Difference between revisions of "Getting started with MATLAB or Python"

From Process Model Formulation and Solution: 3E4
Jump to navigation Jump to search
 
(38 intermediate revisions by 2 users not shown)
Line 1: Line 1:
=Introduction=
{| class="wikitable"
|-
!  MATLAB
! Python
|-
| width="50%" valign="top" |
MATLAB (MATrix LABoratory) is a high level computer language/ interactive software package developed and distributed by MathWorks™. Matlab was first developed in the 1970s by Cleve Molar. Cleve was later joined by John N. Little and Steve Bangert and the three went on to found MathWorks™. MATLAB excels at performing matrix operations and can handle large data sets (stored as matrices) very easily. MATLAB was originally designed as a user friendly interface for LINPACK and EISPACK and so was intended for linear algebra application. Since then MATLAB has greatly expanded it's core abilities to encompass a large array of graphic and numeric applications. These core abilities may in turn be expanded further through the addition of specialized "tool boxes".
| width="50%" valign="top" |
''From Wikipedia'': Python is a high-level programming language whose design philosophy emphasizes code readability. Python aims to combine "remarkable power with very clear syntax", and its standard library of built-in functions is large and comprehensive. 
We will use the [http://numpy.scipy.org/ NumPy and SciPy] modules (the equivalent of a MATLAB toolbox), to provide scientific computing capabilities to Python.  These modules, like MATLAB, allow you to handle large data arrays with little effort.  They provide all the tools we require for this course.
We will also use the [http://matplotlib.sourceforge.net/index.html matplotlib] module, which provides Python with plotting capabilities similar to MATLAB.
|}
=Access / Installation=
{| class="wikitable"
|-
!  MATLAB
! Python
|-
| width="50%" valign="top" |
MATLAB is installed on all computers in the John Hodgins Engineering Student Technology Centre (JHE 233A / 234) as well as the Burke Science Building Labs (BSB 241 / 242 / 244 / 249). The most up to date release of MATLAB is version R2010b. While there are some backward compatibility issues when it comes to older versions of MATLAB, for the level of code that will be encountered in this class students should not encounter issues with portability of code. A MATLAB/SIMULINK student package (version R2010A, good for Windows, Mac, and Linux) is available at the University Book Store for '''$112.95''' (for those students who would like a copy of MATLAB on their laptop or home PC). The student versions comes with the following features:
* MATLAB version R2010a (Everything you need to get started)
* Simulink (Will be useful in your control classes)
* Control System Toolbox
* Signal Processing Toolbox
* Signal Processing Blockset
* Statistics Toolbox
* Optimization Toolbox
* Symbolic Math Toolbox
* Image Processing Toolbox
* Product Demos
I personally have a student copy on my home PC. It's a good buy if you have the extra cash.
| width="50%" valign="top" |
Python is freely available.  The latest stable version that we recommend for the course is version 2.6, because it is compatible with the external libraries that we will use.
[[Course_software | Installation instructions for Python ]] are available on this website.
|}
=Getting Started=
{| border="1" width="100%" cellspacing="10"
<!-- LEFT BLOCK -->
| width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===MATLAB===
*When you launch MATLAB the following window will open:
[[File:MATLAB_start.PNG|center|400px]]
*Do not worry if MATLAB does not open exactly like this. You can easily modify what windows are visible through the '''Desktop''' drop down menu. You can select the windows you want either by toggling them individually (<span style="color:red">Region A</span>) or by selecting a template (<span style="color:blue">Region B</span>). Selecting the ''default'' template should give you the window layout you saw above.
[[File:MATLAB_Desktop_Dropdown.PNG|center|400px]]
*The core MATLAB setup has 6 main sections of interest that you will become very familiar with by the end of this course:
====Command Window====
(1) The command window (or "command line") provides your main means of interacting with the MATLAB software. The command line allows you to enter simple codes that are processed immediately. All program outputs also appear in the command window. While it is technically possible to write an entire program at the command line, this is not recommended. For true programs, scripts and functions will be used.
====Current Directory====
(2) The "current directory" panel lists all of the files (MATLAB or other) that are located in the directory MATLAB is currently accessing. This directory may be changed via the drop down address bar at the top of the main program window. For a program to run properly, ALL CODES required must be in the same directory (for example, if you write a main script file that calls a separate function file, both files must be in the same directory).
====Command History====
(3) The command history records all commands entered in the command window (for possible future reference).
====Workspace====
(4) The workspace lists all variables, matrices, and function handles currently being stored by MATLAB. The workspace also provides basic information on the values being stored, such as size, max, min, etc. Perhaps most useful is the ability to double click variables and matrices in the workspace to open up the '''Variable Editor''' window. The variable editor is essentially a spread sheet representation of your variables that allows for easy manipulation (especially useful for large matrices).
====Editor Window====
(5) The editor window(s) does not immediately open when you launch MATLAB. The editor window is where you write scripts and function files that can be compiled and run in the command window (to be discussed in detail later). To open a new editor window you can either go to File>New>Blank M-File, hit ctrl+N, type "edit" at the command line, or hit the "New M-File" icon at the top left of the main screen (it looks like the little piece of paper with the explosion in the top left corner). To open a pre-existing M-File you need to switch the current directory to the location of the file (as mentioned above) and simply double click on the M-File name in the "current directory" sidebar.
====Plot Window====
(6) When the MATLAB plot tools are used the resulting graphics are displayed in separate plot windows (more on this later).
[[File:MATLAB_Windows_Labeled.PNG|center|400px]]
|}
<!-- RIGHT BLOCK -->
|width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===Python===
|}
|}
=Introduction to the Command Line=
=Introduction to the Command Line=
{| border="1" width="100%" cellspacing="10"


<!-- LEFT BLOCK -->
| width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===MATLAB===
*MATLAB IS CASE SENSITIVE!!! For example, '''A''' is not the same thing as '''a'''. This is very important when working with variable and function names.
*MATLAB IS CASE SENSITIVE!!! For example, '''A''' is not the same thing as '''a'''. This is very important when working with variable and function names.


====Common Command Line Commands and Functions====


=====doc ''function_name''=====
=====doc ''function_name''=====
Line 124: Line 11:
!After
!After
|-align="center"
|-align="center"
|[[File:MATLAB_doc_before.PNG|center|200px]]
|[[File:MATLAB_doc_before.PNG|center|400px]]
|[[File:MATLAB_doc_after.PNG|center|200px]]
|[[File:MATLAB_doc_after.PNG|center|400px]]
|}
|}


Line 135: Line 22:
!After
!After
|-align="center"
|-align="center"
|[[File:MATLAB_edit_before.PNG|center|200px]]
|[[File:MATLAB_edit_before.PNG|center|400px]]
|[[File:MATLAB_edit_after.PNG|center|200px]]
|[[File:MATLAB_edit_after.PNG|center|400px]]
|}
|}


Line 147: Line 34:
!After
!After
|-align="center"
|-align="center"
|[[File:MATLAB_clc_before.PNG|center|200px]]
|[[File:MATLAB_clc_before.PNG|center|400px]]
|[[File:MATLAB_clc_after.PNG|center|200px]]
|[[File:MATLAB_clc_after.PNG|center|400px]]
|}
|}


Line 159: Line 46:
!After
!After
|-align="center"
|-align="center"
|[[File:MATLAB_clear_before.PNG|center|200px]]
|[[File:MATLAB_clear_before.PNG|center|400px]]
|[[File:MATLAB_clear_after.PNG|center|200px]]
|[[File:MATLAB_clear_after.PNG|center|400px]]
|}
|}


Line 187: Line 74:


=====Variables=====
=====Variables=====
 
Variables declared in MATLAB are essentially 1x1 matrices. To declare a variable in MATLAB all you need to do is specify the name of the variable and then set it to a value using the "=" operator (shown below). This is the same for both MATLAB code and the command line. Note below that if you place a semi-colon after the variable declaration then the variable's value is not reprinted (the opposite is also true). Also note that if a variable is not declared for a value then the default '''ans''' variable is assigned to it. Please note that there is only one '''ans''' variable. So as soon as another undeclared value shows up the previous value will be lost. Therefore it is '''always a good idea to give every variable a name''' (to avoid plenty of headaches in the debugging stage). Also note that '''ans''' does not discriminate between variables and larger matrices, so any new undeclared variable or matrix will reassign the value of '''ans'''.
[[File:MATLAB_variables.PNG|center|400px]]
[[File:MATLAB_variables.PNG|center|400px]]


=====Matrices=====
=====Matrices=====
 
Matrices are arrays of values. To declare a matrix in MATLAB all one needs to do is use the "[]" matrix declaration brackets. Within a matrix declaration, the individual row elements are separated by a "," or a blank space (" ") and the rows are separated by a semicolon (thus it is possible to declare a matrix as it would appear on paper...). However, always remember that it is not a requirement to declare each new row on a separate line. As long as the semicolon is there, MATLAB will interpret it as the start of a new row. Note that matrices in MATLAB are declared in row dominant order (similar to C++). That is the first index refers to the row number, the second to the column number, etc. The '''Accessing Submatrices''' section below will outline how to access elements within a matrix. Also note that it is possible to create special types of "ordered" matrices using the colon (:) operator (shown in the '''Dot and Colon Operator''' section).
[[File:MATLAB_Matrices.PNG|center|400px]]
[[File:MATLAB_Matrices.PNG|center|400px]]


=====Matrix Operations=====
=====Matrix Operations=====
MATLAB follows all basic linear algebra rules (obviously :P). That is to say that when you add or subtract two matrices it is done on an element by element basis, while multiplication is done one a row by column basis and division  performs the reverse of this operation. Note however that if you add a "." in front of a *,/, or ^ then the operation '''is performed on an element by element basis'''. Obviously adding a dot in front of a + or - would not do anything extra (and actually results in a compilation error).


[[File:MATLAB_matrix_operations.PNG|center|400px]]
[[File:MATLAB_matrix_operations.PNG|center|400px]]


=====Dot and Colon Operator=====
=====Dot and Colon Operator=====
 
If you add a "." in front of a *,/, or ^ then the operation '''is performed on an element by element basis'''. Obviously adding a dot in front of a + or - would not do anything extra (and actually results in a compilation error). Another usefule operator is the colon operator (:). As will be seen in the next section, the colon operator is instrumental in accessing subsections within a given matrix. However, it is also used in declaring ordered sequences (both in declaring a vector with a known pattern and in declaring the pattern of a for loop. The syntax for using the colon operator to declare an ordered vector is "[start_point:step_size:end_point]". MATLAB will take this code and create a vector that starts at "start_point" and moves up incrementally by "step_size" until it is either equal to or less than "end_point" by a value less than "step_size". This is to say that the sequence will never exceed "end_point". Therefore if a "step_size" was chosen that does not exactly land on "end_point", MATLAB will fall short rather than fall over.
[[File:MATLAB_dot_colon.PNG|center|400px]]
[[File:MATLAB_dot_colon.PNG|center|400px]]


=====Accessing Submatrices=====
=====Accessing Submatrices=====
 
To access a value within a matrix we use the curved matrix access brackets "()". Within these brackets we place the index of the element we wish to access separated by commas (ex. (#,#,...) ). Note that MATLAB starts all its matrix indices at 1 (so it is like Fortran, not C++). MATLAB is also row dominant, so the row index comes first followed by the column index, etc. (This is quizically like C++, not Fortran). To access a subset of a given matrix we use the colon operator (:). In the context of accessing submatrices we simply state "start_point:end_point" (if it makes it easier you can just see it as start_point:1:end_point). You can do this for any of the dimensions within a matrix. If a single ":" is placed in one of the matrix dimensions this translates to "get all of this dimension". Therefore if we wanted to get all the values in row 1 of a matrix, for example (i.e. all the column values of in row 1), we would use (1,:).
[[File:MATLAB_sub_matrices.PNG|center|400px]]
[[File:MATLAB_sub_matrices.PNG|center|400px]]


=====Concatenation=====
=====Concatenation=====
 
Matrices can be combined through a process called concatenation. In horizontal concatenation two matrices are joined from "left to right" (i.e. the new matrix has the same number of rows but a new number of columns equal to the sum of the columns in the two original arrays). Horizontal concatenation of two matrices is achieved through the use of the square matrix declaration brackets "[]". To concatenate the arrays simply place them in the square brackets with a space or comma between them (shown below). Naturally the two matrices being horizontally concatenated must have the same number of rows (or else MATLAB will spit out an error). Vertical concatenation follows the same logic as horizontal concatenation but combines matrices "top to bottom". This time the new matrix has the same number of columns but a new number of rows. To accomplish vertical concatenation the arrays must be placed in a square bracket separated by a semicolon (shown below). This time the arrays must have the same number of columns for the concatenation to be successfull.
[[File:MATLAB_concatenation.PNG|center|400px]]
[[File:MATLAB_concatenation.PNG|center|400px]]
|}
<!-- RIGHT BLOCK -->
|width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===Python===
|}
|}
=Scripts and Functions=
{| border="1" width="100%" cellspacing="10"
<!-- LEFT BLOCK -->
| width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===MATLAB===
|}
<!-- RIGHT BLOCK -->
|width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===Python===
|}
|}
=Code Structures=
{| border="1" width="100%" cellspacing="10"
<!-- LEFT BLOCK -->
| width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===MATLAB===
|}
<!-- RIGHT BLOCK -->
|width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===Python===
|}
|}
=Plotting Data=
{| border="1" width="100%" cellspacing="10"
<!-- LEFT BLOCK -->
| width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===MATLAB===
|}
<!-- RIGHT BLOCK -->
|width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===Python===
|}
|}
=Additional Resources=
{| border="1" width="100%" cellspacing="10"
<!-- LEFT BLOCK -->
| width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===MATLAB===
*[[http://modelling3e4.connectmv.com/mediafiles/mediawiki/7/7c/MATLAB_primer.pdf MATLAB primer]]: A quick guide to MATLAB written by Dr.Prashant Mhaskar.
*[[http://www.mathworks.com/help/pdf_doc/matlab/getstart.pdf Getting Started with MATLAB 7]]: An extensive guide to MATLAB written by MathWorks&trade;.
*[[http://www.math.ufl.edu/help/matlab-tutorial/matlab-tutorial.html Online MATLAB Tutorial]]: An online MATLAB tutorial generated by the University of Florida.
*[[http://www.mathworks.com/help/index.html MathWorks Documentation]]: The central repository for all documentation on products distributed by MathWorks&trade;.
*[[http://www.mathworks.com/matlabcentral/fileexchange/ MATLAB Central File Exchange]]: A database for user generated MATLAB code.
|}
<!-- RIGHT BLOCK -->
|width="50%" valign="top" |
{| border="0" cellpadding="3"
| valign="top" |
===Python===
|}
|}

Latest revision as of 13:40, 24 September 2010

Introduction to the Command Line

  • MATLAB IS CASE SENSITIVE!!! For example, A is not the same thing as a. This is very important when working with variable and function names.


doc function_name

Brings up information on any built-in MATLAB function (i.e. it accesses the help directory entry).

Before After
MATLAB doc before.PNG
MATLAB doc after.PNG
edit

Opens the text editor

Before After
MATLAB edit before.PNG
MATLAB edit after.PNG


clc

Clears the screen

Before After
MATLAB clc before.PNG
MATLAB clc after.PNG


clear

Clears the variables from the workspace (and from memory)

Before After
MATLAB clear before.PNG
MATLAB clear after.PNG


Built in Functions

MATLAB has an array of simple built in functions. Basically all of the rudimentary calculation types that you can think of (sin,cos,exp,etc.) have a built in MATLAB function that may be called at the command line and coded into a script or function file (as the need arises). A few examples are shown in the screenshot below.

MATLAB Common Functions.PNG


Built in Constants

MATLAB also has predefined variables for fundamental constants (ex. pi, Inf, i). Be wary though, as these can be easily overwritten by the user. As a rule of thumb try not to name your variables after any fundamental mathematical constant.

MATLAB constants.PNG


fprintf() and disp()

fprintf() and disp() provide a means of printing text to the command window. While obviously not useful at the command line (as the example below shows), these statements will become useful when you begin writing your own scripts and functions (either as a means of displaying intermediate results or as a means of tracking down bugs). disp() is a simple display function that prints arrays or strings directly to the command window. fprintf() is a more robust function that allows formatting of the string to be printed as well as output specification (for use in printing to files). The default output is set to the command window.

MATLAB fprintf.PNG
Semicolon

Placing a semicolon at the end of a line of code will suppress any output produced by that line. This (along with fprintf() and break points) can help in debugging code. Unless you are debugging it is good practice to end all lines of code with a semi-colon to avoid a bunch of garbage intermediate values being printed out at the command line.

MATLAB semi colon.PNG
Variables

Variables declared in MATLAB are essentially 1x1 matrices. To declare a variable in MATLAB all you need to do is specify the name of the variable and then set it to a value using the "=" operator (shown below). This is the same for both MATLAB code and the command line. Note below that if you place a semi-colon after the variable declaration then the variable's value is not reprinted (the opposite is also true). Also note that if a variable is not declared for a value then the default ans variable is assigned to it. Please note that there is only one ans variable. So as soon as another undeclared value shows up the previous value will be lost. Therefore it is always a good idea to give every variable a name (to avoid plenty of headaches in the debugging stage). Also note that ans does not discriminate between variables and larger matrices, so any new undeclared variable or matrix will reassign the value of ans.

MATLAB variables.PNG
Matrices

Matrices are arrays of values. To declare a matrix in MATLAB all one needs to do is use the "[]" matrix declaration brackets. Within a matrix declaration, the individual row elements are separated by a "," or a blank space (" ") and the rows are separated by a semicolon (thus it is possible to declare a matrix as it would appear on paper...). However, always remember that it is not a requirement to declare each new row on a separate line. As long as the semicolon is there, MATLAB will interpret it as the start of a new row. Note that matrices in MATLAB are declared in row dominant order (similar to C++). That is the first index refers to the row number, the second to the column number, etc. The Accessing Submatrices section below will outline how to access elements within a matrix. Also note that it is possible to create special types of "ordered" matrices using the colon (:) operator (shown in the Dot and Colon Operator section).

MATLAB Matrices.PNG
Matrix Operations

MATLAB follows all basic linear algebra rules (obviously :P). That is to say that when you add or subtract two matrices it is done on an element by element basis, while multiplication is done one a row by column basis and division performs the reverse of this operation. Note however that if you add a "." in front of a *,/, or ^ then the operation is performed on an element by element basis. Obviously adding a dot in front of a + or - would not do anything extra (and actually results in a compilation error).

MATLAB matrix operations.PNG
Dot and Colon Operator

If you add a "." in front of a *,/, or ^ then the operation is performed on an element by element basis. Obviously adding a dot in front of a + or - would not do anything extra (and actually results in a compilation error). Another usefule operator is the colon operator (:). As will be seen in the next section, the colon operator is instrumental in accessing subsections within a given matrix. However, it is also used in declaring ordered sequences (both in declaring a vector with a known pattern and in declaring the pattern of a for loop. The syntax for using the colon operator to declare an ordered vector is "[start_point:step_size:end_point]". MATLAB will take this code and create a vector that starts at "start_point" and moves up incrementally by "step_size" until it is either equal to or less than "end_point" by a value less than "step_size". This is to say that the sequence will never exceed "end_point". Therefore if a "step_size" was chosen that does not exactly land on "end_point", MATLAB will fall short rather than fall over.

MATLAB dot colon.PNG
Accessing Submatrices

To access a value within a matrix we use the curved matrix access brackets "()". Within these brackets we place the index of the element we wish to access separated by commas (ex. (#,#,...) ). Note that MATLAB starts all its matrix indices at 1 (so it is like Fortran, not C++). MATLAB is also row dominant, so the row index comes first followed by the column index, etc. (This is quizically like C++, not Fortran). To access a subset of a given matrix we use the colon operator (:). In the context of accessing submatrices we simply state "start_point:end_point" (if it makes it easier you can just see it as start_point:1:end_point). You can do this for any of the dimensions within a matrix. If a single ":" is placed in one of the matrix dimensions this translates to "get all of this dimension". Therefore if we wanted to get all the values in row 1 of a matrix, for example (i.e. all the column values of in row 1), we would use (1,:).

MATLAB sub matrices.PNG
Concatenation

Matrices can be combined through a process called concatenation. In horizontal concatenation two matrices are joined from "left to right" (i.e. the new matrix has the same number of rows but a new number of columns equal to the sum of the columns in the two original arrays). Horizontal concatenation of two matrices is achieved through the use of the square matrix declaration brackets "[]". To concatenate the arrays simply place them in the square brackets with a space or comma between them (shown below). Naturally the two matrices being horizontally concatenated must have the same number of rows (or else MATLAB will spit out an error). Vertical concatenation follows the same logic as horizontal concatenation but combines matrices "top to bottom". This time the new matrix has the same number of columns but a new number of rows. To accomplish vertical concatenation the arrays must be placed in a square bracket separated by a semicolon (shown below). This time the arrays must have the same number of columns for the concatenation to be successfull.

MATLAB concatenation.PNG