................................... 16
Section 1.9: Reading Input & Writing Output ............................................................................................................ 19
Chapter 2: Functions ................................................................................................................................................. 21
Section 2.1: nargin, nargout ........................................................................................................................................ 21
Chapter 3: Documenting functions .................................................................................................................... 23
Section 3.1: Obtaining a function signature .............................................................................................................. 23
Section 3.2: Simple Function Documentation ........................................................................................................... 23
Section 3.3: Local Function Documentation ............................................................................................................. 23
Section 3.4: Documenting a Function with an Example Script ............................................................................... 24
Chapter 4: Using functions with logical output .......................................................................................... 27
Section 4.1: All and Any with empty arrays .............................................................................................................. 27
Chapter 5: Conditions ............................................................................................................................................... 28
Section 5.1: IF condition ............................................................................................................................................... 28
Section 5.2: IF-ELSE condition .................................................................................................................................... 28
Section 5.3: IF-ELSEIF condition ................................................................................................................................. 29
Section 5.4: Nested conditions ................................................................................................................................... 30
Chapter 6: Set operations ...................................................................................................................................... 33
Section 6.1: Elementary set operations ..................................................................................................................... 33
Chapter 7: For loops .................................................................................................................................................. 34
Section 7.1: Iterate over columns of matrix .............................................................................................................. 34
Section 7.2: Notice: Weird same counter nested loops ........................................................................................... 34
Section 7.3: Iterate over elements of vector ............................................................................................................. 35
Section 7.4: Nested Loops .......................................................................................................................................... 36
Section 7.5: Loop 1 to n ............................................................................................................................................... 37
Section 7.6: Loop over indexes ................................................................................................................................... 38
Chapter 8: Initializing Matrices or arrays ....................................................................................................... 39
Section 8.1: Creating a matrix of 0s ........................................................................................................................... 39
Section 8.2: Creating a matrix of 1s ........................................................................................................................... 39
Section 8.3: Creating an identity matrix .................................................................................................................... 39
Chapter 9: Matrix decompositions ..................................................................................................................... 40
Section 9.1: Schur decomposition .............................................................................................................................. 40
Section 9.2: Cholesky decomposition ........................................................................................................................ 40
Section 9.3: QR decomposition .................................................................................................................................. 41
Section 9.4: LU decomposition ................................................................................................................................... 41
Section 9.5: Singular value decomposition ............................................................................................................... 42
Chapter 10: Object-Oriented Programming .................................................................................................. 43
Section 10.1: Value vs Handle classes ........................................................................................................................ 43
Section 10.2: Constructors .......................................................................................................................................... 43
Section 10.3: Defining a class ..................................................................................................................................... 45
Section 10.4: Inheriting from classes and abstract classes ..................................................................................... 46
Chapter 11: Vectorization ........................................................................................................................................ 50
Section 11.1: Use of bsxfun ........................................................................................................................................... 50
Section 11.2: Implicit array expansion (broadcasting) [R2016b] ............................................................................. 51
Section 11.3: Element-wise operations ....................................................................................................................... 52
Section 11.4: Logical Masking ...................................................................................................................................... 53
Section 11.5: Sum, mean, prod & co ........................................................................................................................... 54
Section 11.6: Get the value of a function of two or more arguments ..................................................................... 55
Chapter 12: Graphics: 2D Line Plots ................................................................................................................... 56
Section 12.1: Split line with NaNs ................................................................................................................................. 56
Section 12.2: Multiple lines in a single plot ................................................................................................................ 56
Section 12.3: Custom colour and line style orders ................................................................................................... 57
Chapter 13: Graphics: 2D and 3D Transformations ................................................................................... 61
Section 13.1: 2D Transformations ............................................................................................................................... 61
Chapter 14: Controlling Subplot coloring in Matlab ................................................................................. 64
Section 14.1: How it's done .......................................................................................................................................... 64
Chapter 15: Image processing .............................................................................................................................. 65
Section 15.1: Basic image I/O ...................................................................................................................................... 65
Section 15.2: Retrieve Images from the Internet ...................................................................................................... 65
Section 15.3: Filtering Using a 2D FFT ....................................................................................................................... 65
Section 15.4: Image Filtering ....................................................................................................................................... 66
Section 15.5: Measuring Properties of Connected Regions ..................................................................................... 67
Chapter 16: Drawing .................................................................................................................................................. 70
Section 16.1: Circles ...................................................................................................................................................... 70
Section 16.2: Arrows ..................................................................................................................................................... 71
Section 16.3: Ellipse ...................................................................................................................................................... 74
Section 16.4: Pseudo 4D plot ...................................................................................................................................... 75
Section 16.5: Fast drawing .......................................................................................................................................... 79
Section 16.6: Polygon(s) .............................................................................................................................................. 80
Chapter 17: Debugging ............................................................................................................................................. 82
Section 17.1: Working with Breakpoints ..................................................................................................................... 82
Section 17.2: Debugging Java code invoked by MATLAB ....................................................................................... 84
Chapter 18: Financial Applications ..................................................................................................................... 87
Section 18.1: Random Walk ......................................................................................................................................... 87
Section 18.2: Univariate Geometric Brownian Motion .............................................................................................. 87
Chapter 19: Fourier Transforms and Inverse Fourier Transforms .................................................... 89
Section 19.1: Implement a simple Fourier Transform in Matlab .............................................................................. 89
Section 19.2: Images and multidimensional FTs ...................................................................................................... 90
Section 19.3: Inverse Fourier Transforms .................................................................................................................. 95
Chapter 20: Ordinary Dierential Equations (ODE) Solvers ................................................................ 97
Section 20.1: Example for odeset ............................................................................................................................... 97
Chapter 21: Interpolation with MATLAB ........................................................................................................... 99
Section 21.1: Piecewise interpolation 2 dimensional ................................................................................................. 99
Section 21.2: Piecewise interpolation 1 dimensional ............................................................................................... 101
Section 21.3: Polynomial interpolation .................................................................................................................... 106
Chapter 22: Integration ......................................................................................................................................... 110
Section 22.1: Integral, integral2, integral3 ............................................................................................................... 110
Chapter 23: Reading large files ......................................................................................................................... 112
Section 23.1: textscan ................................................................................................................................................ 112
Section 23.2: Date and time strings to numeric array fast ................................................................................... 112
Chapter 24: Usage of àccumarray()` Function ........................................................................................ 114
Section 24.1: Apply Filter to Image Patches and Set Each Pixel as the Mean of the Result of Each Patch
............................................................................................................................................................................. 114
Section 24.2: Finding the maximum value among elements grouped by another vector ................................ 115
Chapter 25: Introduction to MEX API .............................................................................................................. 116
Section 25.1: Check number of inputs/outputs in a C++ MEX-file ........................................................................ 116
Section 25.2: Input a string, modify it in C, and output it ...................................................................................... 117
Section 25.3: Passing a struct by field names ........................................................................................................ 118
Section 25.4: Pass a 3D matrix from MATLAB to C ............................................................................................... 118
Chapter 26: Performance and Benchmarking ........................................................................................... 121
Section 26.1: Identifying performance bottlenecks using the Profiler ................................................................. 121
Section 26.2: Comparing execution time of multiple functions ............................................................................ 124
Section 26.3: The importance of preallocation ...................................................................................................... 125
Section 26.4: It's ok to be `single`! ............................................................................................................................ 127
Chapter 27: Multithreading ................................................................................................................................. 130
Section 27.1: Using parfor to parallelize a loop ...................................................................................................... 130
Section 27.2: Executing commands in parallel using a "Single Program, Multiple Data" (SPMD) statement
............................................................................................................................................................................. 130
Section 27.3: Using the batch command to do various computations in parallel ............................................. 131
Section 27.4: When to use parfor ............................................................................................................................ 131
Chapter 28: Using serial ports ........................................................................................................................... 133
Section 28.1: Creating a serial port on Mac/Linux/Windows ............................................................................... 133
Section 28.2: Chosing your communication mode ................................................................................................ 133
Section 28.3: Automatically processing data received from a serial port .......................................................... 136
Section 28.4: Reading from the serial port ............................................................................................................. 137
Section 28.5: Closing a serial port even if lost, deleted or overwritten ............................................................... 137
Section 28.6: Writing to the serial port .................................................................................................................... 137
Chapter 29: Undocumented Features ............................................................................................................ 138
Section 29.1: Color-coded 2D line plots with color data in third dimension ........................................................ 138
Section 29.2: Semi-transparent markers in line and scatter plots ....................................................................... 138
Section 29.3: C++ compatible helper functions ...................................................................................................... 140
Section 29.4: Scatter plot jitter ................................................................................................................................. 141
Section 29.5: Contour Plots - Customise the Text Labels ...................................................................................... 141
Section 29.6: Appending / adding entries to an existing legend ......................................................................... 143
Chapter 30: MATLAB Best Practices ............................................................................................................... 144
Section 30.1: Indent code properly .......................................................................................................................... 144
Section 30.2: Avoid loops ......................................................................................................................................... 145
Section 30.3: Keep lines short .................................................................................................................................. 145
Section 30.4: Use assert ........................................................................................................................................... 146
Section 30.5: Block Comment Operator ................................................................................................................. 146
Section 30.6: Create Unique Name for Temporary File ........................................................................................ 147
Chapter 31: MATLAB User Interfaces .............................................................................................................. 149
Section 31.1: Passing Data Around User Interface ................................................................................................. 149
Section 31.2: Making a button in your UI that pauses callback execution .......................................................... 151
Section 31.3: Passing data around using the "handles" structure ........................................................................ 152
Section 31.4: Performance Issues when Passing Data Around User Interface ................................................... 153
Chapter 32: Useful tricks ....................................................................................................................................... 155
Section 32.1: Extract figure data .............................................................................................................................. 155
Section 32.2: Code Folding Preferences ................................................................................................................. 156
Section 32.3: Functional Programming using Anonymous Functions ................................................................. 158
Section 32.4: Save multiple figures to the same .fig file ........................................................................................ 158
Section 32.5: Comment blocks ................................................................................................................................. 159
Section 32.6: Useful functions that operate on cells and arrays ......................................................................... 160
Chapter 33: Common mistakes and errors ................................................................................................. 163
Section 33.1: The transpose operators .................................................................................................................... 163
Section 33.2: Do not name a variable with an existing function name ............................................................... 163
Section 33.3: Be aware of floating point inaccuracy ............................................................................................. 164
Section 33.4: What you see is NOT what you get: char vs cellstring in the command window ........................ 165
Section 33.5: Undefined Function or Method X for Input Arguments of Type Y ................................................. 166
Section 33.6: The use of "i" or "j" as imaginary unit, loop indices or common variable .................................... 167
Section 33.7: Not enough input arguments ............................................................................................................ 170
Section 33.8: Using `length` for multidimensional arrays ...................................................................................... 171
Section 33.9: Watch out for array size changes .................................................................................................... 171
Credits ............................................................................................................................................................................ 173
You may also like ...................................................................................................................................................... 175
About
Please feel free to share this PDF with anyone for free,
latest version of this book can be downloaded from:
http://GoalKicker.com/MATLABBook
This MATLAB® Notes for Professionals book is compiled from Stack Overflow
Documentation, the content is written by the beautiful people at Stack Overflow.
Text content is released under Creative Commons BY-SA, see credits at the end
of this book whom contributed to the various chapters. Images may be copyright
of their respective owners unless otherwise specified
This is an unofficial free book created for educational purposes and is not
affiliated with official MATLAB® group(s) or company(s) nor Stack Overflow. All
trademarks and registered trademarks are the property of their respective
company owners
The information presented in this book is not guaranteed to be correct nor

accurate, use at your own risk
Please send feedback and corrections to [email protected]
MATLAB® Notes for Professionals 1

Chapter 1: Getting started with MATLAB
Language
Version Release Release Date
1.0 1984-01-01
2 1986-01-01
3 1987-01-01
3.5 1990-01-01
4 1992-01-01
4.2c 1994-01-01
5.0 Volume 8 1996-12-01
5.1 Volume 9 1997-05-01
5.1.1 R9.1 1997-05-02
5.2 R10 1998-03-01
5.2.1 R10.1 1998-03-02
5.3 R11 1999-01-01
5.3.1 R11.1 1999-11-01
6.0 R12 2000-11-01
6.1 R12.1 2001-06-01
6.5 R13 2002-06-01
6.5.1 R13SP2 2003-01-01
6.5.2 R13SP2 2003-01-02
7 R14 2006-06-01
7.0.4 R14SP1 2004-10-01
7.1 R14SP3 2005-08-01
7.2 R2006a 2006-03-01
7.3 R2006b 2006-09-01
7.4 R2007a 2007-03-01
7.5 R2007b 2007-09-01
7.6 R2008a 2008-03-01
7.7 R2008b 2008-09-01
7.8 R2009a 2009-03-01
7.9 R2009b 2009-09-01
7.10 R2010a 2010-03-01
7.11 R2010b 2010-09-01
7.12 R2011a 2011-03-01
7.13 R2011b 2011-09-01
7.14 R2012a 2012-03-01
8.0 R2012b 2012-09-01
8.1 R2013a 2013-03-01
8.2 R2013b 2013-09-01
8.3 R2014a 2014-03-01
8.4 R2014b 2014-09-01
8.5 R2015a 2015-03-01
8.6 R2015b 2015-09-01
9.0 R2016a 2016-03-01
9.1 R2016b 2016-09-14
9.2 R2017a 2017-03-08
See also: MATLAB release history on Wikipedia.
Section 1.1: Indexing matrices and arrays

MATLAB allows for several methods to index (access) elements of matrices and arrays:

Subscript indexing - where you specify the position of the elements you want in each dimension of the
matrix separately.
Linear indexing - where the matrix is treated as a vector, no matter its dimensions. That means, you specify
each position in the matrix with a single number.
Logical indexing - where you use a logical matrix (and matrix of true and false values) with the identical
dimensions of the matrix you are trying to index as a mask to specify which value to return.
These three methods are now explained in more detail using the following 3-by-3 matrix M as an example:
>> M = magic(3)
ans =
8 1 6
3 5 7
4 9 2
Subscript indexing
The most straight-forward method for accessing an element, is to specify its row-column index. For example,
accessing the element on the second row and third column:
>> M(2, 3)
ans =
The number of subscripts provided exactly matches the number of dimensions M has (two in this example).
Note that the order of subscripts is the same as the mathematical convention: row index is the first. Moreover,
MATLAB indices starts with 1 and not 0 like most programming languages.
You can index multiple elements at once by passing a vector for each coordinate instead of a single number. For
example to get the entire second row, we can specify that we want the first, second and third columns:
>> M(2, [1,2,3])
ans =
3 5 7
In MATLAB, the vector [1,2,3] is more easily created using the colon operator, i.e. 1:3. You can use this in indexing
as well. To select an entire row (or column), MATLAB provides a shortcut by allowing you just specify :. For example,
the following code will also return the entire second row
>> M(2, :)
ans =
3 5 7
MATLAB also provides a shortcut for specifying the last element of a dimension in the form of the end keyword. The
end keyword will work exactly as if it was the number of the last element in that dimension. So if you want all the
columns from column 2 to the last column, you can use write the following:
>> M(2, 2:end)

ans =
5 7
Subscript indexing can be restrictive as it will not allow to extract single values from different columns and rows; it
will extract the combination of all rows and columns.
>> M([2,3], [1,3])

ans =
3 7
4 2
For example subscript indexing cannot extract only the elements M(2,1) or M(3,3). To do this we must consider
linear indexing.
Linear indexing
MATLAB allows you to treat n-dimensional arrays as one-dimensional arrays when you index using only one
dimension. You can directly access the first element:
>> M(1)
ans =
Note that arrays are stored in column-major order in MATLAB which means that you access the elements by first
going down the columns. So M(2) is the second element of the first column which is 3 and M(4) will be the first
element of the second column i.e.
>> M(4)
ans =
There exist built-in functions in MATLAB to convert subscript indices to linear indices, and vice versa: sub2ind and
ind2sub respectively. You can manually convert the subscripts (r,c) to a linear index by
idx = r + (c-1)*size(M,1)
To understand this, if we are in the first column then the linear index will simply be the row index. The formula
above holds true for this because for c == 1, (c-1) == 0. In the next columns, the linear index is the row number
plus all the rows of the previous columns.
Note that the end keyword still applies and now refers to the very last element of the array i.e. M(end) == M(end,
end) == 2.
You can also index multiple elements using linear indexing. Note that if you do that, the returned matrix will have
the same shape as the matrix of index vectors.
M(2:4) returns a row vector because 2:4 represents the row vector [2,3,4]:
>> M(2:4)

ans =
3 4 1
As an other example, M([1,2;3,4]) returns a 2-by-2 matrix because [1,2;3,4] is a 2-by-2 matrix as well. See the
below code to convince yourself:
>> M([1,2;3,4])
ans =
8 3
4 1
Note that indexing with : alone will always return a column vector:
>> M(:)
ans =
8
3
4
1
5
9
6
7
2
This example also illustrates the order in which MATLAB returns elements when using linear indexing.
Logical indexing
The third method of indexing is to use a logical matrix, i.e. a matrix containing only true or false values, as a mask
to filter out the elements you don't want. For example, if we want to find all the elements of M that are greater than
5 we can use the logical matrix
>> M > 5
ans =
1 0 1
0 0 1
0 1 0
to index M and return only the values that are greater than 5 as follows:
>> M(M > 5)
ans =
8
9
6
7
If you wanted these number to stay in place (i.e. keep the shape of the matrix), then you could assign to the logic

compliment
>> M(~(M > 5)) = NaN
ans =
8 NaN 6
NaN NaN 7
NaN 9 Nan
We can reduce complicated code blocks containing if and for statements by using logical indexing.
Take the non-vectorized (already shortened to a single loop by using linear indexing):
for elem = 1:numel(M)

if M(elem) > 5
M(elem) = M(elem) - 2;
end
end
This can be shortened to the following code using logical indexing:
idx = M > 5;
M(idx) = M(idx) - 2;
Or even shorter:
M(M > 5) = M(M > 5) - 2;
More on indexing
Higher dimension matrices
All the methods mentioned above generalize into n-dimensions. If we use the three-dimensional matrix M3 =
rand(3,3,3) as an example, then you can access all the rows and columns of the second slice of the third
dimension by writing
>> M(:,:,2)
You can access the first element of the second slice using linear indexing. Linear indexing will only move on to the
second slice after all the rows and all the columns of the first slice. So the linear index for that element is
>> M(size(M,1)*size(M,2)+1)
In fact, in MATLAB, every matrix is n-dimensional: it just happens to be that the size of most of the other n-
dimensions are one. So, if a = 2 then a(1) == 2 (as one would expect), but also a(1, 1) == 2, as does a(1, 1, 1)
== 2, a(1, 1, 1, ..., 1) == 2 and so on. These "extra" dimensions (of size 1), are referred to as singleton
dimensions. The command squeeze will remove them, and one can use permute to swap the order of dimensions
around (and introduce singleton dimensions if required).
An n-dimensional matrix can also be indexed using an m subscripts (where m<=n). The rule is that the first m-1
subscripts behave ordinarily, while the last (m'th) subscript references the remaining (n-m+1) dimensions, just as a
linear index would reference an (n-m+1) dimensional array. Here is an example:
>> M = reshape(1:24,[2,3,4]);
>> M(1,1)

ans =
1
>> M(1,10)
ans =
19
>> M(:,:)
ans =
1 3 5 7 9 11 13 15 17 19 21 23
2 4 6 8 10 12 14 16 18 20 22 24
Returning ranges of elements
With subscript indexing, if you specify more than one element in more than one dimension, MATLAB returns each
possible pair of coordinates. For example, if you try M([1,2],[1,3]) MATLAB will return M(1,1) and M(2,3) but it will
also return M(1,3) and M(2,1). This can seem unintuitive when you are looking for the elements for a list of
coordinate pairs but consider the example of a larger matrix, A = rand(20) (note A is now 20-by-20), where you
want to get the top right hand quadrant. In this case instead of having to specify every coordinate pair in that
quadrant (and this this case that would be 100 pairs), you just specify the 10 rows and the 10 columns you want so
A(1:10, 11:end). Slicing a matrix like this is far more common than requiring a list of coordinate pairs.
In the event that you do want to get a list of coordinate pairs, the simplest solution is to convert to linear indexing.
Consider the problem where you have a vector of column indices you want returned, where each row of the vector
contains the column number you want returned for the corresponding row of the matrix. For example
colIdx = [3;2;1]
So in this case you actually want to get back the elements at (1,3), (2,2) and (3,1). So using linear indexing:
>> colIdx = [3;2;1];

>> rowIdx = 1:length(colIdx);
>> idx = sub2ind(size(M), rowIdx, colIdx);
>> M(idx)
ans =
6 5 4
Returning an element multiple times
With subscript and linear indexing you can also return an element multiple times by repeating it's index so
>> M([1,1,1,2,2,2])
ans =
8 8 8 3 3 3
You can use this to duplicate entire rows and column for example to repeat the first row and last column
>> M([1, 1:end], [1:end, end])
ans =
8 1 6 6
8 1 6 6
3 5 7 7
4 9 2 2

For more information, see here.
Section 1.2: Anonymous functions and function handles

Basics
Anonymous functions are a powerful tool of the MATLAB language. They are functions that exist locally, that is: in
the current workspace. However, they do not exist on the MATLAB path like a regular function would, e.g. in an m-
file. That is why they are called anonymous, although they can have a name like a variable in the workspace.
The @ operator
Use the @ operator to create anonymous functions and function handles. For example, to create a handle to the sin
function (sine) and use it as f:
>> f = @sin
f =
@sin
Now f is a handle to the sin function. Just like (in real life) a door handle is a way to use a door, a function handle is
a way to use a function. To use f, arguments are passed to it as if it were the sin function:
>> f(pi/2)
ans =
1
f accepts any input arguments the sin function accepts. If sin would be a function that accepts zero input
arguments (which it does not, but others do, e.g. the peaks function), f() would be used to call it without input
arguments.
Custom anonymous functions

Anonymous functions of one variable
It is not obviously useful to create a handle to an existing function, like sin in the example above. It is kind of
redundant in that example. However, it is useful to create anonymous functions that do custom things that
otherwise would need to be repeated multiple times or created a separate function for. As an example of a custom
anonymous function that accepts one variable as its input, sum the sine and cosine squared of a signal:
>> f = @(x) sin(x)+cos(x).^2

f =
@(x)sin(x)+cos(x).^2
Now f accepts one input argument called x. This was specified using parentheses (...) directly after the @
operator. f now is an anonymous function of x: f(x). It is used by passing a value of x to f:
>> f(pi)
ans =
1.0000
A vector of values or a variable can also be passed to f, as long as they are used in a valid way within f:
>> f(1:3) % pass a vector to f

ans =
1.1334 1.0825 1.1212
>> n = 5:7;

>> f(n) % pass n to f
ans =
-0.8785 0.6425 1.2254
Anonymous functions of more than one variable
In the same fashion anonymous functions can be created to accept more than one variable. An example of an
anonymous function that accepts three variables:
>> f = @(x,y,z) x.^2 + y.^2 - z.^2

f =
@(x,y,z)x.^2+y.^2-z.^2
>> f(2,3,4)
ans =
-3
Parameterizing anonymous functions
Variables in the workspace can be used within the definition of anonymous functions. This is called parameterizing.
For example, to use a constant c = 2 in an anonymous function:
>> c = 2;
>> f = @(x) c*x
f =
@(x)c*x
>> f(3)
ans =
6
f(3) used the variable c as a parameter to multiply with the provided x. Note that if the value of c is set to
something different at this point, then f(3) is called, the result would not be different. The value of c is the value at
the time of creation of the anonymous function:
>> c = 2;
>> f = @(x) c*x;
>> f(3)
ans =
6
>> c = 3;
>> f(3)
ans =
6
Input arguments to an anonymous function do not refer to workspace variables
Note that using the name of variables in the workspace as one of the input arguments of an anonymous function
(i.e., using @(...)) will not use those variables' values. Instead, they are treated as different variables within the
scope of the anonymous function, that is: the anonymous function has its private workspace where the input
variables never refer to the variables from the main workspace. The main workspace and the anonymous function's
workspace do not know about each other's contents. An example to illustrate this:
>> x = 3 % x in main workspace

x =
3
>> f = @(x) x+1; % here x refers to a private x variable
>> f(5)
ans =
6
>> x
x =

3
The value of x from the main workspace is not used within f. Also, in the main workspace x was left untouched.
Within the scope of f, the variable names between parentheses after the @ operator are independent from the
main workspace variables.
Anonymous functions are stored in variables
An anonymous function (or, more precisely, the function handle pointing at an anonymous function) is stored like
any other value in the current workspace: In a variable (as we did above), in a cell array ({@(x)x.^2,@(x)x+1}), or
even in a property (like h.ButtonDownFcn for interactive graphics). This means the anonymous function can be
treated like any other value. When storing it in a variable, it has a name in the current workspace and can be
changed and cleared just like variables holding numbers.
Put differently: A function handle (whether in the @sin form or for an anonymous function) is simply a value that
can be stored in a variable, just like a numerical matrix can be.
Advanced use
Passing function handles to other functions
Since function handles are treated like variables, they can be passed to functions that accept function handles as
input arguments.
An example: A function is created in an m-file that accepts a function handle and a scalar number. It then calls the
function handle by passing 3 to it and then adds the scalar number to the result. The result is returned.
Contents of funHandleDemo.m:
function y = funHandleDemo(fun,x)
y = fun(3);
y = y + x;
Save it somewhere on the path, e.g. in MATLAB's current folder. Now funHandleDemo can be used as follows, for
example:
>> f = @(x) x^2; % an anonymous function

>> y = funHandleDemo(f,10) % pass f and a scalar to funHandleDemo
y =
19
The handle of another existing function can be passed to funHandleDemo:
>> y = funHandleDemo(@sin,-5)
y =
-4.8589
Notice how @sin was a quick way to access the sin function without first storing it in a variable using f = @sin.
Using bsxfun, cellfun and similar functions with anonymous functions
MATLAB has some built-in functions that accept anonymous functions as an input. This is a way to perform many
calculations with a minimal number of lines of code. For example bsxfun, which performs element-by-element
binary operations, that is: it applies a function on two vectors or matrices in an element-by-element fashion.
Normally, this would require use of for-loops, which often requires preallocation for speed. Using bsxfun this
process is sped up. The following example illustrates this using tic and toc, two functions that can be used to time

how long code takes. It calculates the difference of every matrix element from the matrix column mean.
A = rand(50); % 50-by-50 matrix of random values between 0 and 1
% method 1: slow and lots of lines of code

tic
meanA = mean(A); % mean of every matrix column: a row vector
% pre-allocate result for speed, remove this for even worse performance
result = zeros(size(A));
for j = 1:size(A,1)
result(j,:) = A(j,:) - meanA;
end
toc
clear result % make sure method 2 creates its own result
% method 2: fast and only one line of code

tic
result = bsxfun(@minus,A,mean(A));
toc
Running the example above results in two outputs:
Elapsed time is 0.015153 seconds.

These lines come from the toc functions, which print the elapsed time since the last call to the tic function.
The bsxfun call applies the function in the first input argument to the other two input arguments. @minus is a long
name for the same operation as the minus sign would do. A different anonymous function or handle (@) to any
other function could have been specified, as long as it accepts A and mean(A) as inputs to generate a meaningful
result.
Especially for large amounts of data in large matrices, bsxfun can speed up things a lot. It also makes code look
cleaner, although it might be more difficult to interpret for people who don't know MATLAB or bsxfun. (Note that in
MATLAB R2016a and later, many operations that previously used bsxfun no longer need them; A-mean(A) works
directly and can in some cases be even faster.)
Section 1.3: Matrices and Arrays

In MATLAB, the most basic data type is the numeric array. It can be a scalar, a 1-D vector, a 2-D matrix, or an N-D
multidimensional array.
% a 1-by-1 scalar value

x = 1;
To create a row vector, enter the elements inside brackets, separated by spaces or commas:
% a 1-by-4 row vector

v = [1, 2, 3, 4];
v = [1 2 3 4];
To create a column vector, separate the elements with semicolons:
% a 4-by-1 column vector

v = [1; 2; 3; 4];

To create a matrix, we enter the rows as before separated by semicolons:
% a 2 row-by-4 column matrix

M = [1 2 3 4; 5 6 7 8];
% a 4 row-by-2 column matrix

M = [1 2; ...
4 5; ...
6 7; ...
8 9];
Notice you cannot create a matrix with unequal row / column size. All rows must be the same length, and all
columns must be the same length:
% an unequal row / column matrix

M = [1 2 3 ; 4 5 6 7]; % This is not valid and will return an error
% another unequal row / column matrix

M = [1 2 3; ...
4 5; ...
6 7 8; ...
9 10]; % This is not valid and will return an error
To transpose a vector or a matrix, we use the .'-operator, or the ' operator to take its Hermitian conjugate, which
is the complex conjugate of its transpose. For real matrices, these two are the same:
% create a row vector and transpose it into a column vector

v = [1 2 3 4].'; % v is equal to [1; 2; 3; 4];
% create a 2-by-4 matrix and transpose it to get a 4-by-2 matrix

M = [1 2 3 4; 5 6 7 8].'; % M is equal to [1 5; 2 6; 3 7; 4 8]
% transpose a vector or matrix stored as a variable

A = [1 2; 3 4];
B = A.'; % B is equal to [1 3; 2 4]
For arrays of more than two-dimensions, there is no direct language syntax to enter them literally. Instead we must
use functions to construct them (such as ones, zeros, rand) or by manipulating other arrays (using functions such
as cat, reshape, permute). Some examples:
% a 5-by-2-by-4-by-3 array (4-dimensions)

arr = ones(5, 2, 4, 3);
% a 2-by-3-by-2 array (3-dimensions)

arr = cat(3, [1 2 3; 4 5 6], [7 8 9; 0 1 2]);
% a 5-by-4-by-3-by-2 (4-dimensions)
arr = reshape(1:120, [5 4 3 2]);
Section 1.4: Cell arrays

Elements of the same class can often be concatenated into arrays (with a few rare exceptions, e.g. function
handles). Numeric scalars, by default of class double, can be stored in a matrix.
>> A = [1, -2, 3.14, 4/5, 5^6; pi, inf, 7/0, nan, log(0)]
A =
1.0e+04 *

0.0001 -0.0002 0.0003 0.0001 1.5625
0.0003 Inf Inf NaN -Inf
Characters, which are of class char in MATLAB, can also be stored in array using similar syntax. Such an array is
similar to a string in many other programming languages.
>> s = ['MATLAB ','is ','fun']

s =
MATLAB is fun
Note that despite both of them are using brackets [ and ], the result classes are different. Therefore the operations
that can be done on them are also different.
>> whos
Name Size Bytes Class Attributes
A 2x5 80 double
s 1x13 26 char
In fact, the array s is not an array of the strings 'MATLAB ','is ', and 'fun', it is just one string - an array of 13
characters. You would get the same results if it were defined by any of the following:
>> s = ['MAT','LAB ','is f','u','n'];

>> s = ['M','A','T','L','A','B,' ','i','s',' ','f','u','n'];
A regular MATLAB vector does not let you store a mix of variables of different classes, or a few different strings. This
is where the cell array comes in handy. This is an array of cells that each can contain some MATLAB object, whose
class can be different in every cell if needed. Use curly braces { and } around the elements to store in a cell array.
>> C = {A; s}
C =
[2x5 double]
'MATLAB is fun'
>> whos C
C 2x1 330 cell
Standard MATLAB objects of any classes can be stored together in a cell array. Note that cell arrays require more
memory to store their contents.
Accessing the contents of a cell is done using curly braces { and }.
>> C{1}
ans =
1.0e+04 *
0.0001 -0.0002 0.0003 0.0001 1.5625
0.0003 Inf Inf NaN -Inf
Note that C(1) is different from C{1}. Whereas the latter returns the cell's content (and has class double in out
example), the former returns a cell array which is a sub-array of C. Similarly, if D were an 10 by 5 cell array, then
D(4:8,1:3) would return a sub-array of D whose size is 5 by 3 and whose class is cell. And the syntax C{1:2} does
not have a single returned object, but rater it returns 2 different objects (similar to a MATLAB function with multiple
return values):
>> [x,y] = C{1:2}

x =
1 -2 3.14
0.8 15625
3.14159265358979 Inf Inf
NaN -Inf
y =
MATLAB is fun
Section 1.5: Hello World

Open a new blank document in the MATLAB Editor (in recent versions of MATLAB, do this by selecting the Home tab
of the toolstrip, and clicking on New Script). The default keyboard shortcut to create a new script is Ctrl-n.
Alternatively, typing edit myscriptname.m will open the file myscriptname.m for editing, or offer to create the file if
it does not exist on the MATLAB path.
In the editor, type the following:
disp('Hello, World!');
Select the Editor tab of the toolstrip, and click Save As. Save the document to a file in the current directory called
helloworld.m. Saving an untitled file will bring up a dialog box to name the file.
In the MATLAB Command Window, type the following:
>> helloworld
You should see the following response in the MATLAB Command Window:
Hello, World!
We see that in the Command Window, we are able to type the names of functions or script files that we have
written, or that are shipped with MATLAB, to run them.
Here, we have run the 'helloworld' script. Notice that typing the extension (.m) is unnecessary. The instructions held
in the script file are executed by MATLAB, here printing 'Hello, World!' using the disp function.
Script files can be written in this way to save a series of commands for later (re)use.
Section 1.6: Scripts and Functions

MATLAB code can be saved in m-files to be reused. m-files have the .m extension which is automatically associated
with MATLAB. An m-file can contain either a script or functions.
Scripts
Scripts are simply program files that execute a series of MATLAB commands in a predefined order.
Scripts do not accept input, nor do scripts return output. Functionally, scripts are equivalent to typing commands
directly into the MATLAB command window and being able to replay them.
An example of a script:
length = 10;
width = 3;

area = length * width;
This script will define length, width, and area in the current workspace with the value 10, 3, and 30 respectively.
As stated before, the above script is functionally equivalent to typing the same commands directly into the
command window.
>> length = 10;

>> width = 3;
>> area = length * width;
Functions
Functions, when compared to scripts, are much more flexible and extensible. Unlike scripts, functions can accept
input and return output to the caller. A function has its own workspace, this means that internal operations of the
functions will not change the variables from the caller.
All functions are defined with the same header format:
function [output] = myFunctionName(input)
The function keyword begins every function header. The list of outputs follows. The list of outputs can also be a
comma separated list of variables to return.
function [a, b, c] = myFunctionName(input)
Next is the name of the function that will be used for calling. This is generally the same name as the filename. For
example, we would save this function as myFunctionName.m.
Following the function name is the list of inputs. Like the outputs, this can also be a comma separated list.
function [a, b, c] = myFunctionName(x, y, z)
We can rewrite the example script from before as a reusable function like the following:
function [area] = calcRecArea(length, width)

area = length * width;
end
We can call functions from other functions, or even from script files. Here is an example of our above function
being used in a script file.
l = 100;
w = 20;
a = calcRecArea(l, w);
As before, we create l, w, and a in the workspace with the values of 100, 20, and 2000 respectively.
Section 1.7: Helping yourself

MATLAB comes with many built-in scripts and functions which range from simple multiplication to image
recognition toolboxes. In order to get information about a function you want to use type: help functionname in the
command line. Lets take the help function as an example.
Information on how to use it can be obtained by typing:

>> help help
in the command window. This will return information of the usage of function help. If the information you are
looking for is still unclear you can try the documentation page of the function. Simply type:
>> doc help
in the command window. This will open the browsable documentation on the page for function help providing all
the information you need to understand how the 'help' works.
This procedure works for all built-in functions and symbols.
When developing your own functions you can let them have their own help section by adding comments at the top
of the function file or just after the function declaration.
Example for a simple function multiplyby2 saved in file multiplyby2.m
function [prod]=multiplyby2(num)
% function MULTIPLYBY2 accepts a numeric matrix NUM and returns output PROD
% such that all numbers are multiplied by 2
prod=num*2;
end
or
% function MULTIPLYBY2 accepts a numeric matrix NUM and returns output PROD
% such that all numbers are multiplied by 2
function [prod]=multiplyby2(num)
prod=num*2;
end
This is very useful when you pick up your code weeks/months/years after having written it.
The help and doc function provide a lot of information, learning how to use those features will help you progress
rapidly and use MATLAB efficiently.
Section 1.8: Data Types

There are 16 fundamental data types, or classes, in MATLAB. Each of these classes is in the form of a matrix or
array. With the exception of function handles, this matrix or array is a minimum of 0-by-0 in size and can grow to an
n-dimensional array of any size. A function handle is always scalar (1-by-1).
Important moment in MATLAB is that you don't need to use any type declaration or dimension statements by
default. When you define new variable MATLAB creates it automatically and allocates appropriate memory space.
Example:
a = 123;
b = [1 2 3];
c = '123';
>> whos
a 1x1 8 double

b 1x3 24 double
c 1x3 6 char
If the variable already exists, MATLAB replaces the original data with new one and allocates new storage space if
necessary.
Fundamental data types
Fundamental data types are: numeric, logical, char, cell, struct, table and function_handle.
Numeric data types:
Floating-Point numbers (default)
MATLAB represents floating-point numbers in either double-precision or single-precision format. The default
is double precision, but you can make any number single precision with a simple conversion function:
a = 1.23;
b = single(a);
>> whos
a 1x1 8 double
b 1x1 4 single
Integers
MATLAB has four signed and four unsigned integer classes. Signed types enable you to work with negative
integers as well as positive, but cannot represent as wide a range of numbers as the unsigned types because
one bit is used to designate a positive or negative sign for the number. Unsigned types give you a wider
range of numbers, but these numbers can only be zero or positive.
MATLAB supports 1-, 2-, 4-, and 8-byte storage for integer data. You can save memory and execution time for
your programs if you use the smallest integer type that accommodates your data. For example, you do not
need a 32-bit integer to store the value 100.
a = int32(100);
b = int8(100);
>> whos
a 1x1 4 int32
b 1x1 1 int8
To store data as an integer, you need to convert from double to the desired integer type. If the number being
converted to an integer has a fractional part, MATLAB rounds to the nearest integer. If the fractional part is
exactly 0.5, then from the two equally nearby integers, MATLAB chooses the one for which the absolute
value is larger in magnitude.
a = int16(456);
char

Character arrays provide storage for text data in MATLAB. In keeping with traditional programming
terminology, an array (sequence) of characters is defined as a string. There is no explicit string type in retail
releases of MATLAB.
logical: logical values of 1 or 0, represent true and false respectively. Use for relational conditions and array
indexing. Because it's just TRUE or FALSE it has size of 1 byte.
a = logical(1);
structure. A structure array is a data type that groups variables of different data types using data containers
called fields. Each field can contain any type of data. Access data in a structure using dot notation of the form
structName.fieldName.
field1 = 'first';
field2 = 'second';
value1 = [1 2 3 4 5];
value2 = 'sometext';
s = struct(field1,value1,field2,value2);
In order to access value1, each of the following syntax are equivalent
s.first or s.(field1) or s.('first')
We can explicitly access a field we know will exist with the first method, or either pass a string or create a
string to access the field in the second example. The third example is demostrating that the dot parenthases
notation takes a string, which is the same one stored in the field1 variable.
table variables can be of different sizes and data types, but all variables must have the same number of rows.
Age = [15 25 54]';

Height = [176 190 165]';
Name = {'Mike', 'Pete', 'Steeve'}';
T = table(Name,Age, Height);
cell. It's very useful MATLAB data type: cell array is an array each element of it can be of different data type
and size. It's very strong instrument for manipulating data as you wish.
a = { [1 2 3], 56, 'art'};
or
a = cell(3);
function handles stores a pointer to a function (for example, to anonymous function). It allows you to pass a
function to another function, or call local functions from outside the main function.
There are a lot of instruments to work with each data type and also built-in data type conversion functions
(str2double, table2cell).
Additional data types
There are several additional data types which are useful in some specific cases. They are:

Date and time: arrays to represent dates, time, and duration. datetime('now') returns 21-Jul-2016
16:30:16.
Categorical arrays: it's data type for storing data with values from a set of discrete categories. Useful for
storing nonnumeric data (memory effective). Can be used in a table to select groups of rows.
a = categorical({'a' 'b' 'c'});
Map containers is a data structure that has unique ability to indexing not only through the any scalar
numeric values but character vector. Indices into the elements of a Map are called keys. These keys, along
with the data values associated with them, are stored within the Map.
Time series are data vectors sampled over time, in order, often at regular intervals. It's useful to store the
data connected with timesteps and it has a lot of useful methods to work with.
Section 1.9: Reading Input & Writing Output

Just like all programming language, Matlab is designed to read and write in a large variety of formats. The native
library supports a large number of Text,Image,Video,Audio,Data formats with more formats included in each
version update - check here to see the full list of supported file formats and what function to use to import them.
Before you attempt to load in your file, you must ask yourself what do you want the data to become and how you
expect the computer to organize the data for you. Say you have a txt/csv file in the following format:
Fruit,TotalUnits,UnitsLeftAfterSale,SellingPricePerUnit
Apples,200,67,$0.14
Bananas,300,172,$0.11
Pineapple,50,12,$1.74
We can see that the first column is in the format of Strings, while the second, third are Numeric, the last column is
in the form of Currency. Let's say we want to find how much revenue we made today using Matlab and first we
want to load in this txt/csv file. After checking the link, we can see that String and Numeric type of txt files are
handled by textscan. So we could try:
fileID = fopen('dir/test.txt'); %Load file from dir

C = textscan(fileID,'%s %f %f %s','Delimiter',',','HeaderLines',1); %Parse in the txt/csv
where %s suggest that the element is a String type, %f suggest that the element is a Float type, and that the file is
Delimited by ",". The HeaderLines option asks Matlab to skip the First N lines while the 1 immediately after it means
to skip the first line (the header line).
Now C is the data we have loaded which is in the form of a Cell Array of 4 cells, each containing the column of data
in the txt/csv file.
So first we want to calculate how many fruits we sold today by subtracting the third column from the second
column, this can be done by:
sold = C{2} - C{3}; %C{2} gives the elements inside the second cell (or the second column)
Now we want to multiply this vector by the Price per unit, so first we need to convert that column of Strings into a
column of Numbers, then convert it into a Numeric Matrix using Matlab's cell2mat the first thing we need to do is
to strip-off the "$" sign, there are many ways to do this. The most direct way is using a simple regex:
D = cellfun(@(x)(str2num(regexprep(x, '\$',''))), C{4}, 'UniformOutput', false);%cellfun allows us

to avoid looping through each element in the cell.
Or you can use a loop:
for t=1:size(C{4},1)
D{t} = str2num(regexprep(C{4}{t}, '\$',''));
end
E = cell2mat(D)% converts the cell array into a Matrix
The str2num function turns the string which had "$" signs stripped into numeric types and cell2mat turns the cell
of numeric elements into a matrix of numbers
Now we we can multiply the units sold by the cost per unit:
revenue = sold .* E; %element-wise product is denoted by .* in Matlab
totalrevenue = sum(revenue);

Chapter 2: Functions
Section 2.1: nargin, nargout
In the body of a function nargin and nargout indicate respectively the actual number of input and output supplied
in the call.
We can for example control the execution of a function based on the number of provided input.
myVector.m:
function [res] = myVector(a, b, c)

% Roughly emulates the colon operator
switch nargin
case 1
res = [0:a];
case 2
res = [a:b];
case 3
res = [a:b:c];
otherwise
error('Wrong number of params');
end
end
terminal:
>> myVector(10)
ans =
0 1 2 3 4 5 6 7 8 9 10
>> myVector(10, 20)
ans =
10 11 12 13 14 15 16 17 18 19 20
>> myVector(10, 2, 20)
ans =
10 12 14 16 18 20
In a similar way we can control the execution of a function based on the number of output parameters.
myIntegerDivision:
function [qt, rm] = myIntegerDivision(a, b)

qt = floor(a / b);
if nargout == 2
rm = rem(a, b);
end
end

terminal:
>> q = myIntegerDivision(10, 7)
q = 1
>> [q, r] = myIntegerDivision(10, 7)
q = 1
r = 3

Chapter 3: Documenting functions
Section 3.1: Obtaining a function signature
It is often helpful to have MATLAB print the 1st line of a function, as this usually contains the function signature,
including inputs and outputs:
dbtype <functionName> 1
Example:
>> dbtype fit 1
1 function [fitobj,goodness,output,warnstr,errstr,convmsg] =
fit(xdatain,ydatain,fittypeobj,varargin)
Section 3.2: Simple Function Documentation

function output = mymult(a, b)
% MYMULT Multiply two numbers.
% output = MYMULT(a, b) multiplies a and b.
%
% See also fft, foo, sin.
%
% For more information, see <a href="matlab:web('https://google.com')">Google</a>.
output = a * b;
end
help mymult then provides:
mymult Multiply two numbers.
output = mymult(a, b) multiplies a and b.
See also fft, foo, sin.
For more information, see Google.
fft and sin automatically link to their respective help text, and Google is a link to google.com. foo will not link to
any documentation in this case, as long as there is not a documented function/class by the name of foo on the
search path.
Section 3.3: Local Function Documentation

In this example, documentation for the local function baz (defined in foo.m) can be accessed either by the resulting
link in help foo, or directly through help foo>baz.
function bar = foo

%This is documentation for FOO.
% See also foo>baz
% This won't be printed, because there is a line without % on it.

end

function baz
% This is documentation for BAZ.
end
Section 3.4: Documenting a Function with an Example Script

To document a function, it is often helpful to have an example script which uses your function. The publish function
in Matlab can then be used to generate a help file with embedded pictures, code, links, etc. The syntax for
documenting your code can be found here.
The Function This function uses a "corrected" FFT in Matlab.
function out_sig = myfft(in_sig)
out_sig = fftshift(fft(ifftshift(in_sig)));
end
The Example Script This is a separate script which explains the inputs, outputs, and gives an example explaining
why the correction is necessary. Thanks to Wu, Kan, the original author of this function.
%% myfft
% This function uses the "proper" fft in matlab. Note that the fft needs to
% be multiplied by dt to have physical significance.
% For a full description of why the FFT should be taken like this, refer
% to: Why_use_fftshift(fft(fftshift(x)))__in_Matlab.pdf included in the
% help folder of this code. Additional information can be found:
%
<https://www.mathworks.com/matlabcentral/fileexchange/25473-why-use-fftshift-fft-fftshift-x----in-m
atlab-instead-of-fft-x-->
%
%% Inputs
% *in_sig* - 1D signal
%
%% Outputs
% *out_sig* - corrected FFT of *in_sig*
%
%% Examples
% Generate a signal with an analytical solution. The analytical solution is
% then compared to the fft then to myfft. This example is a modified
% version given by Wu, Kan given in the link aboce.
%%
% Set parameters
fs = 500; %sampling frequency
dt = 1/fs; %time step
T=1; %total time window
t = -T/2:dt:T/2-dt; %time grids
df = 1/T; %freq step
Fmax = 1/2/dt; %freq window
f=-Fmax:df:Fmax-df; %freq grids, not used in our examples, could be used by plot(f, X)
%%
% Generate Gaussian curve
Bx = 10; A = sqrt(log(2))/(2*pi*Bx); %Characteristics of Gaussian curve
x = exp(-t.^2/2/A^2); %Create Gaussian Curve
%%
% Generate Analytical solution
Xan = A*sqrt(2*pi)*exp(-2*pi^2*f.^2*A^2); %X(f), real part of the analytical Fourier transform of
x(t)

%%
% Take FFT and corrected FFT then compare
Xfft = dt *fftshift(fft(x)); %FFT
Xfinal = dt * myfft(x); %Corrected FFT
hold on
plot(f,Xan);
plot(f,real(Xfft));
plot(f,real(Xfinal),'ro');
title('Comparison of Corrected and Uncorrected FFT');
legend('Analytical Solution','Uncorrected FFT','Corrected FFT');
xlabel('Frequency'); ylabel('Amplitude');
DT = max(f) - min(f);
xlim([-DT/4,DT/4]);
The Output The publish option can be found under the "Publish" tab, highlighted in the imageSimple Function
Documentation below.
Matlab will run the script, and save the images which are displayed, as well as the text generated by the command
line. The output can be saved to many different types of formats, including HTML, Latex, and PDF.
The output of the example script given above can be seen in the image below.

Chapter 4: Using functions with logical
output
Section 4.1: All and Any with empty arrays
Special care needs to be taken when there is a possibility that an array become an empty array when it comes to
logical operators. It is often expected that if all(A) is true then any(A) must be true and if any(A) is false, all(A)
must also be false. That is not the case in MATLAB with empty arrays.
>> any([])
ans =
0
>> all([])
ans =
1
So if for example you are comparing all elements of an array with a certain threshold, you need to be aware of the
case where the array is empty:
>> A=1:10;
>> all(A>5)
ans =
0
>> A=1:0;
>> all(A>5)
ans =
1
Use the built-in function isempty to check for empty arrays:
a = [];
isempty(a)
ans =
1

Chapter 5: Conditions
Parameter Description
expression an expression that has logical meaning
Section 5.1: IF condition

Conditions are a fundamental part of almost any part of code. They are used to execute some parts of the code
only in some situations, but not other. Let's look at the basic syntax:
a = 5;
if a > 10 % this condition is not fulfilled, so nothing will happen
disp('OK')
end
if a < 10 % this condition is fulfilled, so the statements between the if...end are executed
disp('Not OK')
end
Output:
Not OK
In this example we see the if consists of 2 parts: the condition, and the code to run if the condition is true. The
code is everything written after the condition and before the end of that if. The first condition was not fulfilled and
hence the code within it was not executed.
Here is another example:
a = 5;
if a ~= a+1 % "~=" means "not equal to"
disp('It''s true!') % we use two apostrophes to tell MATLAB that the ' is part of the string
end
The condition above will always be true, and will display the output It's true!.
We can also write:
a = 5;
if a == a+1 % "==" means "is equal to", it is NOT the assignment ("=") operator
disp('Equal')
end
This time the condition is always false, so we will never get the output Equal.
There is not much use for conditions that are always true or false, though, because if they are always false we can
simply delete this part of the code, and if they are always true then the condition is not needed.
Section 5.2: IF-ELSE condition

In some cases we want to run an alternative code if the condition is false, for this we use the optional else part:
a = 20;
if a < 10
disp('a smaller than 10')

else
disp('a bigger than 10')
end
Here we see that because a is not smaller than 10 the second part of the code, after the else is executed and we
get the output a bigger than 10. Now let's look at another try:
a = 10;
if a > 10
disp('a bigger than 10')
else
disp('a smaller than 10')
end
In this example shows that we did not checked if a is indeed smaller than 10, and we get a wrong message because
the condition only check the expression as it is, and ANY case that does not equals true (a = 10) will cause the
second part to be executed.
This type of error is a very common pitfall for both beginners and experienced programmers, especially when
conditions become complex, and should be always kept in mind
Section 5.3: IF-ELSEIF condition

Using else we can perform some task when the condition is not satisfied. But what if we want to check a second
condition in case that the first one was false. We can do it this way:
a = 9;
if mod(a,2)==0 % MOD - modulo operation, return the remainder after division of 'a' by 2
disp('a is even')
else
if mod(a,3)==0
disp('3 is a divisor of a')
end
end
OUTPUT:
3 is a divisor of a
This is also called "nested condition", but here we have a speciel case that can improve code readability, and reduce
the chance for anr error - we can write:
a = 9;
if mod(a,2)==0
disp('a is even')
elseif mod(a,3)==0
end
OUTPUT:
3 is a divisor of a
using the elseif we are able to check another expression within the same block of condition, and this is not limited
to one try:
a = 25;

if mod(a,2)==0
disp('a is even')
elseif mod(a,3)==0
elseif mod(a,5)==0
end
OUTPUT:
5 is a divisor of a
Extra care should be taken when choosing to use elseif in a row, since only one of them will be executed from all
the if to end block. So, in our example if we want to display all the divisors of a (from those we explicitly check) the
example above won't be good:
a = 15;
if mod(a,2)==0
disp('a is even')
elseif mod(a,3)==0
elseif mod(a,5)==0
end
OUTPUT:
3 is a divisor of a
not only 3, but also 5 is a divisor of 15, but the part that check the division by 5 is not reached if any of the
expressions above it was true.
Finally, we can add one else (and only one) after all the elseif conditions to execute a code when none of the
conditions above are met:
a = 11;
if mod(a,2)==0
disp('a is even')
elseif mod(a,3)==0
elseif mod(a,5)==0
else
disp('2, 3 and 5 are not divisors of a')
end
OUTPUT:
2, 3 and 5 are not divisors of a
Section 5.4: Nested conditions

When we use a condition within another condition we say the conditions are "nested". One special case of nested
conditions is given by the elseif option, but there are numerous other ways to use nested conditons. Let's
examine the following code:
a = 2;
if mod(a,2)==0 % MOD - modulo operation, return the remainder after division of 'a' by 2

disp('a is even')
if mod(a,3)==0
if mod(a,5)==0
end
end
else
disp('a is odd')
end
For a=2, the output will be a is even, which is correct. For a=3, the output will be a is odd, which is also correct,
but misses the check if 3 is a divisor of a. This is because the conditions are nested, so only if the first is true, than
we move to the inner one, and if a is odd, none of the inner conditions are even checked. This is somewhat
opposite to the use of elseif where only if the first condition is false than we check the next one. What about
checking the division by 5? only a number that has 6 as a divisor (both 2 and 3) will be checked for the division by 5,
and we can test and see that for a=30 the output is:
a is even
3 is a divisor of a
5 is a divisor of a
We should also notice two things:
1. The position of the end in the right place for each if is crucial for the set of conditions to work as expected,
so indentation is more than a good recommendation here.
2. The position of the else statement is also crucial, because we need to know in which if (and there could be
several of them) we want to do something in case the expression if false.
Let's look at another example:
for a = 5:10 % the FOR loop execute all the code within it for every a from 5 to 10
ch = num2str(a); % NUM2STR converts the integer a to a character
if mod(a,2)==0
if mod(a,3)==0
disp(['3 is a divisor of ' ch])
elseif mod(a,4)==0
else
disp([ch ' is even'])
end
elseif mod(a,3)==0
else
disp([ch ' is odd'])
end
end
And the output will be:
5 is odd
3 is a divisor of 6
7 is odd
4 is a divisor of 8
3 is a divisor of 9
10 is even

we see that we got only 6 lines for 6 numbers, because the conditions are nested in a way that ensure only one
print per number, and also (although can't be seen directly from the output) no extra checks are preformed, so if a
number is not even there is no point to check if 4 is one of it divisors.

Chapter 6: Set operations
Parameter Details
A,B sets, possibly matrices or vectors
x possible element of a set
Section 6.1: Elementary set operations

It's possible to perform elementary set operations with Matlab. Let's assume we have given two vectors or arrays
A = randi([0 10],1,5);
B = randi([-1 9], 1,5);
and we want to find all elements which are in A and in B. For this we can use
C = intersect(A,B);
C will include all numbers which are part of A and part of B. If we also want to find the position of these elements we
call
[C,pos] = intersect(A,B);
pos is the position of these elements such that C == A(pos).
Another basic operation is the union of two sets
D = union(A,B);
Herby contains D all elements of A and B.
Note that A and B are hereby treated as sets which means that it does not matter how often an element is part of A
or B. To clarify this one can check D == union(D,C).
If we want to obtain the data that is in 'A' but not in 'B' we can use the following function
E = setdiff(A,B);
We want to note again that this are sets such that following statement holds D == union(E,B).
Suppose we want to check if
x = randi([-10 10],1,1);
is an element of either A or B we can execute the command
a = ismember(A,x);
b = ismember(B,x);
If a==1 then x is element of A and x is no element is a==0. The same goes for B. If a==1 && b==1 x is also an element
of C. If a == 1 || b == 1 x is element of D and if a == 1 || b == 0 it's also element of E.

Chapter 7: For loops
Section 7.1: Iterate over columns of matrix
If the right-hand side of the assignment is a matrix, then in each iteration the variable is assigned subsequent
columns of this matrix.
some_matrix = [1, 2, 3; 4, 5, 6]; % 2 by 3 matrix

for some_column = some_matrix
display(some_column)
end
(The row vector version is a normal case of this, because in Matlab a row vector is just a matrix whose columns are
size 1.)
The output would display
1
4
2
5
3
6
i.e. each column of the iterated matrix displayed, each column printed on each call of display.
Section 7.2: Notice: Weird same counter nested loops

This is not something you will see in other programming environments. I came across it some years back and I
couldn't understand why it was happening, but after working with MATLAB for some time I was able to figure it out.
Look at the code snippet below:
for x = 1:10
for x = 1:10
fprintf('%d,', x);
end
fprintf('\n');
end
you wouldn't expect this to work properly but it does, producing the following output:
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
1,2,3,4,5,6,7,8,9,10,
The reason is that, as with everything else in MATLAB, the x counter is also a matrix—a vector to be precise. As
such, x is only a reference to an 'array' (a coherent, consecutive memory structure) which is appropriatelly

referenced with every consequent loop (nested or not). The fact that the nested loop uses the same identifier
makes no difference to how values from that array are referenced. The only problem is that within the nested loop
the outer x is hidden by the nested (local) x and therefore cannot be referenced. However, the functionality of the
nested loop structure remains intact.
Section 7.3: Iterate over elements of vector

The right-hand side of the assignment in a for loop can be any row vector. The left-hand side of the assignment can
be any valid variable name. The for loop assigns a different element of this vector to the variable each run.
other_row_vector = [4, 3, 5, 1, 2];

for any_name = other_row_vector
display(any_name)
end
The output would display
4
3
5
1
2
(The 1:n version is a normal case of this, because in Matlab 1:n is just syntax for constructing a row vector of [1,
2, ..., n].)
Hence, the two following blocks of code are identical:
A = [1 2 3 4 5];
for x = A
disp(x);
end
and
for x = 1:5
disp(x);
end
And the following are identical as well:
A = [1 3 5 7 9];
for x = A
disp(x);
end
and
for x = 1:2:9
disp(x);
end
Any row vector will do. They don't have to be numbers.
my_characters = 'abcde';

for my_char = my_characters
disp(my_char)
end
will output
a
b
c
d
e
Section 7.4: Nested Loops

Loops can be nested, to preform iterated task within another iterated task. Consider the following loops:
ch = 'abc';
m = 3;
for c = ch
for k = 1:m
disp([c num2str(k)]) % NUM2STR converts the number stored in k to a charachter,
% so it can be concataneted with the letter in c
end
end
we use 2 iterators to display all combinations of elements from abc and 1:m, which yields:
a1
a2
a3
b1
b2
b3
c1
c2
c3
We can also use nested loops to combine between tasks to be done each time, and tasks to be done once in a
several iterations:
N = 10;
n = 3;
a1 = 0; % the first element in Fibonacci series
a2 = 1; % the secound element in Fibonacci series
for j = 1:N
for k = 1:n
an = a1 + a2; % compute the next element in Fibonacci series
a1 = a2; % save the previous element for the next iteration
a2 = an; % save ht new element for the next iteration
end
disp(an) % display every n'th element
end
Here we want to compute all the Fibonacci series, but to display only the nth element each time, so we get

3
13
55
233
987
4181
17711
75025
317811
1346269
Another thing we can do is to use the first (outer) iterator within the inner loop. Here is another example:
N = 12;
gap = [1 2 3 4 6];
for j = gap
for k = 1:j:N
fprintf('%d ',k) % FPRINTF prints the number k proceeding to the next the line
end
fprintf('\n') % go to the next line
end
This time we use the nested loop to format the output, and brake the line only when a new gap (j) between the
elements was introduced. We loop through the gap width in the outer loop and use it within the inner loop to
iterate through the vector:
1 2 3 4 5 6 7 8 9 10 11 12
1 3 5 7 9 11
1 4 7 10
1 5 9
1 7
Section 7.5: Loop 1 to n

The simplest case is just preforming a task for a fixed known number of times. Say we want to display the numbers
between 1 to n, we can write:
n = 5;
for k = 1:n
display(k)
end
The loop will execute the inner statement(s), everything between the for and the end, for n times (5 in this
example):
1
2
3
4
5
Here is another example:
n = 5;
for k = 1:n

disp(n-k+1:-1:1) % DISP uses more "clean" way to print on the screen
end
this time we use both the n and k in the loop, to create a "nested" display:
5 4 3 2 1
4 3 2 1
3 2 1
2 1
Section 7.6: Loop over indexes

my_vector = [0, 2, 1, 3, 9];
for i = 1:numel(my_vector)
my_vector(i) = my_vector(i) + 1;
end
Most simple things done with for loops can be done faster and easier by vectorized operations. For example, the
above loop can be replaced by my_vector = my_vector + 1.

Chapter 8: Initializing Matrices or arrays
Parameter Details
sz n (for an n x n matrix)
sz n, m (for an n x m matrix)
sz m,n,...,k (for an m-by-n-by-...-by-k matrix)
datatype 'double' (default), 'single', 'int8', 'uint8', 'int16', 'uint16', 'int32', 'uint32', 'int64', or 'uint64'
arraytype 'distributed'
arraytype 'codistributed'
arraytype 'gpuArray'
Matlab has three important functions to create matrices and set their elements to zeroes, ones, or the identity
matrix. (The identity matrix has ones on the main diagonal and zeroes elsewhere.)
Section 8.1: Creating a matrix of 0s

z1 = zeros(5); % Create a 5-by-5 matrix of zeroes
z2 = zeros(2,3); % Create a 2-by-3 matrix
Section 8.2: Creating a matrix of 1s

o1 = ones(5); % Create a 5-by-5 matrix of ones
o2 = ones(1,3); % Create a 1-by-3 matrix / vector of size 3
Section 8.3: Creating an identity matrix

i1 = eye(3); % Create a 3-by-3 identity matrix
i2 = eye(5,6); % Create a 5-by-6 identity matrix

Chapter 9: Matrix decompositions
Section 9.1: Schur decomposition
If A is a complex and quadratic matrix there exists a unitary Q such that Q*AQ = T = D + N with D being the diagonal
matrix consisting of the eigenvalues and N being strictly upper tridiagonal.
A = [3 6 1
23 13 1
0 3 4];
T = schur(A);
We also display the runtime of schur dependent on the square root of matrix elements:
Section 9.2: Cholesky decomposition

The Cholesky decomposition is a method to decompose an hermitean, positiv definite matrix into an upper
triangular matrix and its transpose. It can be used to solve linear equations systems and and is around twice as fast
as LU-decomposition.
A = [4 12 -16
12 37 -43
-16 -43 98];
R = chol(A);
This returns the upper triangular matrix. The lower one is obtained by transposition.
L = R';
We finally can check whether the decomposition was correct.

b = (A == L*R);
Section 9.3: QR decomposition

This method will decompose a matrix into an upper triangular and an orthogonal matrix.
A = [4 12 -16
12 37 -43
-16 -43 98];
R = qr(A);
This will return the upper triangular matrix while the following will return both matrices.
[Q,R] = qr(A);
The following plot will display the runtime of qr dependent of the square root of elements of the matrix.
Section 9.4: LU decomposition

Hereby a matrix will be decomposed into an upper trangular and an lower triangular matrix. Often it will be used to
increase the performance and stability (if it's done with permutation) of Gauß elimination.
However, quite often does this method not or badly work as it is not stable. For example
A = [8 1 6
3 5 7
4 9 2];
[L,U] = lu(A);
It is sufficient to add an permutation matrix such that PA=LU:
[L,U,P]=lu(A);

In the following we will now plot the runtime of `lu' dependent of the square root of elements of the matrix.
Section 9.5: Singular value decomposition

Given an m times n matrix A with n larger than m. The singular value decomposition
[U,S,V] = svd(A);
computes the matrices U,S,V.
The matrix U consists of the left singular eigenvectors which are the eigenvectors of A*A.' while V consists of the
right singular eigenvalues which are the eigenvectors of A.'*A. The matrix S has the square roots of the eigenvalues
of A*A.' and A.'*A on its diagonal.
If m is larger than n one can use
[U,S,V] = svd(A,'econ');
to perform economy sized singular value decomposition.

Chapter 10: Object-Oriented Programming
Section 10.1: Value vs Handle classes
Classes in MATLAB are divided into two major categories: value classes and handle classes. The major difference is
that when copying an instance of a value class, the underlying data is copied to the new instance, while for handle
classes the new instance points to the original data and changing values in new instance changes them in the
original. A class can be defined as a handle by inheriting from the handle class.
classdef valueClass
properties
data
end
end
and
classdef handleClass < handle

properties
data
end
end
then
>> v1 = valueClass;
>> v1.data = 5;
>> v2 = v1;
>> v2.data = 7;
>> v1.data
ans =
5
>> h1 = handleClass;
>> h1.data = 5;
>> h2 = h1;
>> h2.data = 7;
>> h1.data
ans =
7
Section 10.2: Constructors

A constructor is a special method in a class that is called when an instance of an object is created. It is a regular
MATLAB function that accepts input parameters but it also must follow certain rules.
Constructors are not required as MATLAB creates a default one. In practice, however, this is a place to define a state
of an object. For example, properties can be restricted by specifying attributes. Then, a constructor can initalize
such properties by default or user defined values which in fact can sent by input parameters of a constructor.
Calling a constructor of a simple class
This is a simple class Person.
classdef Person
properties

name
surname
address
end
methods
function obj = Person(name,surname,address)
obj.name = name;
obj.surname = surname;
obj.address = address;
end
end
end
The name of a constructor is the same the name of a class. Consequently, constructors are called by the name of its
class. A class Person can be created as follows:
>> p = Person('John','Smith','London')
p =
Person with properties:
name: 'John'
surname: 'Smith'
address: 'London'
Calling a constructor of a child class
Classes can be inherited from parent classes if the share common properties or methods. When a class is inherited
from another, it is likely that a constructor of a parent class has to be called.
A class Member inherits from a class Person because Member uses the same properties as the class Person but it also
adds payment to its definition.
classdef Member < Person

properties
payment
end
methods
function obj = Member(name,surname,address,payment)
obj = obj@Person(name,surname,address);
obj.payment = payment;
end
end
end
Similarly to the class Person, Member is created by calling its constructor:
>> m = Member('Adam','Woodcock','Manchester',20)
m =
Member with properties:
payment: 20
name: 'Adam'
surname: 'Woodcock'
address: 'Manchester'
A constructor of Person requires three input parameters. Member must respect this fact and therefore call a
constructor of the class Person with three parameters. It is fulfilled by the line:

obj = obj@Person(name,surname,address);
The example above shows the case when a child class needs information for its parent class. This is why a
constructor of Member requires four parameters: three for its parent class and one for itself.
Section 10.3: Defining a class

A class can be defined using classdef in an .m file with the same name as the class. The file can contain the
classdef...end block and local functions for use within class methods.
The most general MATLAB class definition has the following structure:
classdef (ClassAttribute = expression, ...) ClassName < ParentClass1 & ParentClass2 & ...
properties (PropertyAttributes)
PropertyName
end
methods (MethodAttributes)
function obj = methodName(obj,arg2,...)
...
end
end
events (EventAttributes)
EventName
end
enumeration
EnumName
end
end
MATLAB Documentation: Class attributes, Property attributes, Method attributes, Event attributes, Enumeration
class restrictions.
Example class:
A class called Car can be defined in file Car.m as
classdef Car < handle % handle class so properties persist

properties
make
model
mileage = 0;
end
methods
function obj = Car(make, model)
obj.make = make;
obj.model = model;
end
function drive(obj, milesDriven)
obj.mileage = obj.mileage + milesDriven;
end
end
end

Note that the constructor is a method with the same name as the class. <A constructor is a special method of a
class or structure in object-oriented programming that initializes an object of that type. A constructor is an instance
method that usually has the same name as the class, and can be used to set the values of the members of an
object, either to default or to user-defined values.>
An instance of this class can be created by calling the constructor;
>> myCar = Car('Ford', 'Mustang'); //creating an instance of car class
Calling the drive method will increment the mileage
>> myCar.mileage
ans =
0
>> myCar.drive(450);
>> myCar.mileage
ans =
450
Section 10.4: Inheriting from classes and abstract classes

Disclaimer: the examples presented here are only for the purpose of showing the use of abstract classes and
inheritance and may not necessarily be of a practical use. Also, there is no sich thing as polymorphic in MATLAB and
therefore the use of abstract classes is limited. This example is to show who to create a class, inherit from another
class and apply an abstract class to define a common interface.
The use of abstract classes is rather limited in MATLAB but it still can come useful on a couple of occasions.
Let's say we want a message logger. We might create a class similar to the one below:
classdef ScreenLogger
properties(Access=protected)
scrh;
end
methods
function obj = ScreenLogger(screenhandler)
obj.scrh = screenhandler;
end
function LogMessage(obj, varargin)

if ~isempty(varargin)
varargin{1} = num2str(varargin{1});
fprintf(obj.scrh, '%s\n', sprintf(varargin{:}));
end
end
end
end
Properties and methods
In short, properties hold a state of an object whilst methods are like interface and define actions on objects.

The property scrh is protected. This is why it must be initialized in a constructor. There are other methods (getters)
to access this property but it is out of cope of this example. Properties and methods can be access via a variable
that holds a reference to an object by using dot notation followed by a name of a method or a property:
mylogger = ScreenLogger(1); % OK
mylogger.LogMessage('My %s %d message', 'very', 1); % OK
mylogger.scrh = 2; % ERROR!!! Access denied
Properties and methods can be public, private, or protected. In this case, protected means that I will be able to
access to scrh from an inherited class but not from outside. By default all properties and methods are public.
Therefore LogMessage() can freely be used outside the class definition. Also LogMessage defines an interface
meaning this is what we must call when we want an object to log our custom messages.
Application
Let's say I have a script where I utilize my logger:
clc;
% ... a code
logger = ScreenLogger(1);
% ... a code
logger.LogMessage('something');
% ... a code
% ... a code
% ... a code
If I have multiple places where I use the same logger and then want to change it to something more sophisticated,
such as write a message in a file, I would have to create another object:
classdef DeepLogger
properties(SetAccess=protected)
FileName
end
methods
function obj = DeepLogger(filename)
obj.FileName = filename;
end

fid = fopen(obj.fullfname, 'a+t');
fprintf(fid, '%s\n', sprintf(varargin{:}));
fclose(fid);
end
end
end
end
and just change one line of a code into this:
clc;
% ... a code
logger = DeepLogger('mymessages.log');

The above method will simply open a file, append a message at the end of the file and close it. At the moment, to
be consistent with my interface, I need to remember that the name of a method is LogMessage() but it could
equally be anything else. MATLAB can force developper to stick to the same name by using abstract classes. Let's
say we define a common interface for any logger:
classdef MessageLogger
methods(Abstract=true)
LogMessage(obj, varargin);
end
end
Now, if both ScreenLogger and DeepLogger inherit from this class, MATLAB will generate an error if LogMessage() is
not defined. Abstract classes help to build similar classes which can use the same interface.
For the sake of this exmaple, I will make slightly different change. I am going to assume that DeepLogger will do
both logging message on a screen and in a file at the same time. Because ScreenLogger already log messages on
screen, I am going to inherit DeepLogger from the ScreenLoggger to avoid repetition. ScreenLogger doesn't change
at all apart from the first line:
classdef ScreenLogger < MessageLogger

// the rest of previous code
However, DeepLogger needs more changes in the LogMessage method:
classdef DeepLogger < MessageLogger & ScreenLogger

FileName
Path
end
methods
function obj = DeepLogger(screenhandler, filename)
[path,filen,ext] = fileparts(filename);
obj.FileName = [filen ext];
pbj.Path = pathn;
obj = obj@ScreenLogger(screenhandler);
end
LogMessage@ScreenLogger(obj, varargin{:});
fclose(fid);
end
end
end
end
Firstly, I simply initialize properties in the constructor. Secondly, because this class inherits from ScreenLogger I
have to initialize this parrent object as well. This line is even more important because ScreenLogger constructor
requires one parameter to initalize its own object. This line:
obj = obj@ScreenLogger(screenhandler);
simply says "call the consructor of ScreenLogger and initalize it with a screen handler". It is worth noting here that I
have defined scrh as protected. Therefore, I could equally access this property from DeepLogger. If the property
was defined as private. The only way to intialize it would be using consuctor.

Another change is in section methods. Again to avoid repetition, I call LogMessage() from a parent class to log a
message on a screen. If I had to change anything to make improvements in screen logging, now I have to do it in
one place. The rest code is the same as it is a part of DeepLogger.
Because this class also inherits from an abstract class MessageLogger I had to make sure that LogMessage() inside
DeepLogger is also defined. Inheritting from MessageLogger is a little bit tricky here. I think it cases redefinition of
LogMessage mandatory--my guess.
In terms of the code where the a logger is applied, thanks to a common interface in classes, I can rest assure ther
changin this one line in the whole code would not make any issues. The same messages will be log on screen as
before but additionally the code will write such messages to a file.
clc;
% ... a code
logger = DeepLogger(1, 'mylogfile.log');
% ... a code
% ... a code
% ... a code
% ... a code
I hope these examples explained the use of classes, the use of inheritance, and the use of abstract classes.
PS. The solution for the above problem is one of many. Another solution, less complex, would be to make
ScreenLoger to be a component of another logger like FileLogger etc. ScreenLogger would be held in one of the
properties. Its LogMessage would simply call LogMessage of the ScreenLogger and show text on a screen. I have
chosen more complex approach to rather show how classes work in MATLAB. The example code below:
classdef DeepLogger < MessageLogger

FileName
Path
ScrLogger
end
methods
function obj = DeepLogger(screenhandler, filename)
[path,filen,ext] = fileparts(filename);
obj.FileName = [filen ext];
obj.Path = pathn;
obj.ScrLogger = ScreenLogger(screenhandler);
end
obj.LogMessage(obj.ScrLogger, varargin{:}); % <-------- thechange here
fclose(fid);
end
end
end
end

Chapter 11: Vectorization
Section 11.1: Use of bsxfun
Quite often, the reason why code has been written in a for loop is to compute values from 'nearby' ones. The
function bsxfun can often be used to do this in a more succinct fashion.
For example, assume that you wish to perform a columnwise operation on the matrix B, subtracting the mean of
each column from it:
B = round(randn(5)*10); % Generate random data

A = zeros(size(B)); % Preallocate array
for col = 1:size(B,2); % Loop over columns
A(:,col) = B(:,col) - mean(B(:,col)); % Subtract means
end
This method is inefficient if B is large, often due to MATLAB having to move the contents of variables around in
memory. By using bsxfun, one can do the same job neatly and easily in just a single line:
A = bsxfun(@minus, B, mean(B));
Here, @minus is a function handle to the minus operator (-) and will be applied between elements of the two
matrices B and mean(B). Other function handles, even user-defined ones, are possible as well.
Next, suppose you want to add row vector v to each row in matrix A:
v = [1, 2, 3];
A = [8, 1, 6
3, 5, 7
4, 9, 2];
The naive approach is use a loop (do not do this):
B = zeros(3);
for row = 1:3
B(row,:) = A(row,:) + v;
end
Another option would be to replicate v with repmat (do not do this either):
>> v = repmat(v,3,1)
v =
1 2 3
1 2 3
1 2 3
>> B = A + v;
Instead use bsxfun for this task:
>> B = bsxfun(@plus, A, v);

B =
9 3 9
4 7 10

5 11 5
Syntax
bsxfun(@fun, A, B)
where @fun is one of the supported functions and the two arrays A and B respect the two conditions below.
The name bsxfun helps to understand how the function works and it stands for Binary FUNction with Singleton
eXpansion. In other words, if:
1. two arrays share the same dimensions except for one

2. and the discordant dimension is a singleton (i.e. has a size of 1) in either of the two arrays
then the array with the singleton dimension will be expanded to match the dimension of the other array. After the
expansion, a binary function is applied elementwise on the two arrays.
For example, let A be an M-by-N-byK array and B is an M-by-N array. Firstly, their first two dimensions have
corresponding sizes. Secondly, A has K layers while B has implicitly only 1, hence it is a singleton. All conditions are
met and B will be replicated to match the 3rd dimension of A.
In other languages, this is commonly referred to as broadcasting and happens automatically in Python (numpy) and
Octave.
The function, @fun, must be a binary function meaning it must take exactly two inputs.
Remarks
Internally, bsxfun does not replicate the array and executes an efficient loop.
Section 11.2: Implicit array expansion (broadcasting) [R2016b]

MATLAB R2016b featured a generalization of its scalar expansion1,2 mechanism, to also support certain element-
wise operations between arrays of different sizes, as long as their dimension are compatible. The operators that
support implicit expansion are1:
Element-wise arithmetic operators: +, -, .*, .^, ./, .\.

Relational operators: <, <=, >, >=, ==, ~=.
Logical operators: &, |, xor.
Bit-wise functions: bitand, bitor, bitxor.
Elementary math functions: max, min, mod, rem, hypot, atan2, atan2d.
The aforementioned binary operations are allowed between arrays, as long as they have "compatible sizes". Sizes
are considered "compatible" when each dimension in one array is either exactly equal to the same dimension in the
other array, or is equal to 1. Note that trailing singleton (that is, of size 1) dimensions are omitted by MATLAB, even
though there's theoretically an infinite amount of them. In other words - dimensions that appear in one array and
do not appear in the other, are implicitly fit for automatic expansion.
For example, in MATLAB versions before R2016b this would happen:
>> magic(3) + (1:3)

Error using +
Matrix dimensions must agree.
Whereas starting from R2016b the previous operation will succeed:

>> magic(3) + (1:3)
ans =
9 3 9
4 7 10
5 11 5
Examples of compatible sizes:

Description 1st Array Size 2nd Array Size Result Size
Vector and scalar [3x1] [1x1] [3x1]
Row and column vectors [1x3] [2x1] [2x3]
Vector and 2D matrix [1x3] [5x3] [5x3]
N-D and K-D arrays [1x3x3] [5x3x1x4x2] [5x3x3x4x2]
Examples of incompatible sizes:

2nd Array
Description 1st Array Size Possible Workaround
Size
Vectors where a dimension is a multiple of the same
[1x2] [1x8] transpose
dimension in the other array.
Arrays with dimensions that are multiples of each other. [2x2] [8x8] repmat, reshape
N-D arrays that have the right amount of singleton
[2x3x4] [2x4x3] permute
dimensions but they're in the wrong order (#1).
N-D arrays that have the right amount of singleton
[2x3x4x5] [5x2] permute
dimensions but they're in the wrong order (#2).
IMPORTANT:
Code relying on this convention is NOT backward-compatible with any older versions of MATLAB. Therefore, the
explicit invocation of bsxfun1,2 (which achieves the same effect) should be used if code needs to run on older
MATLAB versions. If such a concern does not exist, MATLAB R2016 release notes encourage users to switch from
bsxfun:
Compared to using bsxfun, implicit expansion offers faster speed of execution, better memory usage, and
improved readability of code.
Related reading:
MATLAB documentation on "Compatible Array Sizes for Basic Operations".

NumPy's Broadcasting1,2.
A comparison between the speed of computing using bsxfun vs. implicit array expansion.
Section 11.3: Element-wise operations

MATLAB supports (and encourages) vectorized operations on vectors and matrices.
For example, suppose we have A and B, two n-by-m matrices and we want C to be the element-wise product of the
corresponding elements (i.e., C(i,j) = A(i,j)*B(i,j)).
The un-vectorized way, using nested loops is as follows:
C = zeros(n,m);
for ii=1:n
for jj=1:m
C(ii,jj) = A(ii,jj)*B(ii,jj);
end

end
However, the vectorized way of doing this is by using the element-wise operator .*:
C = A.*B;
For more information on the element-wise multiplication in MATLAB see the documentation of times.
For more information about the difference between array and matrix operations see Array vs. Matrix
Operations in the MATLAB documentation.
Section 11.4: Logical Masking

MATLAB supports the use of logical masking in order to perform selection on a matrix without the use of for loops
or if statements.
A logical mask is defined as a matrix composed of only 1 and 0.
For example:
mask = [1 0 0; 0 1 0; 0 0 1];
is a logical matrix representing the identity matrix.
We can generate a logical mask using a predicate to query a matrix.
A = [1 2 3; 4 5 6; 7 8 9];
B = A > 4;
We first create a 3x3 matrix, A, containing the numbers 1 through 9. We then query A for values that are greater
than 4 and store the result in a new matrix called B.
B is a logical matrix of the form:
B = [0 0 0
0 1 1
1 1 1]
Or 1 when the predicate A > 4 was true. And 0 when it was false.
We can use logical matrices to access elements of a matrix. If a logical matrix is used to select elements, indices
where a 1 appear in the logical matrix will be selected in the matrix you are selecting from.
Using the same B from above, we could do the following:
C = [0 0 0; 0 0 0; 0 0 0];
C(B) = 5;
This would select all of the elements of C where B has a 1 in that index. Those indices in C are then set to 5.
Our C now looks like:
C = [0 0 0
0 5 5
5 5 5]

We can reduce complicated code blocks containing if and for by using logical masks.
Take the non-vectorized code:
A = [1 3 5; 7 9 11; 11 9 7];
for j = 1:length(A)
if A(j) > 5
A(j) = A(j) - 2;
end
end
This can be shortened using logical masking to the following code:
A = [1 3 5; 7 9 11; 11 9 7];
B = A > 5;
A(B) = A(B) - 2;
Or even shorter:
A = [1 3 5; 7 9 11; 11 9 7];
A(A > 5) = A(A > 5) - 2;
Section 11.5: Sum, mean, prod & co

Given a random vector
v = rand(10,1);
if you want the sum of its elements, do NOT use a loop
s = 0;
for ii = 1:10
s = s + v(ii);
end
but use the vectorized capability of the sum() function
s = sum(v);
Functions like sum(), mean(), prod() and others, have the ability to operate directly along rows, columns or other
dimensions.
For instance, given a random matrix
A = rand(10,10);
the average for each column is
m = mean(A,1);
the average for each row is
m = mean(A,2)
All the functions above work only on one dimension, but what if you want to sum the whole matrix? You could use:

s = sum(sum(A))
But what if have an ND-array? applying sum on sum on sum... don't seem like the best option, instead use the :
operator to vectorize your array:
s = sum(A(:))
and this will result in one number which is the sum of all your array, doesn't matter how many dimensions it have.
Section 11.6: Get the value of a function of two or more

arguments
In many application it is necessary to compute the function of two or more arguments.
Traditionally, we use for-loops. For example, if we need to calculate the f = exp(-x^2-y^2) (do not use this if you
need fast simulations):
% code1
x = -1.2:0.2:1.4;
y = -2:0.25:3;
for nx=1:lenght(x)
for ny=1:lenght(y)
f(nx,ny) = exp(-x(nx)^2-y(ny)^2);
end
end
But vectorized version is more elegant and faster:
% code2
[x,y] = ndgrid(-1.2:0.2:1.4, -2:0.25:3);
f = exp(-x.^2-y.^2);
than we can visualize it:
surf(x,y,f)
Note1 - Grids: Usually, the matrix storage is organized row-by-row. But in the MATLAB, it is the column-by-column
storage as in FORTRAN. Thus, there are two simular functions ndgrid and meshgrid in MATLAB to implement the
two aforementioned models. To visualise the function in the case of meshgrid, we can use:
surf(y,x,f)
Note2 - Memory consumption: Let size of x or y is 1000. Thus, we need to store 1000*1000+2*1000 ~ 1e6 elements
for non-vectorized code1. But we need 3*(1000*1000) = 3e6 elements in the case of vectorized code2. In the 3D
case (let z has the same size asx or y), memory consumption increases dramatically: 4*(1000*1000*1000) (~32GB
for doubles) in the case of the vectorized code2 vs ~1000*1000*1000 (just ~8GB) in the case of code1. Thus, we have
to choose either the memory or speed.

Chapter 12: Graphics: 2D Line Plots
Parameter Details
X x-values
Y y-values
LineSpec Line style, marker symbol, and color, specified as a string
Name,Value Optional pairs of name-value arguments to customize line properties
h handle to line graphics object
Section 12.1: Split line with NaNs

Interleave your y or x values with NaNs
x = [1:5; 6:10]';
x(3,2) = NaN
x =
1 6
2 7
3 NaN
4 9
5 10
plot(x)
Section 12.2: Multiple lines in a single plot

In this example we are going to plot multiple lines onto a single axis. Additionally, we choose a different appearance
for the lines and create a legend.
% create sample data

x = linspace(-2,2,100); % 100 linearly spaced points from -2 to 2
y1 = x.^2;
y2 = 2*x.^2;
y3 = 4*x.^2;
% create plot
figure; % open new figure
plot(x,y1, x,y2,'--', x,y3,'-.'); % plot lines
grid minor; % add minor grid
title('Quadratic functions with different curvatures');
xlabel('x');
ylabel('f(x)');

legend('f(x) = x^2', 'f(x) = 2x^2', 'f(x) = 4x^2', 'Location','North');
In the above example, we plotted the lines with a single plot-command. An alternative is to use separate
commands for each line. We need to hold the contents of a figure with hold on the latest before we add the second
line. Otherwise the previously plotted lines will disappear from the figure. To create the same plot as above, we can
use these following commands:
figure; hold on;

plot(x,y1);
plot(x,y2,'--');
plot(x,y3,'-.');
The resulting figure looks like this in both cases:
Section 12.3: Custom colour and line style orders

In MATLAB, we can set new default custom orders, such as a colour order and a line style order. That means new
orders will be applied to any figure that is created after these settings have been applied. The new settings remains
until MATLAB session is closed or new settings has been made.
Default colour and line style order
By default, MATLAB uses a couple of different colours and only a solid line style. Therefore, if plot is called to draw
multiple lines, MATLAB alternates through a colour order to draw lines in different colours.

We can obtain the default colour order by calling get with a global handle 0 followed by this attribute
DefaultAxesColorOrder:
>> get(0, 'DefaultAxesColorOrder')

ans =
0 0.4470 0.7410
0.8500 0.3250 0.0980
0.9290 0.6940 0.1250
0.4940 0.1840 0.5560
0.4660 0.6740 0.1880
0.3010 0.7450 0.9330
0.6350 0.0780 0.1840
Custom colour and line style order
Once we have decided to set a custom colour order AND line style order, MATLAB must alternate through both. The
first change MATLAB applies is a colour. When all colours are exhausted, MATLAB applies the next line style from a
defined line style order and set a colour index to 1. That means MATLAB will begin to alternate through all colours
again but using the next line style in its order. When all line styles and colours are exhausted, obviously MATLAB
begins to cycle from the beginning using the first colour and the first line style.
For this example, I have defined an input vector and an anonymous function to make plotting figures a little bit
easier:
F = @(a,x) bsxfun(@plus, -0.2*x(:).^2, a);

x = (-5:5/100:5-5/100)';
To set a new colour or a new line style orders, we call set function with a global handle 0 followed by an attribute
DefaultAxesXXXXXXX; XXXXXXX can either be ColorOrder or LineStyleOrder. The following command sets a new
colour order to black, red and blue, respectively:
set(0, 'DefaultAxesColorOrder', [0 0 0; 1 0 0; 0 0 1]);

plot(x, F([1 2 3 4 5 6],x));
As you can see, MATLAB alternates only through colours because line style order is set to a solid line by default.
When a set of colours is exhausted, MATLAB starts from the first colour in the colour order.
The following commands set both colour and line style orders:
set(0, 'DefaultAxesColorOrder', [0 0 0; 1 0 0; 0 0 1]);

set(0, 'DefaultAxesLineStyleOrder', {'-' '--'});
plot(x, F([1 2 3 4 5 6],x));

Now, MATLAB alternates through different colours and different line styles using colour as most frequent attribute.

Chapter 13: Graphics: 2D and 3D
Transformations
Section 13.1: 2D Transformations
In this Example we are going to take a square shaped line plotted using line and perform transformations on it.
Then we are going to use the same transformations but in different order and see how it influences the results.
First we open a figure and set some initial parameters (square point coordinates and transformation parameters)
%Open figure and create axis

Figureh=figure('NumberTitle','off','Name','Transformation Example',...
'Position',[200 200 700 700]); %bg is set to red so we know that we can only see the axes
Axesh=axes('XLim',[-8 8],'YLim',[-8,8]);
%Initializing Variables
square=[-0.5 -0.5;-0.5 0.5;0.5 0.5;0.5 -0.5]; %represented by its vertices
Sx=0.5;
Sy=2;
Tx=2;
Ty=2;
teta=pi/4;
Next we construct the transformation matrices (scale, rotation and translation):
%Generate Transformation Matrix

S=makehgtform('scale',[Sx Sy 1]);
R=makehgtform('zrotate',teta);
T=makehgtform('translate',[Tx Ty 0]);
Next we plot the blue suare:
%% Plotting the original Blue Square

OriginalSQ=line([square(:,1);square(1,1)],[square(:,2);square(1,2)],'Color','b','LineWidth',3);
grid on; % Applying grid on the figure
hold all; % Holding all Following graphs to current axes
Next we will plot it again in a different color (red) and apply the transformations:
%% Plotting the Red Square

%Calculate rectangle vertices
HrectTRS=T*R*S;
RedSQ=line([square(:,1);square(1,1)],[square(:,2);square(1,2)],'Color','r','LineWidth',3);
%transformation of the axes
AxesTransformation=hgtransform('Parent',gca,'matrix',HrectTRS);
%seting the line to be a child of transformed axes
set(RedSQ,'Parent',AxesTransformation);
The result should look like this:

Now lets see what happens when we change the transformation order:
%% Plotting the Green Square

HrectRST=R*S*T;
GreenSQ=line([square(:,1);square(1,1)],[square(:,2);square(1,2)],'Color','g','LineWidth',3);
AxesTransformation=hgtransform('Parent',gca,'matrix',HrectRST);
set(GreenSQ,'Parent',AxesTransformation);
%% Plotting the Yellow Square

HrectSRT=S*R*T;
YellowSQ=line([square(:,1);square(1,1)],[square(:,2);square(1,2)],'Color','y','LineWidth',3);
AxesTransformation=hgtransform('Parent',gca,'matrix',HrectSRT);
set(YellowSQ,'Parent',AxesTransformation);

Chapter 14: Controlling Subplot coloring in
Matlab
As I was struggling with this more than once, and the web isn't really clear on what to do, I decided to take what's
out there, adding some of my own in order to explain how to create subplots which have one colorbar and they are
scaled according to it.
I have tested this using latest Matlab but I'm pretty sure it'll work in older versions.
Section 14.1: How it's done

This is a simple code creating 6 3d-subplots and in the end syncing the color displayed in each of them.
c_fin = [0,0];
[X,Y] = meshgrid(1:0.1:10,1:0.1:10);
figure; hold on;

for i = 1 : 6
Z(:,:,i) = i * (sin(X) + cos(Y));
ax(i) = subplot(3,2,i); hold on; grid on;

surf(X, Y, Z(:,:,i));
view(-26,30);
colormap('jet');
ca = caxis;
c_fin = [min(c_fin(1),ca(1)), max(c_fin(2),ca(2))];
end
%%you can stop here to see how it looks before we color-manipulate
c = colorbar('eastoutside');
c.Label.String = 'Units';
set(c, 'Position', [0.9, 0.11, 0.03, 0.815]); %%you may want to play with these values
pause(2); %%need this to allow the last image to resize itself before changing its axes
for i = 1 : 6
pos=get(ax(i), 'Position');
axes(ax(i));
set(ax(i), 'Position', [pos(1) pos(2) 0.85*pos(3) pos(4)]);
set(ax(i),'Clim', c_fin); %%this is where the magic happens
end

Chapter 15: Image processing
Section 15.1: Basic image I/O
>> img = imread('football.jpg');
Use imread to read image files into a matrix in MATLAB.

Once you imread an image, it is stored as an ND-array in memory:
>> size(img)
ans =
256 320 3
The image 'football.jpg' has 256 rows and 320 columns and it has 3 color channels: Red, Green and Blue.
You can now mirror it:
>> mirrored = img(:, end:-1:1, :); %// like mirroring any ND-array in Matlab
And finally, write it back as an image using imwrite:
>> imwrite(mirrored, 'mirrored_football.jpg');
Section 15.2: Retrieve Images from the Internet

As long as you have an internet connection, you can read images from an hyperlink
I=imread('https://cdn.sstatic.net/Sites/stackoverflow/company/img/logos/so/so-logo.png');
Section 15.3: Filtering Using a 2D FFT

Like for 1D signals, it's possible to filter images by applying a Fourier transformation, multiplying with a filter in the
frequency domain, and transforming back into the space domain. Here is how you can apply high- or low-pass
filters to an image with Matlab:
Let image be the original, unfiltered image, here's how to compute its 2D FFT:
ft = fftshift(fft2(image));
Now to exclude a part of the spectrum, one need to set its pixel values to 0. The spatial frequency contained in the
original image is mapped from the center to the edges (after using fftshift). To exclude the low frequencies, we
will set the central circular area to 0.
Here's how to generate a disc-shaped binary mask with radius D using built-in function:
[x y ~] = size(ft);
D = 20;
mask = fspecial('disk', D) == 0;
mask = imresize(padarray(mask, [floor((x/2)-D) floor((y/2)-D)], 1, 'both'), [x y]);
Masking the frequency domain image can be done by multiplying the FFT point-wise with the binary mask obtained
above:

masked_ft = ft .* mask;
Now, let's compute the inverse FFT:
filtered_image = ifft2(ifftshift(masked_ft), 'symmetric');
The high frequencies in an image are the sharp edges, so this high-pass filter can be used to sharpen images.
Section 15.4: Image Filtering

Let's say you have an image rgbImg, e.g., read in using imread.
>> rgbImg = imread('pears.png');

>> figure, imshow(rgbImg), title('Original Image')
Use fspecial to create a 2D filter:
>> h = fspecial('disk', 7);

>> figure, imshow(h, []), title('Filter')

Use imfilter to apply the filter on the image:
>> filteredRgbImg = imfilter(rgbImg, h);

>> figure, imshow(filteredRgbImg), title('Filtered Image')
Section 15.5: Measuring Properties of Connected Regions

Starting with a binary image, bwImg, which contains a number of connected objects.
>> bwImg = imread('blobs.png');

>> figure, imshow(bwImg), title('Binary Image')

To measure properties (e.g., area, centroid, etc) of every object in the image, use regionprops:
>> stats = regionprops(bwImg, 'Area', 'Centroid');
stats is a struct array which contains a struct for every object in the image. Accessing a measured property of an
object is simple. For example, to display the area of the first object, simply,
>> stats(1).Area
ans =
35
Visualize the object centroids by overlaying them on the original image.
>> figure, imshow(bwImg), title('Binary Image With Centroids Overlaid')

>> hold on
>> for i = 1:size(stats)
scatter(stats(i).Centroid(1), stats(i).Centroid(2), 'filled');
end

Chapter 16: Drawing
Section 16.1: Circles
The easiest option to draw a circle, is - obviously - the rectangle function.
%// radius
r = 2;
%// center
c = [3 3];
pos = [c-r 2*r 2*r];

rectangle('Position',pos,'Curvature',[1 1])
axis equal
but the curvature of the rectangle has to be set to 1!
The position vector defines the rectangle, the first two values x and y are the lower left corner of the rectangle.
The last two values define width and height of the rectangle.
pos = [ [x y] width height ]
The lower left corner of the circle - yes, this circle has corners, imaginary ones though - is the center c = [3 3]
minus the radius r = 2 which is [x y] = [1 1]. Width and height are equal to the diameter of the circle, so
width = 2*r; height = width;
In case the smoothness of the above solution is not sufficient, there is no way around using the obvious way of
drawing an actual circle by use of trigonometric functions.
%// number of points

n = 1000;

%// running variable
t = linspace(0,2*pi,n);
x = c(1) + r*sin(t);
y = c(2) + r*cos(t);
%// draw line

line(x,y)
%// or draw polygon if you want to fill it with color

%// fill(x,y,[1,1,1])
axis equal
Section 16.2: Arrows

Firstly, one can use quiver, where one doesn't have to deal with unhandy normalized figure units by use of
annotation
drawArrow = @(x,y) quiver( x(1),y(1),x(2)-x(1),y(2)-y(1),0 )
x1 = [10 30];
y1 = [10 30];
drawArrow(x1,y1); hold on
x2 = [25 15];
y2 = [15 25];
drawArrow(x2,y2)

Important is the 5th argument of quiver: 0 which disables an otherwise default scaling, as this function is usually
used to plot vector fields. (or use the property value pair 'AutoScale','off')
One can also add additional features:
drawArrow = @(x,y,varargin) quiver( x(1),y(1),x(2)-x(1),y(2)-y(1),0, varargin{:} )

drawArrow(x1,y1); hold on
drawArrow(x2,y2,'linewidth',3,'color','r')
If different arrowheads are desired, one needs to use annotations (this answer is may helpful How do I change the
arrow head style in quiver plot?).

The arrow head size can be adjust with the 'MaxHeadSize' property. It's not consistent unfortunately. The axes
limits need to be set afterwards.
x1 = [10 30];
y1 = [10 30];
drawArrow(x1,y1,{'MaxHeadSize',0.8,'Color','b','LineWidth',3}); hold on
x2 = [25 15];
y2 = [15 25];
drawArrow(x2,y2,{'MaxHeadSize',10,'Color','r','LineWidth',3}); hold on
xlim([1, 100])
ylim([1, 100])
There is another tweak for adjustable arrow heads:
function [ h ] = drawArrow( x,y,xlimits,ylimits,props )
xlim(xlimits)
ylim(ylimits)
h = annotation('arrow');
set(h,'parent', gca, ...
'position', [x(1),y(1),x(2)-x(1),y(2)-y(1)], ...
'HeadLength', 10, 'HeadWidth', 10, 'HeadStyle', 'cback1', ...
props{:} );
end
which you can call from your script as follows:
drawArrow(x1,y1,[1, 100],[1, 100],{'Color','b','LineWidth',3}); hold on

drawArrow(x2,y2,[1, 100],[1, 100],{'Color','r','LineWidth',3}); hold on

Section 16.3: Ellipse
To plot an ellipse you can use its equation. An ellipse has a major and a minor axis. Also we want to be able to plot
the ellipse on different center points. Therefore we write a function whose inputs and outputs are:
Inputs:
r1,r2: major and minor axis respectively
C: center of the ellipse (cx,cy)
Output:
[x,y]: points on the circumference of the ellipse
You can use the following function to get the points on an ellipse and then plot those points.
function [x,y] = getEllipse(r1,r2,C)

beta = linspace(0,2*pi,100);
x = r1*cos(beta) - r2*sin(beta);
y = r1*cos(beta) + r2*sin(beta);
x = x + C(1,1);
y = y + C(1,2);
end
Exmaple:
[x,y] = getEllipse(1,0.3,[2 3]);

plot(x,y);

Section 16.4: Pseudo 4D plot
A (m x n) matrix can be representes by a surface by using surf;
The color of the surface is automatically set as function of the values in the (m x n) matrix. If the colormap is not
specified, the default one is applied.
A colorbar can be added to display the current colormap and indicate the mapping of data values into the
colormap.
In the following example, the z (m x n) matrix is generated by the function:
z=x.*y.*sin(x).*cos(y);
over the interval [-pi,pi]. The x and y values can be generated using the meshgrid function and the surface is
rendered as follows:
% Create a Figure
figure
% Generate the `x` and `y` values in the interval `[-pi,pi]`
[x,y] = meshgrid([-pi:.2:pi],[-pi:.2:pi]);
% Evaluate the function over the selected interval
z=x.*y.*sin(x).*cos(y);
% Use surf to plot the surface
S=surf(x,y,z);
xlabel('X Axis');

ylabel('Y Axis');
zlabel('Z Axis');
grid minor
colormap('hot')
colorbar
Figure 1
Now it could be the case that additional information are linked to the values of the z matrix and they are store in
another (m x n) matrix
It is possible to add these additional information on the plot by modifying the way the surface is colored.
This will allows having kinda of 4D plot: to the 3D representation of the surface generated by the first (m x n)
matrix, the fourth dimension will be represented by the data contained in the second (m x n) matrix.
It is possible to create such a plot by calling surf with 4 input:
surf(x,y,z,C)
where the C parameter is the second matrix (which has to be of the same size of z) and is used to define the color
of the surface.
In the following example, the C matrix is generated by the function:
C=10*sin(0.5*(x.^2.+y.^2))*33;
over the interval [-pi,pi]
The surface generated by C is

Figure 2
Now we can call surf with four input:
figure
surf(x,y,z,C)
% shading interp
xlabel('X Axis');
ylabel('Y Axis');
zlabel('Z Axis');
grid minor
colormap('hot')
colorbar

Figure 3
Comparing Figure 1 and Figure 3, we can notice that:
the shape of the surface corresponds to the z values (the first (m x n) matrix)
the colour of the surface (and its range, given by the colorbar) corresponds to the C values (the first (m x n)
matrix)

Figure 4
Of course, it is possible to swap z and C in the plot to have the shape of the surface given by the C matrix and the
color given by the z matrix:
figure
surf(x,y,C,z)
% shading interp
xlabel('X Axis');
ylabel('Y Axis');
zlabel('Z Axis');
grid minor
colormap('hot')
colorbar
and to compare Figure 2 with Figure 4
Section 16.5: Fast drawing

There are three main ways to do sequential plot or animations: plot(x,y), set(h , 'XData' , y, 'YData' , y)
and animatedline. If you want your animation to be smooth, you need efficient drawing, and the three methods
are not equivalent.
% Plot a sin with increasing phase shift in 500 steps

x = linspace(0 , 2*pi , 100);
figure
tic
for thetha = linspace(0 , 10*pi , 500)
y = sin(x + thetha);
plot(x,y)
drawnow
end

toc
I get 5.278172 seconds. The plot function basically deletes and recreates the line object each time. A more efficient
way to update a plot is to use the XData and YData properties of the Line object.
tic
h = []; % Handle of line object
if isempty(h)
% If Line still does not exist, create it
h = plot(x,y);
else
% If Line exists, update it
set(h , 'YData' , y)
end
drawnow
end
toc
Now I get 2.741996 seconds, much better!
animatedline is a relatively new function, introduced in 2014b. Let's see how it fares:
tic
h = animatedline;
clearpoints(h)
addpoints(h , x , y)
drawnow
end
toc
3.360569 seconds, not as good as updating an existing plot, but still better than plot(x,y).
Of course, if you have to plot a single line, like in this example, the three methods are almost equivalent and give
smooth animations. But if you have more complex plots, updating existing Line objects will make a difference.
Section 16.6: Polygon(s)

Create vectors to hold the x- and y-locations of vertices, feed these into patch.
Single Polygon
X=rand(1,4); Y=rand(1,4);
h=patch(X,Y,'red');

Multiple Polygons
Each polygon's vertices occupy one column of each of X, Y.
X=rand(4,3); Y=rand(4,3);
for i=2:3
X(:,i)=X(:,i)+(i-1); % create horizontal offsets for visibility
end
h=patch(X,Y,'red');

Chapter 17: Debugging
Parameter Details
Name of .m file (without extension), e.g. fit. This parameter is (Required) unless setting special
file
conditional breakpoint types such as dbstop if error or dbstop if naninf.
Line number where the breakpoint should be placed. If the specified line does not contain runnable
location
code, the breakpoint will be placed on the first valid line after the specified one.
Any expression or combination thereof that evaluates to a boolean value. Examples: ind == 1, nargin
expression
< 4 && isdir('Q:\').
Section 17.1: Working with Breakpoints

Definition
In software development, a breakpoint is an intentional stopping or pausing place in a program, put in place
for debugging purposes.
More generally, a breakpoint is a means of acquiring knowledge about a program during its execution. During
the interruption, the programmer inspects the test environment (general purpose registers, memory, logs, files,
etc.) to find out whether the program is functioning as expected. In practice, a breakpoint consists of one or
more conditions that determine when a program's execution should be interrupted.
-Wikipedia
Breakpoints in MATLAB
Motivation
In MATLAB, when execution pauses at a breakpoint, variables existing in the current workspace (a.k.a. scope) or any
of the calling workspaces, can be inspected (and usually also modified).
Types of Breakpoints
MATLAB allow users to place two types of breakpoints in .m files:
Standard (or "unrestricted") breakpoints (shown in red) - pause execution whenever the marked line is
reached.
"Conditional" breakpoints (shown in yellow) - pause execution whenever the marked line is reached AND the
condition defined in the breakpoint is evaluated as true.
Placing Breakpoints
Both types of breakpoints can be created in several ways:

Using the MATLAB Editor GUI, by right clicking the horizontal line next to the line number.
Using the dbstop command:
% Create an unrestricted breakpoint:

dbstop in file at location
% Create a conditional breakpoint:
dbstop in file at location if expression
% Examples and special cases:

dbstop in fit at 99 % Standard unrestricted breakpoint.
dbstop in fit at 99 if nargin==3 % Standard conditional breakpoint.
dbstop if error % This special type of breakpoint is not limited to a specific file, and
% will trigger *whenever* an error is encountered in "debuggable" code.
dbstop in file % This will create an unrestricted breakpoint on the first executable line
% of "file".
dbstop if naninf % This special breakpoint will trigger whenever a computation result
% contains either a NaN (indicates a division by 0) or an Inf
Using keyboard shortcuts: the default key for creating a standard breakpoint on Windows is F12 ; the
default key for conditional breakpoints is unset.
Disabling and Re-enabling Breakpoints
Disable a breakpoint to temporarily ignore it: disabled breakpoints do not pause execution. Disabling a breakpoint
can be done in several ways:
Right click on the red/yellow breakpoint circle > Disable Breakpoint.

Left click on a conditional (yellow) breakpoint.
In the Editor tab > Breakpoints > Enable\Disable.
Removing Breakpoints
All breakpoints remain in a file until removed, either manually or automatically. Breakpoints are cleared
automatically when ending the MATLAB session (i.e. terminating the program). Clearing breakpoints manually is
done in one of the following ways:
Using the dbclear command:
dbclear all
dbclear in file
dbclear in file at location
dbclear if condition
Left clicking a standard breakpoint icon, or a disabled conditional breakpoint icon.
Right clicking on any breakpoint > Clear Breakpoint.
In the Editor tab > Breakpoints > Clear All.
In pre-R2015b versions of MATLAB, using the command clear.
Resuming Execution

When execution is paused at a breakpoint, there are two ways to continue executing the program:
Execute the current line and pause again before the next line.
F10 1 in the Editor, dbstep in the Command Window, "Step" in Ribbon > Editor > DEBUG.
Execute until the next breakpoint (if there are no more breakpoints, the execution proceeds until the end of
the program).
F12 1 in the Editor, dbcont in the Command Window, "Continue" in Ribbon > Editor > DEBUG.
1 - default on Windows.
Section 17.2: Debugging Java code invoked by MATLAB

Overview
In order to debug Java classes that are called during MATLAB execution, it is necessary to perform two steps:
1. Run MATLAB in JVM debugging mode.

2. Attach a Java debugger to the MATLAB process.
When MATLAB is started in JVM debugging mode, the following message appears in the command window:
JVM is being started with debugging enabled.

Use "jdb -connect com.sun.jdi.SocketAttach:port=4444" to attach debugger.
MATLAB end
Windows:
Create a shortcut to the MATLAB executable (matlab.exe) and add the -jdb flag at the end as shown below:

When running MATLAB using this shortcut JVM debugging will be enabled.
Alternatively the java.opts file can be created/updated. This file is stored in "matlab-root\bin\arch", where "matlab-
root" is the MATLAB installation directoy and "arch" is the architecture (e.g. "win32").
The following should be added in the file:
-Xdebug
-Xrunjdwp:transport=dt_socket,address=1044,server=y,suspend=n
Debugger end
IntelliJ IDEA
Attaching this debugger requires the creation of a "remote debugging" configuration with the port exposed by
MATLAB:

Then the debugger is started:
If everything is working as expected, the following message will appear in the console:

Chapter 18: Financial Applications
Section 18.1: Random Walk
The following is an example that displays 5 one-dimensional random walks of 200 steps:
y = cumsum(rand(200,5) - 0.5);
plot(y)
legend('1', '2', '3', '4', '5')
title('random walks')
In the above code, y is a matrix of 5 columns, each of length 200. Since x is omitted, it defaults to the row numbers
of y (equivalent to using x=1:200 as the x-axis). This way the plot function plots multiple y-vectors against the same
x-vector, each using a different color automatically.
Section 18.2: Univariate Geometric Brownian Motion

The dynamics of the Geometric Brownian Motion (GBM) are described by the following stochastic differential
equation (SDE):
I can use the exact solution to the SDE
to generate paths that follow a GBM.

Given daily parameters for a year-long simulation
mu = 0.08/250;
sigma = 0.25/sqrt(250);
dt = 1/250;
npaths = 100;
nsteps = 250;
S0 = 23.2;
we can get the Brownian Motion (BM) W starting at 0 and use it to obtain the GBM starting at S0
% BM
epsilon = randn(nsteps, npaths);
W = [zeros(1,npaths); sqrt(dt)*cumsum(epsilon)];
% GBM
t = (0:nsteps)'*dt;
Y = bsxfun(@plus, (mu-0.5*sigma.^2)*t, sigma*W);
Y = S0*exp(Y);
Which produces the paths
plot(Y)

Chapter 19: Fourier Transforms and
Inverse Fourier Transforms
Parameter Description
X this is your input Time-Domain signal, it should be a vector of numerics.
this is the NFFT parameter known as Transform Length, think of it as resolution of your FFT result, it
n
MUST be a number that is a power of 2 (i.e. 64,128,256...2^N)
this is the dimension you want to compute FFT on, use 1 if you want to compute your FFT in the
dim horizontal direction and 2 if you want to compute your FFT in the vertical direction - Note this
parameter is usually left blank, as the function is capable of detecting the direction of your vector.
Section 19.1: Implement a simple Fourier Transform in Matlab

Fourier Transform is probably the first lesson in Digital Signal Processing, it's application is everywhere and it is a
powerful tool when it comes to analyze data (in all sectors) or signals. Matlab has a set of powerful toolboxes for
Fourier Transform. In this example, we will use Fourier Transform to analyze a basic sine-wave signal and generate
what is sometimes known as a Periodogram using FFT:
%Signal Generation
A1=10; % Amplitude 1
w1=2*pi*0.2; % Angular frequency 1
Ts=1; % Sampling time
N=64; % Number of process samples to be generated
K=5; % Number of independent process realizations
sgm=1; % Standard deviation of the noise
n=repmat([0:N-1].',1,K); % Generate resolution
phi1=repmat(rand(1,K)*2*pi,N,1); % Random phase matrix 1
x=A1*sin(w1*n*Ts+phi1)+A2*sin(w2*n*Ts+phi2)+sgm*randn(N,K); % Resulting Signal
NFFT=256; % FFT length

F=fft(x,NFFT); % Fast Fourier Transform Result
Z=1/N*abs(F).^2; % Convert FFT result into a Periodogram

Note that the Discrete Fourier Transform is implemented by Fast Fourier Transform (fft) in Matlab, both will yield
the same result, but FFT is a fast implementation of DFT.
figure
w=linspace(0,2,NFFT);
plot(w,10*log10(Z)),grid;
xlabel('w [\pi rad/s]')
ylabel('Z(f) [dB]')
title('Frequency Range: [ 0 , \omega_s ]')
Section 19.2: Images and multidimensional FTs

In medical imaging, spectroscopy, image processing, cryptography and other areas of science and engineering it is
often the case that one wishes to compute multidimensional Fourier transforms of images. This is quite
straightforward in Matlab: (multidimensional) images are just n-dimensional matrices, after all, and Fourier
transforms are linear operators: one just iteratively Fourier transforms along other dimensions. Matlab provides
fft2 and ifft2 to do this in 2-d, or fftn in n-dimensions.
One potential pitfall is that the Fourier transform of images are usually shown "centric ordered", i.e. with the origin
of k-space in the middle of the picture. Matlab provides the fftshift command to swap the location of the DC
components of the Fourier transform appropriately. This ordering notation makes it substantially easier to perform
common image processing techniques, one of which is illustrated below.
Zero filling
One "quick and dirty" way to interpolate a small image to a larger size is to Fourier transform it, pad the Fourier
transform with zeros, and then take the inverse transform. This effectively interpolates between each pixel with a
sinc shaped basis function, and is commonly used to up-scale low resolution medical imaging data. Let's start by
loading a built-in image example
%Load example image

I=imread('coins.png'); %Load example data -- coins.png is builtin to Matlab
I=double(I); %Convert to double precision -- imread returns integers
imageSize = size(I); % I is a 246 x 300 2D image
%Display it
imagesc(I); colormap gray; axis equal;
%imagesc displays images scaled to maximum intensity
We can now obtain the Fourier transform of I. To illustrate what fftshift does, let's compare the two methods:
% Fourier transform
%Obtain the centric- and non-centric ordered Fourier transform of I
k=fftshift(fft2(fftshift(I)));
kwrong=fft2(I);
%Just for the sake of comparison, show the magnitude of both transforms:
figure; subplot(2,1,1);
imagesc(abs(k),[0 1e4]); colormap gray; axis equal;
subplot(2,1,2);
imagesc(abs(kwrong),[0 1e4]); colormap gray; axis equal;
%(The second argument to imagesc sets the colour axis to make the difference clear).

We now have obtained the 2D FT of an example image. To zero-fill it, we want to take each k-space, pad the edges
with zeros, and then take the back transform:
%Zero fill
kzf = zeros(imageSize .* 2); %Generate a 492x600 empty array to put the result in
kzf(end/4:3*end/4-1,end/4:3*end/4-1) = k; %Put k in the middle
kzfwrong = zeros(imageSize .* 2); %Generate a 492x600 empty array to put the result in
kzfwrong(end/4:3*end/4-1,end/4:3*end/4-1) = kwrong; %Put k in the middle
%Show the differences again

%Just for the sake of comparison, show the magnitude of both transforms:
imagesc(abs(kzf),[0 1e4]); colormap gray; axis equal;
subplot(2,1,2);
imagesc(abs(kzfwrong),[0 1e4]); colormap gray; axis equal;
%(The second argument to imagesc sets the colour axis to make the difference clear).
At this point, the result fairly unremarkable:

Once we then take the back-transforms, we can see that (correctly!) zero-filling data provides a sensible method for
interpolation:
% Take the back transform and view

Izf = fftshift(ifft2(ifftshift(kzf)));
Izfwrong = ifft2(kzfwrong);
imagesc(abs(Izf)); colormap gray; axis equal;
title('Zero-filled image');
subplot(1,3,2);
imagesc(abs(Izfwrong)); colormap gray; axis equal;
title('Incorrectly zero-filled image');
subplot(1,3,3);
imagesc(I); colormap gray; axis equal;
title('Original image');
set(gcf,'color','w');

Note that the zero-filled image size is double that of the original. One can zero fill by more than a factor of two in
each dimension, although obviously doing so does not arbitrarily increase the size of an image.
Hints, tips, 3D and beyond
The above example holds for 3D images (as are often generated by medical imaging techniques or confocal
microscopy, for example), but require fft2 to be replaced by fftn(I, 3), for example. Due to the somewhat
cumbersome nature of writing fftshift(fft(fftshift(... several times, it is quite common to define functions
such as fft2c locally to provide easier syntax locally -- such as:
function y = fft2c(x)
y = fftshift(fft2(fftshift(x)));
Note that the FFT is fast, but large, multidimensional Fourier transforms will still take time on a modern computer.
It is additionally inherently complex: the magnitude of k-space was shown above, but the phase is absolutely vital;
translations in the image domain are equivalent to a phase ramp in the Fourier domain. There are several far more
complex operations that one may wish to do in the Fourier domain, such as filtering high or low spatial frequencies
(by multiplying it with a filter), or masking out discrete points corresponding to noise. There is correspondingly a
large quantity of community generated code for handling common Fourier operations available on Matlab's main
community repository site, the File Exchange.

Section 19.3: Inverse Fourier Transforms
One of the major benefit of Fourier Transform is its ability to inverse back in to the Time Domain without losing
information. Let us consider the same Signal we used in the previous example:
Ts=1; % Sampling time
N=64; % Number of process samples to be generated
K=1; % Number of independent process realizations
sgm=1; % Standard deviation of the noise
n=repmat([0:N-1].',1,K); % Generate resolution
x=A1*sin(w1*n*Ts+phi1)+A2*sin(w2*n*Ts+phi2)+sgm*randn(N,K); % Resulting Signal
NFFT=256; % FFT length

F=fft(x,NFFT); % FFT result of time domain signal
If we open F in Matlab, we will find that it is a matrix of complex numbers, a real part and an imaginary part. By
definition, in order to recover the original Time Domain signal, we need both the Real (which represents Magnitude
variation) and the Imaginary (which represents Phase variation), so to return to the Time Domain, one may simply
want to:
TD = ifft(F,NFFT); %Returns the Inverse of F in Time Domain
Note here that TD returned would be length 256 because we set NFFT to 256, however, the length of x is only 64, so
Matlab will pad zeros to the end of the TD transform. So for example, if NFFT was 1024 and the length was 64, then
TD returned will be 64 + 960 zeros. Also note that due to floating point rounding, you might get something like 3.1 *
10e-20 but for general purposed: For any X, ifft(fft(X)) equals X to within roundoff error.
Let us say for a moment that after the transformation, we did something and are only left with the REAL part of the
FFT:
R = real(F); %Give the Real Part of the FFT

TDR = ifft(R,NFFT); %Give the Time Domain of the Real Part of the FFT
This means that we are losing the imaginary part of our FFT, and therefore, we are losing information in this reverse
process. To preserve the original without losing information, you should always keep the imaginary part of the FFT
using imag and apply your functions to either both or the real part.
figure
subplot(3,1,1)
plot(x);xlabel('time samples');ylabel('magnitude');title('Original Time Domain Signal')
subplot(3,1,2)
plot(TD(1:64));xlabel('time samples');ylabel('magnitude');title('Inverse Fourier Transformed - Time
Domain Signal')
subplot(3,1,3)
plot(TDR(1:64));xlabel('time samples');ylabel('magnitude');title('Real part of IFFT transformed
Time Domain Signal')

Chapter 20: Ordinary Dierential
Equations (ODE) Solvers
Section 20.1: Example for odeset
First we initialize our initial value problem we want to solve.
odefun = @(t,y) cos(y).^2*sin(t);

tspan = [0 16*pi];
y0=1;
We then use the ode45 function without any specified options to solve this problem. To compare it later we plot the
trajectory.
[t,y] = ode45(odefun, tspan, y0);

plot(t,y,'-o');
We now set a narrow relative and a narrow absolut limit of tolerance for our problem.
options = odeset('RelTol',1e-2,'AbsTol',1e-2);
[t,y] = ode45(odefun, tspan, y0, options);
plot(t,y,'-o');
We set tight relative and narrow absolut limit of tolerance.
plot(t,y,'-o');
We set narrow relative and tight absolute limit of tolerance. As in the previous examples with narrow limits of
tolerance one sees the trajectory being completely different from the first plot without any specific options.
plot(t,y,'-o');
We set tight relative and tight absolut limit of tolerance. Comparing the result with the other plot will underline the
errors made calculating with narrow tolerance limits.
plot(t,y,'-o');
The following should demonstrate the trade-off between precision and run-time.
tic;
time1 = toc;
plot(t,y,'-o');
For comparison we tighten the limit of tolerance for absolute and relative error. We now can see that without large
gain in precision it will take considerably longer to solve our initial value problem.

tic;
time2 = toc;
plot(t,y,'-o');

Chapter 21: Interpolation with MATLAB
Section 21.1: Piecewise interpolation 2 dimensional
We initialize the data:
[X,Y] = meshgrid(1:2:10);
Z = X.*cos(Y) - Y.*sin(X);
The surface looks like the following.
Now we set the points where we want to interpolate:
[Vx,Vy] = meshgrid(1:0.25:10);
We now can perform nearest interpolation,
Vz = interp2(X,Y,Z,Vx,Vy,'nearest');

linear interpolation,
Vz = interp2(X,Y,Z,Vx,Vy,'linear');
cubic interpolation
Vz = interp2(X,Y,Z,Vx,Vy,'cubic');

or spline interpolation:
Vz = interp2(X,Y,Z,Vx,Vy,'spline');
Section 21.2: Piecewise interpolation 1 dimensional

We will use the following data:
x = 1:5:50;
y = randi([-10 10],1,10);

Hereby x and y are the coordinates of the data points and z are the points we need information about.
z = 0:0.25:50;
One way to find the y-values of z is piecewise linear interpolation.
z_y = interp1(x,y,z,'linear');

Hereby one calculates the line between two adjacent points and gets z_y by assuming that the point would be an
element of those lines.
interp1 provides other options too like nearest interpolation,
z_y = interp1(x,y,z, 'nearest');
next interpolation,
z_y = interp1(x,y,z, 'next');

previous interpolation,
z_y = interp1(x,y,z, 'previous');
Shape-preserving piecewise cubic interpolation,
z_y = interp1(x,y,z, 'pchip');

cubic convolution, z_y = interp1(x,y,z, 'v5cubic');
and spline interpolation
z_y = interp1(x,y,z, 'spline');

Hereby are nearst, next and previous interpolation piecewise constant interpolations.
Section 21.3: Polynomial interpolation

We initialize the data we want to interpolate:
x = 0:0.5:10;
y = sin(x/2);
This means the underlying function for the data in the interval [0,10] is sinusoidal. Now the coefficients of the
approximating polynómials are being calculated:
p1 = polyfit(x,y,1);
Hereby is x the x-value and y the y-value of our data points and the third number is the order/degree of the
polynomial. We now set the grid we want to compute our interpolating function on:
zx = 0:0.1:10;
and calculate the y-values:
zy1 = polyval(p1,zx);
One can see that the approximation error for the sample gets smaller when the degree of the polynomial increases.

While the approximation of the straight line in this example has larger errors the order 3 polynomial approximates
the sinus function in this intervall relatively good.
The interpolation with order 5 and order 10 polynomials has almost no apprroximation error.
However if we consider the out of sample performance one sees that too high orders tend to overfit and therefore
perform badly out of sample. We set

zx = -10:0.1:40;
p10 = polyfit(X,Y,10);
p20 = polyfit(X,Y,20);
and
If we take a look at the plot we see that the out of sample performance is best for the order 1
and keeps getting worse with increasing degree.

Chapter 22: Integration
Section 22.1: Integral, integral2, integral3
1 dimensional
To integrate a one dimensional function
f = @(x) sin(x).^3 + 1;
within the range
xmin = 2;
xmax = 8;
one can call the function
q = integral(f,xmin,xmax);
it's also possible to set boundarys for relative and absolute errors
q = integral(f,xmin,xmax, 'RelTol',10e-6, 'AbsTol',10-4);
2 dimensional
If one wants to integrate a two dimensional function
f = @(x,y) sin(x).^y ;
within the range
xmin = 2;
xmax = 8;
ymin = 1;
ymax = 4;
one calls the function
q = integral2(f,xmin,xmax,ymin,ymax);
Like in the other case it's possible to limit the tolerances
q = integral2(f,xmin,xmax,ymin,ymax, 'RelTol',10e-6, 'AbsTol',10-4);
3 dimensional
Integrating a three dimensional function
f = @(x,y,z) sin(x).^y - cos(z) ;
within the range
xmin = 2;
xmax = 8;

ymin = 1;
ymax = 4;
zmin = 6;
zmax = 13;
is performed by calling
q = integral3(f,xmin,xmax,ymin,ymax, zmin, zmax);
Again it's possible to limit the tolerances
q = integral3(f,xmin,xmax,ymin,ymax, zmin, zmax, 'RelTol',10e-6, 'AbsTol',10-4);

Chapter 23: Reading large files
Section 23.1: textscan
Assume you have formatted data in a large text file or string, e.g.
Data,2015-09-16,15:41:52;781,780.000000,0.0034,2.2345
Data,2015-09-16,15:41:52;791,790.000000,0.1255,96.5948
Data,2015-09-16,15:41:52;801,800.000000,1.5123,0.0043
one may use textscan to read this quite fast. To do so, get a file identifier of the text file with fopen:
fid = fopen('path/to/myfile');
Assume for the data in this example, we want to ignore the first column "Data", read the date and time as strings,
and read the rest of the columns as doubles, i.e.
Data , 2015-09-16 , 15:41:52;801 , 800.000000 , 1.5123 , 0.0043

ignore string string double double double
To do this, call:
data = textscan(fid,'%*s %s %s %f %f %f','Delimiter',',');
The asterisk in %*s means "ignore this column". %s means "interpret as a string". %f means "interpret as doubles
(floats)". Finally, 'Delimiter',',' states that all commas should be interpreted as the delimiter between each
column.
To sum up:
data = textscan(fid,'%*s %s %s %f %f %f','Delimiter',',');
data now contains a cell array with each column in a cell.
Section 23.2: Date and time strings to numeric array fast

Converting date and time strings to numeric arrays can be done with datenum, though it may take as much as half
the time of reading a large data file.
Consider the data in example Textscan. By, again, using textscan and interpret date and time as integers, they can
rapidly be converted into a numeric array.
I.e. a line in the example data would be interpreted as:
Data , 2015 - 09 - 16 , 15 : 41 : 52 ; 801 , 800.000000 , 1.5123 , 0.0043

ignore double double double double double double double double double double
which will be read as:
data = textscan(fid,'%*s %f %f %f %f %f %f %f %f %f %f','Delimiter',',-:;');
fclose(fid);

Now:
y = data{1}; % year
m = data{2}; % month
d = data{3}; % day
H = data{4}; % hours
M = data{5}; % minutes
S = data{6}; % seconds
F = data{7}; % milliseconds
% Translation from month to days

ms = [0,31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334];
n = length(y); % Number of elements

Time = zeros(n,1); % Declare numeric time array
% Algorithm for calculating numeric time array

for k = 1:n
Time(k) = y(k)*365 + ms(m(k)) + d(k) + floor(y(k)/4)...
- floor(y(k)/100) + floor(y(k)/400) + (mod(y(k),4)~=0)...
- (mod(y(k),100)~=0) + (mod(y(k),400)~=0)...
+ (H(k)*3600 + M(k)*60 + S(k) + F(k)/1000)/86400 + 1;
end
Using datenum on 566,678 elements required 6.626570 seconds, whilst the method above required 0.048334
seconds, i.e. 0.73% of the time for datenum or ~137 times faster.

Chapter 24: Usage of àccumarray()`
Function
Parameter Details
subscriptArray Subscript matrix, specified as a vector of indices, matrix of indices, or cell array of index vectors.
valuesArray Data, specified as a vector or a scalar.
sizeOfOutput Size of output array, specified as a vector of positive integers.
Function to be applied to each set of items during aggregation, specified as a function handle or
funcHandle
[].
fillVal Fill value, for when subs does not reference each element in the output.
isSparse Should the output be a sparse array?
accumarray allows to aggregate items of an array in various ways, potentially applying some function to the items in
the process. accumarray can be thought of as a lightweight reducer (see also: Introduction to MapReduce ).
This topic will contain common scenarios where accumarray is especially useful.
Section 24.1: Apply Filter to Image Patches and Set Each Pixel
as the Mean of the Result of Each Patch
Many modern Image Processing algorithms use patches are their basic element to work on.
For instance one could denoise patches (See BM3D Algorithm).
Yet when building the image form the processed patches we have many results for the same pixel.
One way to deal with it is taking the average (Empirical Mean) of all values of the same pixel.
The following code shows how to break an image into patches and them reconstruct the image from patches using
the average by using [accumarray()][1]:
numRows = 5;
numCols = 5;
numRowsPatch = 3;
numColsPatch = 3;
% The Image
mI = rand([numRows, numCols]);
% Decomposing into Patches - Each pixel is part of many patches (Neglecting

% boundariwes, each pixel is part of (numRowsPatch * numColsPatch) patches).
mY = ImageToColumnsSliding(mI, [numRowsPatch, numColsPatch]);
% Here one would apply some operation which work on patches
% Creating image of the index of each pixel

mPxIdx = reshape(1:(numRows * numCols), [numRows, numCols]);
% Creating patches of the same indices

mSubsAccu = ImageToColumnsSliding(mPxIdx, [numRowsPatch, numColsPatch]);
% Reconstruct the image - Option A

mO = accumarray(mSubsAccu(:), mY(:)) ./ accumarray(mSubsAccu(:), 1);
% Reconstruct the image - Option B

mO = accumarray(mSubsAccu, mY(:), [(numRows * numCols), 1], @(x) mean(x));

% Rehsape the Vector into the Image
mO = reshape(mO, [numRows, numCols]);
Section 24.2: Finding the maximum value among elements

grouped by another vector
This is an official MATLAB example
Consider the following code:
month = [1;1;2;3;8;1;3;4;9;11;9;12;3;4;11];
temperature = [57;61;60;62;45;59;64;66;40;56;38;65;61;64;55];
maxTemp = accumarray(month,temperature,[],@max);
The image below demonstrates the computation process done by accumarray in this case:
In this example, all values that have the same month are first collected, and then the function specified by the 4th
input to accumarray (in this case, @max) is applied to each such set.

Chapter 25: Introduction to MEX API
Section 25.1: Check number of inputs/outputs in a C++ MEX-
file
In this example we will write a basic program that checks the number of inputs and outputs passed to a MEX-
function.
As a starting point, we need to create a C++ file implementing the "MEX gateway". This is the function executed
when the file is called from MATLAB.
testinputs.cpp
// MathWorks provided header file
#include "mex.h"
// gateway function
void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[])
{
// This function will error if number of inputs its not 3 or 4
// This function will error if number of outputs is more than 1
// Check inputs:
if (nrhs < 3 || nrhs > 4) {
mexErrMsgIdAndTxt("Testinputs:ErrorIdIn",
"Invalid number of inputs to MEX file.");
}
// Check outputs:
if (nlhs > 1) {
mexErrMsgIdAndTxt("Testinputs:ErrorIdOut",
"Invalid number of outputs to MEX file.");
}
}
First, we include the mex.h header which contains definitions of all the required functions and data types to work
with the MEX API. Then we implement the function mexFunction as shown, where its signature must not change,
independent of the inputs/outputs actually used. The function parameters are as follows:
nlhs: Number of outputs requested.

*plhs[]: Array containing all the outputs in MEX API format.
nrhs: Number of inputs passed.
*prhs[]: Array containing all the inputs in MEX API format.
Next, we check the number of inputs/outputs arguments, and if the validation fails, an error is thrown using
mexErrMsgIdAndTxt function (it expects somename:iD format identifier, a simple "ID" won't work).
Once the file is compiled as mex testinputs.cpp, the function can be called in MATLAB as:
>> testinputs(2,3)
Error using testinputs. Invalid number of inputs to MEX file.
>> testinputs(2,3,5)
>> [~,~] = testinputs(2,3,3)

Error using testinputs. Invalid number of outputs to MEX file.

Section 25.2: Input a string, modify it in C, and output it
In this example, we illustrate string manipulation in MATLAB MEX. We will create a MEX-function that accepts a
string as input from MATLAB, copy the data into C-string, modify it and convert it back to mxArray returned to the
MATLAB side.
The main objective of this example is to show how strings can be converted to C/C++ from MATLAB and vice versa.
stringIO.cpp
#include "mex.h"
#include <cstring>
void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[])
{
// check number of arguments
if (nrhs != 1 || nlhs > 1) {
mexErrMsgIdAndTxt("StringIO:WrongNumArgs", "Wrong number of arguments.");
}
// check if input is a string

if (mxIsChar(prhs[0])) {
mexErrMsgIdAndTxt("StringIO:TypeError", "Input is not a string");
}
// copy characters data from mxArray to a C-style string (null-terminated)

char *str = mxArrayToString(prhs[0]);
// manipulate the string in some way

if (strcmp("theOneString", str) == 0) {
str[0] = 'T'; // capitalize first letter
} else {
str[0] = ' '; // do something else?
}
// return the new modified string

plhs[0] = mxCreateString(str);
// free allocated memory

mxFree(str);
}
The relevant functions in this example are:
mxIsChar to test if an mxArray is of mxCHAR type.

mxArrayToString to copy the data of a mxArray string to a char * buffer.
mxCreateString to create an mxArray string from a char*.
As a side note, if you only want to read the string, and not modify it, remember to declare it as const char* for
speed and robustness.
Finally, once compiled we can call it from MATLAB as:
>> mex stringIO.cpp
>> strOut = stringIO('theOneString')

strOut =
TheOneString
>> strOut = stringIO('somethingelse')

strOut=
omethingelse
Section 25.3: Passing a struct by field names

This example illustrates how to read various-type struct entries from MATLAB, and pass it to C equivalent type
variables.
While it is possible and easy to figure out from the example how to load fields by numbers, it is here achieved via
comparing the field names to strings. Thus the struct fields can be addressed by their field names and variables in it
can be read by C.
structIn.c
#include "mex.h"
#include <string.h> // strcmp
void mexFunction (int nlhs, mxArray *plhs[],

int nrhs, const mxArray *prhs[])
{
// helpers
double* double_ptr;
unsigned int i; // loop variable
// to be read variables
bool optimal;
int randomseed;
unsigned int desiredNodes;
if (!mxIsStruct(prhs[0])) {
mexErrMsgTxt("First argument has to be a parameter struct!");
}
for (i=0; i<mxGetNumberOfFields(prhs[0]); i++) {
if (0==strcmp(mxGetFieldNameByNumber(prhs[0],i),"randomseed")) {
mxArray *p = mxGetFieldByNumber(prhs[0],0,i);
randomseed = *mxGetPr(p);
}
if (0==strcmp(mxGetFieldNameByNumber(prhs[0],i),"optimal")) {
optimal = (bool)*mxGetPr(p);
}
if (0==strcmp(mxGetFieldNameByNumber(prhs[0],i),"numNodes")) {
desiredNodes = *mxGetPr(p);
}
}
}
The loop over i runs over every field in the given struct, while the if(0==strcmp)-parts compare the matlab field's
name to the given string. If it is a match, the corresponding value is extracted to a C variable.
Section 25.4: Pass a 3D matrix from MATLAB to C

In this example we illustrate how to take a double real-type 3D matrix from MATLAB, and pass it to a C double*
array.
The main objectives of this example are showing how to obtain data from MATLAB MEX arrays and to highlight
some small details in matrix storage and handling.

matrixIn.cpp
#include "mex.h"
void mexFunction(int nlhs , mxArray *plhs[],

int nrhs, mxArray const *prhs[]){
// check amount of inputs
if (nrhs!=1) {
mexErrMsgIdAndTxt("matrixIn:InvalidInput", "Invalid number of inputs to MEX file.");
}
// check type of input

if( !mxIsDouble(prhs[0]) || mxIsComplex(prhs[0])){
mexErrMsgIdAndTxt("matrixIn:InvalidType", "Input matrix must be a double, non-complex
array.");
}
// extract the data

double const * const matrixAux= static_cast<double const *>(mxGetData(prhs[0]));
// Get matrix size

const mwSize *sizeInputMatrix= mxGetDimensions(prhs[0]);
// allocate array in C. Note: its 1D array, not 3D even if our input is 3D

double* matrixInC= (double*)malloc(sizeInputMatrix[0] *sizeInputMatrix[1] *sizeInputMatrix[2]*
sizeof(double));
// MATLAB is column major, not row major (as C). We need to reorder the numbers
// Basically permutes dimensions
// NOTE: the ordering of the loops is optimized for fastest memory access!
// This improves the speed in about 300%
const int size0 = sizeInputMatrix[0]; // Const makes compiler optimization kick in

const int size1 = sizeInputMatrix[1];
const int size2 = sizeInputMatrix[2];
for (int j = 0; j < size2; j++)

{
int jOffset = j*size0*size1; // this saves re-computation time
for (int k = 0; k < size0; k++)
{
int kOffset = k*size1; // this saves re-computation time
for (int i = 0; i < size1; i++)
{
int iOffset = i*size0;
matrixInC[i + jOffset + kOffset] = matrixAux[iOffset + jOffset + k];
}
}
}
// we are done!
// Use your C matrix here
// free memory
free(matrixInC);
return;
}
The relevant concepts to be aware of:

MATLAB matrices are all 1D in memory, no matter how many dimensions they have when used in MATLAB.
This is also true for most (if not all) main matrix representation in C/C++ libraries, as allows optimization and
faster execution.
You need to explicitly copy matrices from MATLAB to C in a loop.
MATLAB matrices are stored in column major order, as in Fortran, but C/C++ and most modern languages
are row major. It is important to permute the input matrix , or else the data will look completely different.
The relevant function in this example are:
mxIsDouble checks if input is double type.

mxIsComplex checks if input is real or imaginary.
mxGetData returns a pointer to the real data in the input array. NULL if there is no real data.
mxGetDimensions returns an pointer to a mwSize array, with the size of the dimension in each index.

Chapter 26: Performance and
Benchmarking
Section 26.1: Identifying performance bottlenecks using the
Profiler
The MATLAB Profiler is a tool for software profiling of MATLAB code. Using the Profiler, it is possible to obtain a
visual representation of both execution time and memory consumption.
Running the Profiler can be done in two ways:
Clicking the "Run and Time" button in the MATLAB GUI while having some .m file open in the editor (added in
R2012b).
Programmatically, using:
profile on
<some code we want to test>
profile off
Below is some sample code and the result of its profiling:
function docTest
for ind1 = 1:100

[~] = var(...
sum(...
randn(1000)));
end
spy

From the above we learn that the spy function takes about 25% of the total execution time. In the case of "real
code", a function that takes such a large percentage of execution time would be a good candidate for optimization,
as opposed to functions analogous to var and cla whose optimization should be avoided.
Moreover, it is possible to click on entries in the Function Name column to see a detailed breakdown of execution
time for that entry. Here's the example of clicking spy:

It is also possible to profile memory consumption by executing profile('-memory') before running the Profiler.

Section 26.2: Comparing execution time of multiple functions
The widely used combination of tic and toc can provide a rough idea of the execution time of a function or code
snippets.
For comparing several functions it shouldn't be used. Why? It is almost impossible to provide equal conditions for all
code snippets to compare within a script using above solution. Maybe the functions share the same function space
and common variables, so later called functions and code snippets already take advantage of previously initialized
variables and functions. Also the there is no insight whether the JIT compiler would handle these subsequently
called snippets equally.
The dedicated function for benchmarks is timeit. The following example illustrates its use.
There are the array A and the matrix B. It should be determined which row of B is the most similar to A by counting
the number of different elements.
function t = bench()
A = [0 1 1 1 0 0];
B = perms(A);
% functions to compare
fcns = {
@() compare1(A,B);
@() compare2(A,B);
@() compare3(A,B);

@() compare4(A,B);
};
% timeit
t = cellfun(@timeit, fcns);
end
function Z = compare1(A,B)
Z = sum( bsxfun(@eq, A,B) , 2);
end
Z = sum(bsxfun(@xor, A, B),2);
end
A = logical(A);
Z = sum(B(:,~A),2) + sum(~B(:,A),2);
end
Z = pdist2( A, B, 'hamming', 'Smallest', 1 );
end
This way of benchmark was first seen in this answer.
Section 26.3: The importance of preallocation

Arrays in MATLAB are held as continuous blocks in memory, allocated and released automatically by MATLAB.
MATLAB hides memory management operations such as resizing of an array behind easy to use syntax:
a = 1:4
a =
1 2 3 4
a(5) = 10 % or alternatively a = [a, 10]
a =
1 2 3 4 10
It is important to understand that the above is not a trivial operation, a(5) = 10 will cause MATLAB to allocate a
new block of memory of size 5, copy the first 4 numbers over, and set the 5'th to 10. That's a O(numel(a))
operation, and not O(1).
Consider the following:
clear all
n=12345678;
a=0;
tic
for i = 2:n
a(i) = sqrt(a(i-1)) + i;
end
toc
a is reallocated n times in this loop (excluding some optimizations undertaken by MATLAB)! Note that MATLAB gives
us a warning:

"The variable 'a' appears to change size on every loop iteration. Consider preallocating for speed."
What happens when we preallocate?
a=zeros(1,n);
tic
for i = 2:n
a(i) = sqrt(a(i-1)) + i;
end
toc
We can see the runtime is reduced by an order of magnitude.
Methods for preallocation:
MATLAB provides various functions for allocation of vectors and matrices, depending on the specific requirements
of the user. These include: zeros, ones, nan, eye, true etc.
a = zeros(3) % Allocates a 3-by-3 matrix initialized to 0

a =
0 0 0
0 0 0
0 0 0
a = zeros(3, 2) % Allocates a 3-by-2 matrix initialized to 0

a =
0 0
0 0
0 0
a = ones(2, 3, 2) % Allocates a 3 dimensional array (2-by-3-by-2) initialized to 1

a(:,:,1) =
1 1 1
1 1 1
a(:,:,2) =
1 1 1
1 1 1
a = ones(1, 3) * 7 % Allocates a row vector of length 3 initialized to 7

a =
7 7 7
A data type can also be specified:
a = zeros(2, 1, 'uint8'); % allocates an array of type uint8
It is also easy to clone the size of an existing array:
a = ones(3, 4); % a is a 3-by-4 matrix of 1's

b = zeros(size(a)); % b is a 3-by-4 matrix of 0's
And clone the type:
a = ones(3, 4, 'single'); % a is a 3-by-4 matrix of type single

b = zeros(2, 'like', a); % b is a 2-by-2 matrix of type single
note that 'like' also clones complexity and sparsity.
Preallocation is implicitly achieved using any function that returns an array of the final required size, such as rand,
gallery, kron, bsxfun, colon and many others. For example, a common way to allocate vectors with linearly varying
elements is by using the colon operator (with either the 2- or 3-operand variant1):
a = 1:3
a =
1 2 3
a = 2:-3:-4
a =
2 -1 -4
Cell arrays can be allocated using the cell() function in much the same way as zeros().
a = cell(2,3)
a =
[] [] []
[] [] []
Note that cell arrays work by holding pointers to the locations in memory of cell contents. So all preallocation tips
apply to the individual cell array elements as well.
Further reading:
Official MATLAB documentation on "Preallocating Memory".

Official MATLAB documentation on "How MATLAB Allocates Memory".
Preallocation performance on Undocumented matlab.
Understanding Array Preallocation on Loren on the Art of MATLAB
Section 26.4: It's ok to be `single`!

Overview:
The default data type for numeric arrays in MATLAB is double. double is a floating point representation of numbers,
and this format takes 8 bytes (or 64 bits) per value. In some cases, where e.g. dealing only with integers or when
numerical instability is not an imminent issue, such high bit depth may not be required. For this reason, it is advised
to consider the benefits of single precision (or other appropriate types):
Faster execution time (especially noticeable on GPUs).

Half the memory consumption: may succeed where double fails due to an out-of-memory error; more
compact when storing as files.
Converting a variable from any supported data type to single is done using:

sing_var = single(var);
Some commonly used functions (such as: zeros, eye, ones, etc.) that output double values by default, allow
specifying the type/class of the output.
Converting variables in a script to a non-default precision/type/class:
As of July 2016, there exists no documented way to change the default MATLAB data type from double.
In MATLAB, new variables usually mimic the data types of variables used when creating them. To illustrate this,
consider the following example:
A = magic(3);
B = diag(A);
C = 20*B;
>> whos C
C 3x1 24 double
A = single(magic(3)); % A is converted to "single"

B = diag(A);
C = B*double(20); % The stricter type, which in this case is "single", prevails
D = single(size(C)); % It is generally advised to cast to the desired type explicitly.
>> whos C
C 3x1 12 single
Thus, it may seem sufficient to cast/convert several initial variables to have the change permeate throughout the
code - however this is discouraged (see Caveats & Pitfalls below).
Caveats & Pitfalls:
1. Repeated conversions are discouraged due to the introduction of numeric noise (when casting from single
to double) or loss of information (when casting from double to single, or between certain integer types), e.g.
:
double(single(1.2)) == double(1.2)
ans =
0
This can be mitigated somewhat using typecast. See also Be aware of floating point inaccuracy.
2. Relying solely on implicit data-typing (i.e. what MATLAB guesses the type of the output of a computation
should be) is discouraged due to several undesired effects that might arise:
Loss of information: when a double result is expected, but a careless combination of single and double
operands yields single precision.
Unexpectedly high memory consumption: when a single result is expected but a careless computation
results in a double output.
Unnecessary overhead when working with GPUs: when mixing gpuArray types (i.e. variables stored in
VRAM) with non-gpuArray variables (i.e. those usually stored in RAM) the data will have to be
transferred one way or the other before the computation can be performed. This operation takes
time, and can be very noticeable in repetative computations.

Errors when mixing floating-point types with integer types: functions like mtimes (*) are not defined for
mixed inputs of integer and floating point types - and will error. Functions like times (.*) are not
defined at all for integer-type inputs - and will again error.
>> ones(3,3,'int32')*ones(3,3,'int32')
Error using *
MTIMES is not fully supported for integer classes. At least one input must be scalar.
>> ones(3,3,'int32').*ones(3,3,'double')
Error using .*
Integers can only be combined with integers of the same class, or scalar doubles.
For better code readability and reduced risk of unwanted types, a defensive approach is advised, where
variables are explicitly cast to the desired type.
See Also:
MATLAB Documentation: Floating-Point Numbers.

Mathworks' Technical Article: Best Practices for Converting MATLAB Code to Fixed Point.

Chapter 27: Multithreading
Section 27.1: Using parfor to parallelize a loop
You can use parfor to execute the iterations of a loop in parallel:
Example:
poolobj = parpool(2); % Open a parallel pool with 2 workers
s = 0; % Performing some parallel Computations

parfor i=0:9
s = s + 1;
end
disp(s) % Outputs '10'
delete(poolobj); % Close the parallel pool
Note: parfor cannot be nested directly. For parfor nesting use a function in fisrt parfor and add second parfor in
that function.
Example:
parfor i = 1:n
[op] = fun_name(ip);
end
function [op] = fun_name(ip)

parfor j = 1:length(ip)
% Some Computation
end
Section 27.2: Executing commands in parallel using a "Single

Program, Multiple Data" (SPMD) statement
Unlike a parallel for-loop (parfor), which takes the iterations of a loop and distributes them among multiple
threads, a single program, multiple data (spmd) statement takes a series of commands and distributes them to all
the threads, so that each thread performs the command and stores the results. Consider this:
poolobj = parpool(2); % open a parallel pool with two workers
spmd
q = rand(3); % each thread generates a unique 3x3 array of random numbers
end
q{1} % q is called like a cell vector

q{2} % the values stored in each thread may be accessed by their index
delete(poolobj) % if the pool is closed, then the data in q will no longer be accessible
It is important to note that each thread may be accessed during the spmd block by its thread index (also called lab
index, or labindex):
poolobj = parpool(2); % open a parallel pool with two workers
spmd

q = rand(labindex + 1); % each thread generates a unique array of random numbers
end
size(q{1}) % the size of q{1} is 2x2

size(q{2}) % the size of q{2} is 3x3
delete(poolobj) % q is no longer accessible
In both examples, q is a composite object, which may be initialized with the command q = Composite(). It is
important to note that composite objects are only accessible while the pool is running.
Section 27.3: Using the batch command to do various

computations in parallel
To use multi-threading in MATLAB one can use the batch command. Note that you must have the Parallel
Computing toolbox installed.
For a time-consuming script, for example,
for ii=1:1e8
A(ii)=sin(ii*2*pi/1e8);
end
to run it in batch mode one would use the following:
job=batch("da")
which is enables MATLAB to run in batch mode and makes it possible to use MATLAB in the meantime to do other
things, such as add more batch processes.
To retrieve the results after finishing the job and load the array A into the workspace:
load(job, 'A')
Finally, open the "monitor job gui" from Home ? Environment ? Parallel ? Monitor jobs and delete the job through:
delete(job)
To load a function for batch processing, simply use this statement where fcn is the function name, N is number of
output arrays and x1, ..., xn are input arrays:
j=batch(fcn, N, {x1, x2, ..., xn})
Section 27.4: When to use parfor

Basically, parfor is recommended in two cases: lots of iterations in your loop (i.e., like 1e10), or if each iteration
takes a very long time (e.g., eig(magic(1e4))). In the second case you might want to consider using spmd . The
reason parfor is slower than a for loop for short ranges or fast iterations is the overhead needed to manage all
workers correctly, as opposed to just doing the calculation.
Also a lot of functions have implicit multi-threading built-in, making a parfor loop not more efficient, when using
these functions, than a serial for loop, since all cores are already being used. parfor will actually be a detriment in
this case, since it has the allocation overhead, whilst being as parallel as the function you are trying to use.

Consider the following example to see the behaviour of for as opposed to that of parfor. First open the parallel
pool if you've not already done so:
gcp; % Opens a parallel pool using your current settings
Then execute a couple of large loops:
n = 1000; % Iteration number

EigenValues = cell(n,1); % Prepare to store the data
Time = zeros(n,1);
for ii = 1:n
tic
EigenValues{ii,1} = eig(magic(1e3)); % Might want to lower the magic if it takes too long
Time(ii,1) = toc; % Collect time after each iteration
end
figure; % Create a plot of results

plot(1:n,t)
title 'Time per iteration'
ylabel 'Time [s]'
xlabel 'Iteration number[-]';
Then do the same with parfor instead of for. You will notice that the average time per iteration goes up. Do realise
however that the parfor used all available workers, thus the total time (sum(Time)) has to be divided by the
number of cores in your computer.
So, whilst the time to do each separate iteration goes up using parfor with respect to using for, the total time goes
down considerably.

Chapter 28: Using serial ports
Serial port parameter what it does
Sets the baudrate. The most common today is 57600, but 4800, 9600, and 115200 are
BaudRate
frequently seen as well
The number of bytes kept in memory. Matlab has a FIFO, which means that new bytes
InputBufferSize will be discarded. The default is 512 bytes, but it can easily be set to 20MB without issue.
There are only a few edge cases where the user would want this to be small
BytesAvailable The number of bytes waiting to be read
ValuesSent The number of bytes sent since the port was opened
ValuesReceived The number of bytes read since the port was opened
Specify the callback function to execute when a specified number of bytes is available in
BytesAvailableFcn
the input buffer, or a terminator is read
Specify the number of bytes that must be available in the input buffer to generate a
BytesAvailableFcnCount
bytes-available event
Specify if the bytes-available event is generated after a specified number of bytes is
BytesAvailableFcnMode
available in the input buffer, or after a terminator is read
Serial ports are a common interface for communicating with external sensors or embedded systems such as
Arduinos. Modern serial communications are often implemented over USB connections using USB-serial adapters.
MATLAB provides built-in functions for serial communications, including RS-232 and RS-485 protocols. These
functions can be used for hardware serial ports or "virtual" USB-serial connections. The examples here illustrate
serial communications in MATLAB.
Section 28.1: Creating a serial port on Mac/Linux/Windows

% Define serial port with a baud rate of 115200
rate = 115200;
if ispc
s = serial('COM1', 'BaudRate',rate);
elseif ismac
% Note that on OSX the serial device is uniquely enumerated. You will
% have to look at /dev/tty.* to discover the exact signature of your
% serial device
s = serial('/dev/tty.usbserial-A104VFT7', 'BaudRate',rate);
elseif isunix
s = serial('/dev/ttyusb0', 'BaudRate',rate);
end
% Set the input buffer size to 1,000,000 bytes (default: 512 bytes).
s.InputBufferSize = 1000000;
% Open serial port

fopen(s);
Section 28.2: Chosing your communication mode

Matlab supports synchronous and asynchronous communication with a serial port. It is important to chose the right
communication mode. The choice will depend on:
how the instrument you are communicating with behave.

what other functions your main program (or GUI) will have to do aside from managing the serial port.
I'll define 3 different cases to illustrate, from the simplest to the most demanding. For the 3 examples, the
instrument I am connecting to is a circuit board with an inclinometer, which can work in the 3 modes I will be
describing below.

Mode 1: Synchronous (Master/Slave)
This mode is the simplest one. It correspond to the case where the PC is the Master and the instrument is the slave.
The instrument does not send anything to the serial port on it's own, it only replies an answer after being asked a
question/command by the Master (the PC, your program). For example:
The PC sends a command: "Give me a measurement now"

The instrument receive the command, take the measurement then send back the measurement value to the
serial line: "The inclinometer value is XXX".
OR
The PC sends a command: "Change from mode X to mode Y"

The instrument receive the command, execute it, then send a confirmation message back to the serial line:
"Command executed" (or "Command NOT executed"). This is commonly called an ACK/NACK reply (for
"Acknowledge(d)" / "NOT Acknowledged").
Summary: in this mode, the instrument (the Slave) only send data to the serial line immediately after having been
asked by the PC (the Master)
Mode 2: Asynchronous
Now suppose I started my instrument, but it is more than just a dumb sensor. It constantly monitor it's own
inclination and as long as it is vertical (within a tolerance, let's say +/-15 degrees), it stays silent. If the device is tilted
by more than 15 degrees and get close to horizontal, it sends an alarm message to the serial line, immediately
followed by a reading of the inclination. As long as the inclination is above the threshold, it continues to send an
inclination reading every 5s.
If your main program (or GUI) is constantly "waiting" for message arriving on the serial line, it can do that well ... but
it cannot do anything else in the meantime. If the main program is a GUI, it is highly frustrating to have a GUI
seemingly "frozen" because it won't accept any input from the user. Essentially, it became the Slave and the
instrument is the Master. Unless you have a fancy way of controlling your GUI from the instrument, this is
something to avoid. Fortunately, the asynchronous communication mode will let you:
define a separate function which tells your program what to do when a message is received
keep this function in a corner, it will only be called and executed when a message arrives on the serial line.
The rest of the time the GUI can execute any other code it has to run.
Summary: In this mode, the instrument may send message to the serial line at anytime (but not necessarily all the
time). The PC does not wait permanently for a message to process. It is allowed to run any other code. Only when a

message arrives, it activates a function which will then read and process this message.
Mode 3: Streaming (Real time)
Now let's unleash the full power of my instrument. I put it in a mode where it will constantly send measurements to
the serial line. My program want to receive these packets and display that on a curve or a digital display. If it only
send a value every 5s as above, no problem, keep the above mode. But my instrument at full whack sends a data
point to the serial line at 1000Hz, i.e. it sends a new value every single millisecond. If I stay in the asynchronous mode
described above, there is a high risk (actually a guaranteed certainty) that the special function we defined to
process every new packet will take more than 1ms to execute (if you want to plot or display the value, graphic
functions are quite slow, not even considering filtering or FFT'ing the signal). It means the function will start to
execute, but before it finishes, a new packet will arrive and trigger the function again. The second function is placed
in a queue for execution, and will only starts when the first one is done ... but by this time a few new packets
arrived and each placed a function to execute in the queue. You can quickly foresee the result: By the time I am
plotting the 5th points, I have already hundreds waiting to be plotted too ... the gui slows down, eventually freezes,
the stack grows, the buffers fill up, until something gives. Eventually you are left with a completely frozen program
or simply a crashed one.
To overcome this, we will disconnect even further the synchronisation link between the PC and the instrument. We
will let the instrument send data at it's own pace, without immediately triggering a function at each packet arrival.
The serial port buffer will just accumulate the packets received. The PC will only collect data in the buffer at a pace it
can manage (a regular interval, set up on the PC side), do something with it (while the buffer is getting refilled by
the instrument), then collect a new batch of data from the buffer ... and so on.
Summary: In this mode, the instrument sends data continuously, which are collected by the serial port buffer. At
regular interval, the PC collect data from the buffer and do something with it. There is no hard synchronisation link
between the PC and the instrument. Both execute their tasks on their own timing.

Section 28.3: Automatically processing data received from a
serial port
Some devices connected through a serial port send data to your program at a constant rate (streaming data) or
send data at unpredictable intervals. You can configure the serial port to execute a function automatically to handle
data whenever it arrives. This is called the "callback function" for the serial port object.
There are two properties of the serial port that must be set to use this feature: the name of the function you want
for the callback (BytesAvailableFcn), and the condition which should trigger executing the callback function
(BytesAvailableFcnMode).
There are two ways to trigger a callback function:
1. When a certain number of bytes have been received at the serial port (typically used for binary data)
2. When a certain character is received at the serial port (typically used for text or ASCII data)
Callback functions have two required input arguments, called obj and event. obj is the serial port. For example, if
you want to print the data received from the serial port, define a function for printing the data called newdata:
function newdata(obj,event)
[d,c] = fread(obj); % get the data from the serial port
% Note: for ASCII data, use fscanf(obj) to return characters instead of binary values
fprintf(1,'Received %d bytes\n',c);
disp(d)
end
For example, to execute the newdata function whenever 64 bytes of data are received, configure the serial port like
this:
s = serial(port_name);
s.BytesAvailableFcnMode = 'byte';
s.BytesAvailableFcnCount = 64;
s.BytesAvailableFcn = @newdata;
With text or ASCII data, the data is typically divided into lines with a "terminator character", just like text on a page.
To execute the newdata function whenever the carriage return character is received, configure the serial port like
this:
s = serial(port_name);
s.BytesAvailableFcnMode = 'terminator';
s.Terminator = 'CR'; % the carriage return, ASCII code 13

s.BytesAvailableFcn = @newdata;
Section 28.4: Reading from the serial port

Assuming you created the serial port object s as in this example, then
% Read one byte

data = fread(s, 1);
% Read all the bytes, version 1

data = fread(s);
% Read all the bytes, version 2

data = fread(s, s.BytesAvailable);
% Close the serial port

fclose(s);
Section 28.5: Closing a serial port even if lost, deleted or

overwritten
Assuming you created the serial port object s as in this example, then to close it
fclose(s)
However, sometimes you can accidentally lose the port (e.g. clear, overwrite, change scope, etc...), and fclose(s)
will no longer work. The solution is easy
fclose(instrfindall)
More info at instrfindall().
Section 28.6: Writing to the serial port

Assuming you created the serial port object s as in this example, then
% Write one byte

fwrite(s, 255);
% Write one 16-bit signed integer

fwrite(s, 32767, 'int16');
% Write an array of unsigned 8-bit integers

fwrite(s,[48 49 50],'uchar');
% Close the serial port

fclose(s);

Chapter 29: Undocumented Features
Section 29.1: Color-coded 2D line plots with color data in third
dimension
In MATLAB versions prior to R2014b, using the old HG1 graphics engine, it was not obvious how to create color
coded 2D line plots. With the release of the new HG2 graphics engine arose a new undocumented feature
introduced by Yair Altman:
n = 100;
x = linspace(-10,10,n); y = x.^2;
p = plot(x,y,'r', 'LineWidth',5);
% modified jet-colormap
cd = [uint8(jet(n)*255) uint8(ones(n,1))].';
drawnow
set(p.Edge, 'ColorBinding','interpolated', 'ColorData',cd)
Section 29.2: Semi-transparent markers in line and scatter

plots
Since Matlab R2014b it is easily possible to achieve semi-transparent markers for line and scatter plots using
undocumented features introduced by Yair Altman.
The basic idea is to get the hidden handle of the markers and apply a value < 1 for the last value in the
EdgeColorData to achieve the desired transparency.
Here we go for scatter:
%// example data

x = linspace(0,3*pi,200);

y = cos(x) + rand(1,200);
%// plot scatter, get handle

h = scatter(x,y);
drawnow; %// important
%// get marker handle

hMarkers = h.MarkerHandle;
%// get current edge and face color

edgeColor = hMarkers.EdgeColorData
faceColor = hMarkers.FaceColorData
%// set face color to the same as edge color

faceColor = edgeColor;
%// opacity
opa = 0.3;
%// set marker edge and face color

hMarkers.EdgeColorData = uint8( [edgeColor(1:3); 255*opa] );
hMarkers.FaceColorData = uint8( [faceColor(1:3); 255*opa] );
and for a line plot
%// example data

x = linspace(0,3*pi,200);
y1 = cos(x);
y2 = sin(x);
%// plot scatter, get handle

h1 = plot(x,y1,'o-','MarkerSize',15); hold on
h2 = plot(x,y2,'o-','MarkerSize',15);

drawnow; %// important
%// get marker handle

h1Markers = h1.MarkerHandle;
h2Markers = h2.MarkerHandle;
%// get current edge and face color

edgeColor1 = h1Markers.EdgeColorData;
edgeColor2 = h2Markers.EdgeColorData;
%// set face color to the same as edge color

faceColor1 = edgeColor1;
faceColor2 = edgeColor2;
%// opacity
opa = 0.3;
%// set marker edge and face color

h1Markers.EdgeColorData = uint8( [edgeColor1(1:3); 255*opa] );
h1Markers.FaceColorData = uint8( [faceColor1(1:3); 255*opa] );
h2Markers.EdgeColorData = uint8( [edgeColor2(1:3); 255*opa] );
h2Markers.FaceColorData = uint8( [faceColor2(1:3); 255*opa] );
The marker handles, which are used for the manipulation, are created with the figure. The drawnow command is
ensuring the creation of the figure before subsequent commands are called and avoids errors in case of delays.
Section 29.3: C++ compatible helper functions

The use of Matlab Coder sometimes denies the use of some very common functions, if they are not compatible to
C++. Relatively often there exist undocumented helper functions, which can be used as replacements.
Here is a comprehensive list of supported functions..
And following a collection of alternatives, for non-supported functions:

The functions sprintf and sprintfc are quite similar, the former returns a character array, the latter a cell string:
str = sprintf('%i',x) % returns '5' for x = 5

str = sprintfc('%i',x) % returns {'5'} for x = 5
However, sprintfc is compatible with C++ supported by Matlab Coder, and sprintf is not.
Section 29.4: Scatter plot jitter

The scatter function has two undocumented properties 'jitter' and 'jitterAmount' that allow to jitter the data
on the x-axis only. This dates back to Matlab 7.1 (2005), and possibly earlier.
To enable this feature set the 'jitter' property to 'on' and set the 'jitterAmount' property to the desired
absolute value (the default is 0.2).
This is very useful when we want to visualize overlapping data, for example:
scatter(ones(1,10), ones(1,10), 'jitter', 'on', 'jitterAmount', 0.5);
Read more on Undocumented Matlab
Section 29.5: Contour Plots - Customise the Text Labels

When displaying labels on contours Matlab doesn't allow you to control the format of the numbers, for example to
change to scientific notation.
The individual text objects are normal text objects but how you get them is undocumented. You access them from

the TextPrims property of the contour handle.
figure
[X,Y]=meshgrid(0:100,0:100);
Z=(X+Y.^2)*1e10;
[C,h]=contour(X,Y,Z);
h.ShowText='on';
drawnow();
txt = get(h,'TextPrims');
v = str2double(get(txt,'String'));
for ii=1:length(v)
set(txt(ii),'String',sprintf('%0.3e',v(ii)))
end
Note: that you must add a drawnow command to force Matlab to draw the contours, the number and location of the
txt objects are only determined when the contours are actually drawn so the text objects are only created then.
The fact the txt objects are created when the contours are drawn means that they are recalculated every time the
plot is redrawn (for example figure resize). To manage this you need to listen to the undocumented event
MarkedClean:
function customiseContour
figure
[X,Y]=meshgrid(0:100,0:100);
Z=(X+Y.^2)*1e10;
[C,h]=contour(X,Y,Z);
h.ShowText='on';
% add a listener and call your new format function
addlistener(h,'MarkedClean',@(a,b)ReFormatText(a))
end
function ReFormatText(h)
% get all the text items from the contour
t = get(h,'TextPrims');
for ii=1:length(t)
% get the current value (Matlab changes this back when it
% redraws the plot)
v = str2double(get(t(ii),'String'));
% Update with the format you want - scientific for example
set(t(ii),'String',sprintf('%0.3e',v));
end
end

Example tested using Matlab r2015b on Windows
Section 29.6: Appending / adding entries to an existing legend

Existing legends can be difficult to manage. For example, if your plot has two lines, but only one of them has a
legend entry and that should stay this way, then adding a third line with a legend entry can be difficult. Example:
figure hold on fplot(@sin) fplot(@cos) legend sin % Add only a legend entry for sin hTan = fplot(@tan); % Make sure
to get the handle, hTan, to the graphics object you want to add to the legend
Now, to add a legend entry for tan, but not for cos, any of the following lines won't do the trick; they all fail in some
way:
legend tangent % Replaces current legend -> fail legend -DynamicLegend % Undocumented feature, adds 'cos',
which shouldn't be added -> fail legend sine tangent % Sets cos DisplayName to 'tangent' -> fail legend sine ''
tangent % Sets cos DisplayName, albeit empty -> fail legend(f)
Luckily, an undocumented legend property called PlotChildren keeps track of the children of the parent figure1.
So, the way to go is to explicitly set the legend's children through its PlotChildren property as follows:
hTan.DisplayName = 'tangent'; % Set the graphics object's display name l = legend; l.PlotChildren(end + 1) = hTan; %
Append the graphics handle to legend's plot children
The legend updates automatically if an object is added or removed from its PlotChildren property.
1 Indeed: figure. You can add any figure's child with the DisplayName property to any legend in the figure, e.g. from
a different subplot. This is because a legend in itself is basically an axes object.
Tested on MATLAB R2016b

Chapter 30: MATLAB Best Practices
Section 30.1: Indent code properly
Proper indentation gives not only the aesthetic look but also increases the readability of the code.
For example, consider the following code:
%no need to understand the code, just give it a look

n = 2;
bf = false;
while n>1
for ii = 1:n
for jj = 1:n
if ii+jj>30
bf = true;
break
end
end
if bf
break
end
end
if bf
break
end
n = n + 1;
end
As you can see, you need to give a careful look to see which loop and if statements are ending where.
With smart indentation, you'll get this look:
n = 2;
bf = false;
while n>1
for ii = 1:n
for jj = 1:n
if ii+jj>30
bf = true;
break
end
end
if bf
break
end
end
if bf
break
end
n = n + 1;
end
This clearly indicates the starting and ending of loops/if statement.
You can do smart indentation by:

• selecting all your code ( Ctrl + A )
• and then pressing Ctrl + I or clicking from edit bar.

Section 30.2: Avoid loops
Most of the time, loops are computationally expensive with Matlab. Your code will be orders of magnitudes faster if
you use vectorization. It also often makes your code more modular, easily modifiable, and easier to debug. The
major downside is that you have to take time to plan the data structures, and dimension errors are easier to come
by.
Examples
Don't write
for t=0:0.1:2*pi
R(end+1)=cos(t);
end
but
t=0:0.1:2*pi;
R=cos(t)
Don't write
for i=1:n
for j=1:m
c(i,j)=a(i)+2*b(j);
end
end
But something similar to
c=repmat(a.',1,m)+2*repmat(b,n,1)
For more details, see vectorization
Section 30.3: Keep lines short

Use the continuation character (ellipsis) ... to continue long statement.
Example:
MyFunc( parameter1,parameter2,parameter3,parameter4, parameter5, parameter6,parameter7, parameter8,

parameter9)
can be replaced by:
MyFunc( parameter1, ...

parameter2, ...
parameter3, ...

parameter4, ...
parameter5, ...
parameter6, ...
parameter7, ...
parameter8, ...
parameter9)
Section 30.4: Use assert

Matlab allows some very trivial mistakes to go by silently, which might cause an error to be raised much later in the
run - making debugging hard. If you assume something about your variables, validate it.
function out1 = get_cell_value_at_index(scalar1,cell2)

assert(isscalar(scalar1),'1st input must be a scalar')
assert(iscell(cell2),'2nd input must be a cell array')
assert(numel(cell2) >= scalar1),'2nd input must have more elements than the value of the 1st
input')
assert(~isempty(cell2{scalar1}),'2nd input at location is empty')
out1 = cell2{scalar1};
Section 30.5: Block Comment Operator

It is a good practice to add comments that describe the code. It is helpful for others and even for the coder when
returned later. A single line can be commented using the % symbol or using the shortkey Ctrl+R. To uncomment a
previously commented line remove the % symbol or use shortkey Crtl+T.
While commenting a block of code can be done by adding a % symbol at the beginning of each line, newer versions
of MATLAB (after 2015a) let you use the Block Comment Operator %{ code %}. This operator increases the
readability of the code. It can be used for both code commenting and function help documentation. The Block can
be folded and unfolded to increase the readability of the code.
As it can be seen the %{ and %} operators must appear alone on the lines. Do not include any other text on these
lines.

function y = myFunction(x)
%{
myFunction Binary Singleton Expansion Function
y = myFunction(x) applies the element-by-element binary operation
specified by the function handle FUNC to arrays A and B, with implicit
expansion enabled.
%}
%% Compute z(x, y) = x.*sin(y) on a grid:

% x = 1:10;
y = x.';
%{
z = zeros(numel(x),numel(y));
for ii=1:numel(x)
for jj=1:numel(y)
z(ii,jj) = x(ii)*sin(y(jj));
end
end
%}
z = bsxfun(@(x, y) x.*sin(y), x, y);

y = y + z;
end
Section 30.6: Create Unique Name for Temporary File

While coding a script or a function, it can be the case that one or more than one temporary file be needed in order
to, for example, store some data.
In order to avoid overwriting an existing file or to shadow a MATLAB function the tempname function can be used
to generate a unique name for a temporary file in the system temporary folder.
my_temp_file=tempname
The filename is generated without the extension; it can be added by concatenating the desired extension to the
name generated by tempname
my_temp_file_with_ext=[tempname '.txt']
The locaton of the system temporary folder can be retrieved by caling the tempdir function.
If, during the execution of the function / script, the temporary file is no longer needed, it can be deleted by using
the function delete
Since delete does not ask for confirmation, it might be useful to set on the option to move the file to be deleted in
the recycle folder.
This can be done by using the function recycle this way:
recycle('on')
In the following example, a possible usage of the functions tempname, delete and recycle is proposed.
%
% Create some example data

%
theta=0:.1:2*pi;
x=cos(theta);
y=sin(theta);
%
% Generate the temporary filename
%
my_temp_file=[tempname '.mat'];
%
% Split the filename (path, name, extension) and display them in a message box
[tmp_file_path,tmp_file_name, tmp_file_ext]=fileparts(my_temp_file)
uiwait(msgbox(sprintf('Path= %s\nName= %s\nExt= %s', ...
tmp_file_path,tmp_file_name,tmp_file_ext),'TEMPORARY FILE'))
%
% Save the varaibles in a temporary file
%
save(my_temp_file,'x','y','theta')
%
% Load the varaibles from the temporary file
%
load(my_temp_file)
%
% Set the reclycle option on
%
recycle('on')
%
% Delete the temporary file
%
delete(my_temp_file)
Caveat
The temporary filename is generated by using the java.util.UUID.randomUUID method (randomUUID).
If MATLAB is run without JVM, the temporary filename is generated by using

matlab.internal.timing.timing based on the CPU counter and time. In this case the temporary filename is not
guaranteed to be unique.

Chapter 31: MATLAB User Interfaces
Section 31.1: Passing Data Around User Interface
Most advanced user interfaces require the user to be able to pass information between the various functions which
make up a user interface. MATLAB has a number of different methods to do so.
guidata
MATLAB's own GUI Development Environment (GUIDE) prefers to use a struct named handles to pass data
between callbacks. This struct contains all of the graphics handles to the various UI components as well as user-
specified data. If you aren't using a GUIDE-created callback which automatically passes handles, you can retrieve
the current value using guidata
% hObject is a graphics handle to any UI component in your GUI

handles = guidata(hObject);
If you want to modify a value stored in this data structure, you can modify but then you must store it back within
the hObject for the changes to be visible by other callbacks. You can store it by specifying a second input argument
to guidata.
% Update the value

handles.myValue = 2;
% Save changes
guidata(hObject, handles)
The value of hObject doesn't matter as long as it is a UI component within the same figure because ultimately the
data is stored within the figure containing hObject.
Best for:
Storing the handles structure, in which you can store all the handles of your GUI components.
Storing "small" other variables which need to be accessed by most callbacks.
Not recommended for:
Storing large variables which do not have to be accessed by all callbacks and sub-functions (use
setappdata/getappdata for these).
setappdata/getappdata
Similar to the guidata approach, you can use setappdata and getappdata to store and retrieve values from within a
graphics handle. The advantage of using these methods is that you can retrieve only the value you want rather than
an entire struct containing all stored data. It is similar to a key/value store.
To store data within a graphics object
% Create some data you would like to store

myvalue = 2
% Store it using the key 'mykey'

setappdata(hObject, 'mykey', myvalue)
And to retrieve that same value from within a different callback

value = getappdata(hObject, 'mykey');
Note: If no value was stored prior to calling getappdata, it will return an empty array ([]).
Similar to guidata, the data is stored in the figure that contains hObject.
Best for:
Storing large variables which do not have to be accessed by all callbacks and sub-functions.
UserData
Every graphics handle has a special property, UserData which can contain any data you wish. It could contain a cell
array, a struct, or even a scalar. You can take advantage of this property and store any data you wish to be
associated with a given graphics handle in this field. You can save and retrieve the value using the standard get/set
methods for graphics objects or dot notation if you're using R2014b or newer.
% Create some data to store

mydata = {1, 2, 3};
% Store it within the UserData property

set(hObject, 'UserData', mydata)
% Of if you're using R2014b or newer:

% hObject.UserData = mydata;
Then from within another callback, you can retrieve this data:
their_data = get(hObject, 'UserData');
% Or if you're using R2014b or newer:

% their_data = hObject.UserData;
Best for:
Storing variables with a limited scope (variables which are likely to be used only by the object in which they
are stored, or objects having a direct relationship to it).
Nested Functions
In MATLAB, a nested function can read and modify any variable defined in the parent function. In this way, if you
specify a callback to be a nested function, it can retrieve and modify any data stored in the main function.
function mygui()
hButton = uicontrol('String', 'Click Me', 'Callback', @callback);
% Create a counter to keep track of the number of times the button is clicked
nClicks = 0;
% Callback function is nested and can therefore read and modify nClicks
function callback(source, event)
% Increment the number of clicks
nClicks = nClicks + 1;
% Print the number of clicks so far

fprintf('Number of clicks: %d\n', nClicks);
end
end

Best for:
Small, simple GUIs. (for quick prototyping, to not have to implement the guidata and/or set/getappdata
methods).
Not recommended for:
Medium, large or complex GUIs.
GUI created with GUIDE.
Explicit input arguments
If you need to send data to a callback function and don't need to modify the data within the callback, you can
always consider passing the data to the callback using a carefully crafted callback definition.
You could use an anonymous function which adds inputs
% Create some data to send to mycallback

data = [1, 2, 3];
% Pass data as a third input to mycallback

set(hObject, 'Callback', @(source, event)mycallback(source, event, data))
Or you could use the cell array syntax to specify a callback, again specifying additional inputs.
set(hObject, 'Callback', {@mycallback, data})
Best for:
When the callback needs data to perform some operations but the data variable does not need to be
modified and saved in a new state.
Section 31.2: Making a button in your UI that pauses callback

execution
Sometimes we'd like to pause code execution to inspect the state of the application (see Debugging). When running
code through the MATLAB editor, this can be done using the "Pause" button in the UI or by pressing Ctrl + c
(on Windows). However, when a computation was initiated from a GUI (via the callback of some uicontrol), this
method does not work anymore, and the callback should be interrupted via another callback. Below is a
demonstration of this principle:
function interruptibleUI
dbclear in interruptibleUI % reset breakpoints in this file
figure('Position',[400,500,329,160]);
uicontrol('Style', 'pushbutton',...
'String', 'Compute',...
'Position', [24 55 131 63],...
'Callback', @longComputation,...
'Interruptible','on'); % 'on' by default anyway
'String', 'Pause #1',...
'Position', [180 87 131 63],...
'Callback', @interrupt1);

'String', 'Pause #2',...
'Position', [180 12 131 63],...
'Callback', @interrupt2);
end
function longComputation(src,event)
superSecretVar = rand(1);
pause(15);
print('done!'); % we'll use this to determine if code kept running "in the background".
end
function interrupt1(src,event) % depending on where you want to stop

dbstop in interruptibleUI at 27 % will stop after print('done!');
dbstop in interruptibleUI at 32 % will stop after **this** line.
end
function interrupt2(src,event) % method 2

keyboard;
dbup; % this will need to be executed manually once the code stops on the previous line.
end
To make sure you understand this example do the following:
1. Paste the above code into a new file called and save it as interruptibleUI.m, such that the code starts on
the very first line of the file (this is important for the 1st method to work).
2. Run the script.
3. Click Compute and shortly afterwards click either Pause #1 or on Pause #2 .
4. Make sure you can find the value of superSecretVar.
Section 31.3: Passing data around using the "handles"

structure
This is an example of a basic GUI with two buttons that change a value stored in the GUI's handles structure.
function gui_passing_data()
% A basic GUI with two buttons to show a simple use of the 'handles'
% structure in GUI building
% Create a new figure.

f = figure();
% Retrieve the handles structure

handles = guidata(f);
% Store the figure handle

handles.figure = f;
% Create an edit box and two buttons (plus and minus),

% and store their handles for future use
handles.hedit = uicontrol('Style','edit','Position',[10,200,60,20] , 'Enable', 'Inactive');
handles.hbutton_plus = uicontrol('Style','pushbutton','String','+',...
'Position',[80,200,60,20] , 'Callback' , @ButtonPress);
handles.hbutton_minus = uicontrol('Style','pushbutton','String','-',...
'Position',[150,200,60,20] , 'Callback' , @ButtonPress);

% Define an initial value, store it in the handles structure and show
% it in the Edit box
handles.value = 1;
set(handles.hedit , 'String' , num2str(handles.value))
% Store handles
guidata(f, handles);
function ButtonPress(hObject, eventdata)

% A button was pressed
% Retrieve the handles
handles = guidata(hObject);
% Determine which button was pressed; hObject is the calling object

switch(get(hObject , 'String'))
case '+'
% Add 1 to the value
handles.value = handles.value + 1;
set(handles.hedit , 'String', num2str(handles.value))
case '-'
% Substract 1 from the value
handles.value = handles.value - 1;
end
% Display the new value

set(handles.hedit , 'String', num2str(handles.value))
% Store handles
guidata(hObject, handles);
To test the example, save it in a file called gui_passing_data.m and launch it with F5. Please note that in such a
simple case, you would not even need to store the value in the handles structure because you could directly access
it from the edit box's String property.
Section 31.4: Performance Issues when Passing Data Around

User Interface
Two main techniques allow passing data between GUI functions and Callbacks: setappdata/getappdata and guidata
(read more about it). The former should be used for larger variables as it is more time efficient. The following
example tests the two methods' efficiency.
A GUI with a simple button is created and a large variable (10000x10000 double) is stored both with guidata and
with setappdata. The button reloads and stores back the variable using the two methods while timing their
execution. The running time and percentage improvement using setappdata are displayed in the command
window.
function gui_passing_data_performance()
% A basic GUI with a button to show performance difference between
% guidata and setappdata
% Create a new figure.

f = figure('Units' , 'normalized');
% Retrieve the handles structure

handles = guidata(f);
% Store the figure handle

handles.figure = f;
handles.hbutton = uicontrol('Style','pushbutton','String','Calculate','units','normalized',...
'Position',[0.4 , 0.45 , 0.2 , 0.1] , 'Callback' , @ButtonPress);
% Create an uninteresting large array

data = zeros(10000);
% Store it in appdata
setappdata(handles.figure , 'data' , data);
% Store it in handles
handles.data = data;
% Save handles
guidata(f, handles);
function ButtonPress(hObject, eventdata)
% Calculate the time difference when using guidata and appdata

t_handles = timeit(@use_handles);
t_appdata = timeit(@use_appdata);
% Absolute and percentage difference

t_diff = t_handles - t_appdata;
t_perc = round(t_diff / t_handles * 100);
disp(['Difference: ' num2str(t_diff) ' ms / ' num2str(t_perc) ' %'])
function use_appdata()
% Retrieve the data from appdata

data = getappdata(gcf , 'data');
% Do something with data %
% Store the value again

setappdata(gcf , 'data' , data);
function use_handles()
% Retrieve the data from handles

handles = guidata(gcf);
data = handles.data;
% Do something with data %
% Store it back in the handles

handles.data = data;
guidata(gcf, handles);
On my Xeon [email protected] GHz I get Difference: 0.00018957 ms / 73 %, thus using getappdata/setappdata I get a
performance improvement of 73%! Note that the result does not change if a 10x10 double variable is used,
however, result will change if handles contains many fields with large data.

Chapter 32: Useful tricks
Section 32.1: Extract figure data
On a few occasions, I have had an interesting figure I saved but I lost an access to its data. This example shows a
trick how to achieve extract information from a figure.
The key functions are findobj and get. findobj returns a handler to an object given attributes or properties of the
object, such as Type or Color, etc. Once a line object has been found, get can return any value held by properties. It
turns out that the Line objects hold all data in following properties: XData, YData, and ZData; the last one is usually
0 unless a figure contains a 3D plot.
The following code creates an example figure that shows two lines a sin function and a threshold and a legend
t = (0:1/10:1-1/10)';
y = sin(2*pi*t);
plot(t,y);
hold on;
plot([0 0.9],[0 0], 'k-');
hold off;
legend({'sin' 'threshold'});
The first use of findobj returns two handlers to both lines:
findobj(gcf, 'Type', 'Line')

ans =
2x1 Line array:
Line (threshold)
Line (sin)
To narrow the result, findobj can also use combination of logical operators -and, -or and property names. For
instance, I can find a line object whose DiplayName is sin and read its XData and YData.
lineh = findobj(gcf, 'Type', 'Line', '-and', 'DisplayName', 'sin');

xdata = get(lineh, 'XData');
ydata = get(lineh, 'YData');
and check if the data are equal.
isequal(t(:),xdata(:))
ans =
1
isequal(y(:),ydata(:))
ans =
1
Similarly, I can narrow my results by excluding the black line (threshold):
lineh = findobj(gcf, 'Type', 'Line', '-not', 'Color', 'k');

xdata = get(lineh, 'XData');
ydata = get(lineh, 'YData');
and last check confirms that data extracted from this figure are the same:

isequal(t(:),xdata(:))
ans =
1
isequal(y(:),ydata(:))
ans =
1
Section 32.2: Code Folding Preferences

It is possible to change Code Folding preference to suit your need. Thus code folding can be set enable/unable for
specific constructs (ex: if block, for loop, Sections ...).
To change folding preferences, go to Preferences -> Code Folding:
Then you can choose which part of the code can be folded.
Some information:
Note that you can also expand or collapse all of the code in a file by placing your cursor anywhere within the
file, right-click, and then select Code Folding > Expand All or Code Folding > Fold All from the context menu.
Note that folding is persistent, in the sense that part of the code that has been expanded/collapsed will keep
their status after Matlab or the m-file has been closed and is re-open.
Example: To enable folding for sections:
An interesting option is to enable to fold Sections. Sections are delimited by two percent signs (%%).
Example: To enable it check the "Sections" box:

Then instead of seeing a long source code similar to :
You will be able to fold sections to have a general overview of your code :

Section 32.3: Functional Programming using Anonymous
Functions
Anonymous functions can be used for functional programming. The main problem to solve is that there is no native
way for anchoring a recursion, but this can still be implemented in a single line:
if_ = @(bool, tf) tf{2-bool}();
This function accepts a boolean value and a cell array of two functions. The first of those functions is evaluated if
the boolean value evaluates as true, and the second one if the boolean value evaluates as false. We can easily write
the factorial function now:
fac = @(n,f) if_(n>1, {@()n*f(n-1,f), @()1});
The problem here is that we cannot directly invoke a recursive call, as the function is not yet assigned to a variable
when the right hand side is evaluated. We can however complete this step by writing
factorial_ = @(n)fac(n,fac);
Now @(n)fac(n,fac) evaulates the factorial function recursively. Another way to do this in functional programming
using a y-combinator, which also can easily be implemented:
y_ = @(f)@(n)f(n,f);
With this tool, the factorial function is even shorter:
factorial_ = y_(fac);
Or directly:
factorial_ = y_(@(n,f) if_(n>1, {@()n*f(n-1,f), @()1}));
Section 32.4: Save multiple figures to the same .fig file

By putting multiple figure handles into a graphics array, multiple figures can be saved to the same .fig file

h(1) = figure;
scatter(rand(1,100),rand(1,100));
h(2) = figure;
h(3) = figure;
savefig(h,'ThreeRandomScatterplots.fig');
close(h);
This creates 3 scatterplots of random data, each part of graphic array h. Then the graphics array can be saved using
savefig like with a normal figure, but with the handle to the graphics array as an additional argument.
An interesting side note is that the figures will tend to stay arranged in the same way that they were saved when
you open them.
Section 32.5: Comment blocks

If you want to comment part of your code, then comment blocks may be useful. Comment block starts with a %{ in
a new line and ends with %} in another new line:
a = 10;
b = 3;
%{
c = a*b;
d = a-b;
%}
This allows you fo fold the sections that you commented to make the code more clean and compact.
These blocks are also useful for toggling on/off parts of your code. All you have to do to uncomment the block is
add another % before it strats:
a = 10;
b = 3;
%%{ <-- another % over here
c = a*b;
d = a-b;
%}
Sometimes you want to comment out a section of the code, but without affecting its indentation:
for k = 1:a
b = b*k;
c = c-b;
d = d*c;
disp(b)
end
Usually, when you mark a block of code and press Ctrl+r for commenting it out (by that adding the % automatically
to all lines, then when you press later Ctrl+i for auto indentation, the block of code moves from its correct
hierarchical place, and moved too much to the right:
for k = 1:a
b = b*k;

% c = c-b;
% d = d*c;
disp(b)
end
A way to solve this is to use comment blocks, so the inner part of the block stays correctly indented:
for k = 1:a
b = b*k;
%{
c = c-b;
d = d*c;
%}
disp(b)
end
Section 32.6: Useful functions that operate on cells and

arrays
This simple example provides an explanation on some functions I found extremely useful since I have started using
MATLAB: cellfun, arrayfun. The idea is to take an array or cell class variable, loop through all its elements and
apply a dedicated function on each element. An applied function can either be anonymous, which is usually a case,
or any regular function define in a *.m file.
Let's start with a simple problem and say we need to find a list of *.mat files given the folder. For this example, first
let's create some *.mat files in a current folder:
for n=1:10; save(sprintf('mymatfile%d.mat',n)); end
After executing the code, there should be 10 new files with extension *.mat. If we run a command to list all *.mat
files, such as:
mydir = dir('*.mat');
we should get an array of elements of a dir structure; MATLAB should give a similar output to this one:
10x1 struct array with fields:

name
date
bytes
isdir
datenum
As you can see each element of this array is a structure with couple of fields. All information are indeed important
regarding each file but in 99% I am rather interested in file names and nothing else. To extract information from a
structure array, I used to create a local function that would involve creating temporal variables of a correct size, for
loops, extracting a name from each element, and save it to created variable. Much easier way to achieve exactly the
same result is to use one of the aforementioned functions:
mydirlist = arrayfun(@(x) x.name, dir('*.mat'), 'UniformOutput', false)

mydirlist =
'mymatfile1.mat'
'mymatfile10.mat'
'mymatfile2.mat'
'mymatfile3.mat'
'mymatfile4.mat'

'mymatfile5.mat'
'mymatfile6.mat'
'mymatfile7.mat'
'mymatfile8.mat'
'mymatfile9.mat'
How this function works? It usually takes two parameters: a function handle as the first parameter and an array. A
function will then operate on each element of a given array. The third and fourth parameters are optional but
important. If we know that an output will not be regular, it must be saved in cell. This must be point out setting
false to UniformOutput. By default this function attempts to return a regular output such as a vector of numbers.
For instance, let's extract information about how much of disc space is taken by each file in bytes:
mydirbytes = arrayfun(@(x) x.bytes, dir('*.mat'))

mydirbytes =
34560
34560
34560
34560
34560
34560
34560
34560
34560
34560
or kilobytes:
mydirbytes = arrayfun(@(x) x.bytes/1024, dir('*.mat'))

mydirbytes =
33.7500
33.7500
33.7500
33.7500
33.7500
33.7500
33.7500
33.7500
33.7500
33.7500
This time the output is a regular vector of double. UniformOutput was set to true by default.
cellfun is a similar function. The difference between this function and arrayfun is that cellfun operates on cell
class variables. If we wish to extract only names given a list of file names in a cell 'mydirlist', we would just need to
run this function as follows:
mydirnames = cellfun(@(x) x(1:end-4), mydirlist, 'UniformOutput', false)

mydirnames =
'mymatfile1'
'mymatfile10'
'mymatfile2'
'mymatfile3'
'mymatfile4'
'mymatfile5'
'mymatfile6'
'mymatfile7'
'mymatfile8'
'mymatfile9'

Again, as an output is not a regular vector of numbers, an output must be saved in a cell variable.
In the example below, I combine two functions in one and return only a list of file names without an extension:
cellfun(@(x) x(1:end-4), arrayfun(@(x) x.name, dir('*.mat'), 'UniformOutput', false),

'UniformOutput', false)
ans =
'mymatfile1'
'mymatfile10'
'mymatfile2'
'mymatfile3'
'mymatfile4'
'mymatfile5'
'mymatfile6'
'mymatfile7'
'mymatfile8'
'mymatfile9'
It is crazy but very possible because arrayfun returns a cell which is expected input of cellfun; a side note to this is
that we can force any of those functions to return results in a cell variable by setting UniformOutput to false,
explicitly. We can always get results in a cell. We may not be able to get results in a regular vector.
There is one more similar function that operates on fields a structure: structfun. I have not particularly found it as
useful as the other two but it would shine in some situations. If for instance one would like to know which fields are
numeric or non-numeric, the following code can give the answer:
structfun(@(x) ischar(x), mydir(1))
The first and the second field of a dir structure is of a char type. Therefore, the output is:
1
1
0
0
0
Also, the output is a logical vector of true / false. Consequently, it is regular and can be saved in a vector; no need
to use a cell class.

Chapter 33: Common mistakes and errors
Section 33.1: The transpose operators
.' is the correct way to transpose a vector or matrix in MATLAB.
' is the correct way to take the complex conjugate transpose (a.k.a. Hermitian conjugate) of a vector or
matrix in MATLAB.
Note that for the transpose .', there is a period in front of the apostrophe. This is in keeping with the syntax for
the other element-wise operations in MATLAB: * multiplies matrices, .* multiplies elements of matrices together. The
two commands are very similar, but conceptually very distinct. Like other MATLAB commands, these operators are
"syntactical sugar" that gets turned into a "proper" function call at runtime. Just as == becomes an evaluation of the
eq function, think of .' as the shorthand for transpose. If you would only write ' (without the point), you are in fact
using the ctranspose command instead, which calculates the complex conjugate transpose, which is also known as
the Hermitian conjugate, often used in physics. As long as the transposed vector or matrix is real-valued, the two
operators produce the same result. But as soon as we deal with complex numbers, we will inevitably run into
problems if we do not use the "correct" shorthand. What "correct" is depends on your application.
Consider the following example of a matrix C containing complex numbers:
>> C = [1i, 2; 3*1i, 4]

C =
0.0000 + 1.0000i 2.0000 + 0.0000i
0.0000 + 3.0000i 4.0000 + 0.0000i
Let's take the transpose using the shorthand .' (with the period). The output is as expected, the transposed form of
C.
>> C.'
ans =
0.0000 + 1.0000i 0.0000 + 3.0000i
2.0000 + 0.0000i 4.0000 + 0.0000i
Now, let's use ' (without the period). We see, that in addition to the transposition, the complex values have been
transformed to their complex conjugates as well.
>> C'
ans =
0.0000 - 1.0000i 0.0000 - 3.0000i
2.0000 + 0.0000i 4.0000 + 0.0000i
To sum up, if you intend to calculate the Hermitian conjugate, the complex conjugate transpose, then use '
(without the period). If you just want to calculate the transpose without complex-conjugating the values, use .'
(with the period).
Section 33.2: Do not name a variable with an existing function

name
There is already a function sum(). As a result, if we name a variable with the same name
sum = 1+3;
and if we try to use the function while the variable still exists in the workspace

A = rand(2);
sum(A,1)
we will get the cryptic error:
Subscript indices must either be real positive integers or logicals.
clear() the variable first and then use the function
clear sum
sum(A,1)
ans =
1.0826 1.0279
How can we check if a function already exists to avoid this conflict?
Use which() with the -all flag:
which sum -all

sum is a variable.
built-in (C:\Program Files\MATLAB\R2016a\toolbox\matlab\datafun\@double\sum) % Shadowed double
method
...
This output is telling us that sum is first a variable and that the following methods (functions) are shadowed by it, i.e.
MATLAB will first try to apply our syntax to the variable, rather than using the method.
Section 33.3: Be aware of floating point inaccuracy

Floating-point numbers cannot represent all real numbers. This is known as floating point inaccuracy.
There are infinitely many floating points numbers and they can be infinitely long (e.g. ?), thus being able to
represent them perfectly would require infinitely amount of memory. Seeing this was a problem, a special
representation for "real number" storage in computer was designed, the IEEE 754 standard. In short, it describes
how computers store this type of numbers, with an exponent and mantissa, as,
floatnum = sign * 2êxponent * mantissa
With limited amount of bits for each of these, only a finite precision can be achieved. The smaller the number,
smaller the gap between possible numbers (and vice versa!). You can try your real numbers in this online demo.
Be aware of this behavior and try to avoid all floating points comparison and their use as stopping conditions in
loops. See below two examples:
Examples: Floating point comparison done WRONG:

>> 0.1 + 0.1 + 0.1 == 0.3
ans =
logical
It is poor practice to use floating point comparison as shown by the precedent example. You can overcome it by
taking the absolute value of their difference and comparing it to a (small) tolerance level.

Below is another example, where a floating point number is used as a stopping condition in a while loop:**
k = 0.1;
while k <= 0.3
disp(num2str(k));
k = k + 0.1;
end
% --- Output: ---

0.1
0.2
It misses the last expected loop (0.3 <= 0.3).
Example: Floating point comparison done RIGHT:

x = 0.1 + 0.1 + 0.1;
y = 0.3;
tolerance = 1e-10; % A "good enough" tolerance for this case.
if ( abs( x - y ) <= tolerance )

disp('x == y');
else
disp('x ~= y');
end
% --- Output: ---

x == y
Several things to note:
As expected, now x and y are treated as equivalent.

In the example above, the choice of tolerance was done arbitrarily. Thus, the chosen value might not be
suitable for all cases (especially when working with much smaller numbers). Choosing the bound intelligently
can be done using the eps function, i.e. N*eps(max(x,y)), where N is some problem-specific number. A
reasonable choice for N, which is also permissive enough, is 1E2 (even though, in the above problem N=1
would also suffice).
Further reading:
See these questions for more information about floating point inaccuracy:
Why is 24.0000 not equal to 24.0000 in MATLAB?

Is floating point math broken?
Section 33.4: What you see is NOT what you get: char vs
cellstring in the command window
This a basic example aimed at new users. It does not focus on explaining the difference between char and
cellstring.
It might happen that you want to get rid of the ' in your strings, although you never added them. In fact, those are
artifacts that the command window uses to distinguish between some types.
A string will print
s = 'dsadasd'
s =

dsadasd
A cellstring will print
c = {'dsadasd'};
c =
'dsadasd'
Note how the single quotes and the indentation are artifacts to notify us that c is a cellstring rather than a
char. The string is in fact contained in the cell, i.e.
c{1}
ans =
dsadasd
Section 33.5: Undefined Function or Method X for Input

Arguments of Type Y
This is MATLAB's long-winded way of saying that it cannot find the function that you're trying to call. There are a
number of reasons you could get this error:
That function was introduced after your current version of MATLAB
The MATLAB online documentation provides a very nice feature which allows you to determine in what version a
given function was introduced. It is located in the bottom left of every page of the documentation:
Compare this version with your own current version (ver) to determine if this function is available in your particular
version. If it's not, try searching the archived versions of the documentation to find a suitable alternative in your
version.
You don't have that toolbox!
The base MATLAB installation has a large number of functions; however, more specialized functionality is packaged
within toolboxes and sold separately by the Mathworks. The documentation for all toolboxes is visible whether you
have the toolbox or not so be sure to check and see if you have the appropriate toolbox.
To check which toolbox a given function belongs to, look to the top left of the online documentation to see if a
specific toolbox is mentioned.

You can then determine which toolboxes your version of MATLAB has installed by issuing the ver command which
will print a list of all installed toolboxes.
If you do not have that toolbox installed and want to use the function, you will need to purchase a license for that
particular toolbox from The Mathworks.
MATLAB cannot locate the function
If MATLAB still can't find your function, then it must be a user-defined function. It is possible that it lives in another
directory and that directory should be added to the search path for your code to run. You can check whether
MATLAB can locate your function by using which which should return the path to the source file.
Section 33.6: The use of "i" or "j" as imaginary unit, loop

indices or common variable
Recommendation
Because the symbols i and j can represent significantly different things in MATLAB, their use as loop indices has
split the MATLAB user community since ages. While some historic performance reasons could help the balance lean
to one side, this is no longer the case and now the choice rest entirely on you and the coding practices you choose
to follow.
The current official recommendations from Mathworks are:
Since i is a function, it can be overridden and used as a variable. However, it is best to avoid using
i and j for variable names if you intend to use them in complex arithmetic.
For speed and improved robustness in complex arithmetic, use 1i and 1j instead of i and j.
Default
In MATLAB, by default, the letters i and j are built-in function names, which both refer to the imaginary unit in the
complex domain.

So by default i = j = sqrt(-1).
>> i
ans =
0.0000 + 1.0000i
>> j
ans =
0.0000 + 1.0000i
and as you should expect:
>> i^2
ans =
-1
Using them as a variable (for loop indices or other variable)
MATLAB allows using built-in function name as a standard variable. In this case the symbol used will not point to
the built-in function any more but to your own user defined variable. This practice, however, is not generally
recommended as it can lead to confusion, difficult debugging and maintenance (see other example do-not-name-a-
variable-with-an-existing-function-name).
If you are ultra pedantic about respecting conventions and best practices, you will avoid using them as loop indices
in this language. However, it is allowed by the compiler and perfectly functional so you may also choose to keep old
habits and use them as loop iterators.
>> A = nan(2,3);
>> for i=1:2 % perfectly legal loop construction
for j = 1:3
A(i, j) = 10 * i + j;
end
end
Note that loop indices do not go out of scope at the end of the loop, so they keep their new value.
>> [ i ; j ]
ans =
2
3
In the case you use them as variable, make sure they are initialised before they are used. In the loop above
MATLAB initialise them automatically when it prepare the loop, but if not initialised properly you can quickly see
that you may inadvertently introduce complex numbers in your result.
If later on, you need to undo the shadowing of the built-in function (=e.g. you want i and j to represent the
imaginary unit again), you can clear the variables:
>> clear i j
You understand now the Mathworks reservation about using them as loop indices if you intend to use them in
complex arithmetic. Your code would be riddled with variable initialisations and clear commands, best way to
confuse the most serious programmer (yes you there!...) and program accidents waiting to happen.
If no complex arithmetic is expected, the use of i and j is perfectly functional and there is no performance penalty.
Using them as imaginary unit:

If your code has to deal with complex numbers, then i and j will certainly come in handy. However, for the sake of
disambiguation and even for performances, it is recommended to use the full form instead of the shorthand
syntax. The full form is 1i (or 1j).
>> [ i ; j ; 1i ; 1j]
ans =
0.0000 + 1.0000i
0.0000 + 1.0000i
0.0000 + 1.0000i
0.0000 + 1.0000i
They do represent the same value sqrt(-1), but the later form:
is more explicit, in a semantic way.

is more maintainable (someone looking at your code later will not have to read up the code to find whether i
or j was a variable or the imaginary unit).
is faster (source: Mathworks).
Note that the full syntax 1i is valid with any number preceding the symbol:
>> a = 3 + 7.8j
a =
3.0000 + 7.8000i
This is the only function which you can stick with a number without an operator between them.
Pitfalls
While their use as imaginary unit OR variable is perfectly legal, here is just a small example of how confusing it could
get if both usages get mixed:
Let's override i and make it a variable:
>> i=3
i =
3
Now i is a variable (holding the value 3), but we only overrid the shorthand notation of the imaginary unit, the full
form is still interpreted correctly:
>> 3i
ans =
0.0000 + 3.0000i
Which now lets us build the most obscure formulations. I let you assess the readability of all the following
constructs:
>> [ i ; 3i ; 3*i ; i+3i ; i+3*i ]

ans =
3.0000 + 0.0000i
0.0000 + 3.0000i
9.0000 + 0.0000i
3.0000 + 3.0000i
12.0000 + 0.0000i
As you can see, each value in the array above return a different result. While each result is valid (provided that was

the initial intent), most of you will admit that it would be a proper nightmare to read a code riddled with such
constructs.
Section 33.7: Not enough input arguments

Often beginning MATLAB developers will use MATLAB's editor to write and edit code, in particular custom functions
with inputs and outputs. There is a Run button at the top that is available in recent versions of MATLAB:
Once the developer finishes with the code, they are often tempted to push the Run button. For some functions this
will work fine, but for others they will receive a Not enough input arguments error and be puzzled about why the
error occurs.
The reason why this error may not happen is because you wrote a MATLAB script or a function that takes in no
input arguments. Using the Run button will run a test script or run a function assuming no input arguments. If your
function requires input arguments, the Not enough input arguments error will occur as you have written a
functions that expects inputs to go inside the function. Therefore, you cannot expect the function to run by simply
pushing the Run button.
To demonstrate this issue, suppose we have a function mult that simply multiplies two matrices together:
function C = mult(A, B)
C = A * B;
end
In recent versions of MATLAB, if you wrote this function and pushed the Run button, it will give you the error we
expect:
>> mult
Not enough input arguments.
Error in mult (line 2)

C = A * B;
There are two ways to resolve this issue:
Method #1 - Through the Command Prompt
Simply create the inputs you need in the Command Prompt, then run the function using those inputs you have
created:
A = rand(5,5);
B = rand(5,5);
C = mult(A,B);
Method #2 - Interactively through the Editor
Underneath the Run button, there is a dark black arrow. If you click on that arrow, you can specify the variables you
would like to get from the MATLAB workspace by typing the way you want to call the function exactly as how you
have seen in method #1. Be sure that the variables you are specifying inside the function exist in the MATLAB
workspace:

Section 33.8: Using `length` for multidimensional arrays
A common mistake MATLAB coders have, is using the length function for matrices (as opposed to vectors, for
which it is intended). The length function, as mentioned in its documentation, "returns the length of the largest array
dimension" of the input.
For vectors, the return value of length has two different meanings:
1. The total number of elements in the vector.

2. The largest dimension of the vector.
Unlike in vectors, the above values would not be equal for arrays of more than one non-singleton (i.e. whose size is
larger than 1) dimension. This is why using length for matrices is ambiguous. Instead, using one of the following
functions is encouraged, even when working with vectors, to make the intention of the code perfectly clear:
1. size(A) - returns a row vector whose elements contain the amount of elements along the corresponding
dimension of A.
2. numel(A) - returns the number of elements in A. Equivalent to prod(size(A)).
3. ndims(A) - returns the number of dimensions in the array A. Equivalent to numel(size(A)).
This is especially important when writing "future-proof", vectorized library functions, whose inputs are not known in
advance, and can have various sizes and shapes.
Section 33.9: Watch out for array size changes

Some common operations in MATLAB, like differentiation or integration, output results that have a different
amount of elements than the input data has. This fact can easily be overlooked, which would usually cause errors
like Matrix dimensions must agree. Consider the following example:
t = 0:0.1:10; % Declaring a time vector

y = sin(t); % Declaring a function
dy_dt = diff(y); % calculates dy/dt for y = sin(t)
Let's say we want to plot these results. We take a look at the array sizes and see:
size(y) is 1x101
size(t) is 1x101
But:
size(dy_dt) is 1x100
The array is one element shorter!
Now imagine you have measurement data of positions over time and want to calculate jerk(t), you will get an array 3

elements less than the time array (because the jerk is the position differentiated 3 times).
vel = diff(y); % calculates velocity vel=dy/dt for y = sin(t) size(vel)=1x100

acc = diff(vel); % calculates acceleration acc=d(vel)/dt size(acc)=1x99
jerk = diff(acc); % calculates jerk jerk=d(acc)/dt size(jerk)=1x98
And then operations like:
x = jerk .* t; % multiplies jerk and t element wise
return errors, because the matrix dimensions do not agree.
To calculate operations like above you have to adjust the bigger array size to fit the smaller one. You could also run
a regression (polyfit) with your data to get a polynomial for your data.
Dimension Mismatch Errors
Dimension mismatch errors typically appear when:
Not paying attention to the shape of returned variables from function/method calls. In many inbuilt MATLAB
functions, matrices are converted into vectors to speed up the calculations, and the returned variable might
still be a vector rather than the matrix we expected. This is also a common scenario when logical masking is
involved.
Using incompatible array sizes while invoking implicit array expansion.

Credits
Thank you greatly to all the people from Stack Overflow Documentation who helped provide this content,
more changes can be sent to [email protected] for new content to be published or updated
adjpayot Chapter 1
Adriaan Chapter 27
agent_C.Hdj Chapter 7
Alexander Korovin Chapter 11
alexforrence Chapters 10 and 3
ammportal Chapter 5
Amro Chapters 1, 12 and 18
Ander Biguri Chapters 25, 33, 28, 3 and 15
Batsu Chapter 2
Cape Code Chapter 15
ceiltechbladhm Chapter 3
Celdor Chapters 12, 10 and 32
chrisb2244 Chapter 1
Christopher Creutzig Chapter 1
codeaviator Chapter 29
daleonpz Chapter 26
Dan Chapters 1 and 11
daren shan Chapters 10 and 27
Dev Chapters 11, 33, 10, 17, 26, 3, 29, 31 and 24
drhagen Chapter 7
DVarga Chapters 1 and 17
EBH Chapters 1, 11, 7, 5 and 32
edwinksl Chapter 33
Eric Chapter 3
Erik Chapters 1, 29 and 32
excaza Chapter 1
flawr Chapter 1
Franck Dernoncourt Chapters 18 and 27
fyrepenguin Chapters 1 and 32
GameOfThrows Chapters 1 and 19
girish_m Chapter 15
Hardik_Jain Chapter 27
Hoki Chapters 33, 28 and 31
il_raffa Chapters 26, 30 and 16
itzik Ben Shabat Chapter 13
JCKaz Chapter 23
jenszvs Chapter 17
Jim Chapter 27
Justin Chapter 17
Kenn Sebesta Chapter 28
Landak Chapters 1, 33, 19 and 4
Lior Chapter 1
Malick Chapters 33, 30 and 32
matlabgui Chapter 29
Matt Chapters 1, 12, 11 and 33
MayeulC Chapter 30
McLemon Chapter 30
mhopeng Chapter 28
mike Chapter 25

Mikhail_Sam Chapters 1 and 4
Mohsen Nosratinia Chapters 10 and 4
nahomyaja Chapter 33
nitsua60 Chapter 16
NKN Chapters 33, 30 and 16
Noa Regev Chapter 14
Oleg Chapters 12, 11, 33, 28 and 18
pseudoDust Chapter 26
R. Joiny Chapter 33
rajah9 Chapter 8
rayryeng Chapter 33
Royi Chapter 24
S. Radev Chapter 4
Sam Roberts Chapter 1
Sardar Usama Chapter 30
Shai Chapters 1, 11, 6 and 15
StefanM Chapters 6, 20, 9, 21 and 22
Suever Chapters 33 and 31
thewaywewalk Chapters 26, 29 and 16
Tim Chapter 33
Trilarion Chapter 15
Trogdor Chapter 10
Tyler Chapters 1, 11 and 7
Umar Chapter 33
Zep Chapters 31 and 16

You may also like

MATLAB Notes For Professionals

Uploaded by

Copyright:

Available Formats

MATLAB Notes For Professionals

Uploaded by

Document Information

Original Title

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

MATLAB Notes For Professionals

Uploaded by

Copyright:

Available Formats

MATLAB

Notes for Professionals

The information presented in this book is not guaranteed to be correct nor

Please send feedback and corrections to [email protected]

MATLAB® Notes for Professionals 1

Section 1.1: Indexing matrices and arrays

MATLAB® Notes for Professionals 2

>> M(2, [1,2,3])

>> M(2, 2:end)

MATLAB® Notes for Professionals 3

>> M([2,3], [1,3])

MATLAB® Notes for Professionals 4

>> M(M > 5)

MATLAB® Notes for Professionals 5

>> M(~(M > 5)) = NaN

for elem = 1:numel(M)

This can be shortened to the following code using logical indexing:

M(M > 5) = M(M > 5) - 2;

Higher dimension matrices

MATLAB® Notes for Professionals 6

Returning ranges of elements

>> colIdx = [3;2;1];

Returning an element multiple times

>> M([1, 1:end], [1:end, end])

MATLAB® Notes for Professionals 7

Section 1.2: Anonymous functions and function handles

Custom anonymous functions

>> f = @(x) sin(x)+cos(x).^2

>> f(1:3) % pass a vector to f

MATLAB® Notes for Professionals 8

Anonymous functions of more than one variable

>> f = @(x,y,z) x.^2 + y.^2 - z.^2

Parameterizing anonymous functions

Input arguments to an anonymous function do not refer to workspace variables

>> x = 3 % x in main workspace

MATLAB® Notes for Professionals 9

Anonymous functions are stored in variables

>> f = @(x) x^2; % an anonymous function

The handle of another existing function can be passed to funHandleDemo:

Using bsxfun, cellfun and similar functions with anonymous functions

MATLAB® Notes for Professionals 10

A = rand(50); % 50-by-50 matrix of random values between 0 and 1

% method 1: slow and lots of lines of code

% method 2: fast and only one line of code

Running the example above results in two outputs:

Elapsed time is 0.015153 seconds.

Section 1.3: Matrices and Arrays

% a 1-by-1 scalar value

% a 1-by-4 row vector

To create a column vector, separate the elements with semicolons:

% a 4-by-1 column vector

MATLAB® Notes for Professionals 11

% a 2 row-by-4 column matrix

% a 4 row-by-2 column matrix

% an unequal row / column matrix

% another unequal row / column matrix

% create a row vector and transpose it into a column vector

% create a 2-by-4 matrix and transpose it to get a 4-by-2 matrix

% transpose a vector or matrix stored as a variable