About matlab

x = [1,3,10]; y = [2 5 8]; x(4) = 12 y = [y 10]

x = [1;3;10];

x = []

x = [1: 4; 1 22 33 44];

2. It is possible not to initialize the matrix in advance, but use the operator equal to write some values ​​in X(i, j) even outside the allowable dimension.

This is prohibited in C++, but for Matlab, the matrix will optionally be expanded to the minimum size such that the entered element will be included in the matrix. The rest of the added elements are initialized to zero.

x(10,9) = 12;

However, it is impossible to read without an appropriate shape initialization.

3. You can write/read not only one variable but several at once

x = [1: 3; 4: 6]; display (x); x (1, [2,3]) = 100; display (x); y = [1: 3; 4: 6]; display (y ([1,2], 3)); display (y ([1,2], [3,1,2]));

5. The last index when referring to a line, the column can be replaced with the keyword end . 

x = [1: end; 4: 6; 7: 9];

display (x (1: end, [1,2,3,3]));

6. Expression 1: end has the same meaning as ":", as in Python for vectors in which it performs complete slicing.

7. Matrices can be combined into a block matrix

x = [[1: 3; 4: 6], [1: 3; 4: 6]]

8. Matrix transposition is performed using the <'> operator

x = [[1: 3; 4: 6]]'

9. To transpose a vector, you need to use parentheses or square brackets before applying the transpose operation '

(1:4)'

10. The componentwise multiplication of matrices is performed by the operator <.*>

X .* Y

11. Matrix multiplication (which is the usual multiplication of a matrix <m, n> * <n, p>) is written as

x * y

12. When performing a binary operation on a matrix and a scalar the scalar is expanded into a matrix of the appropriate size and the binary operation is performed componentwise.

13. When ":" is passed to the matrix during indexing, the matrix is ​​interpreted as a one-dimensional column vector.

This column vector is constructed by concatenating matrix column by column

x = [1: 3; 4: 6]

x (:)

14. Converting a vector to a matrix (useful when saving to a binary file)

a = [1 2 3 4 5 6]; dims = [2 3]; reshape (a, dims)

15. Repeat matrix

repmat ([1 2 3], 3,2)

ans =

     1 2 3 1 2 3

     1 2 3 1 2 3

     1 2 3 1 2 3

% import numpy as np; np.tile(np.arange(1,4), (3,2)) 

% Analog for numpy, a popular numeric library for Python

15. Reshape matrix

reshape ([1 2 3 4 5 6], [3 2])

ans =

     1 4

     2 5

     3 6

% import numpy as np; np.arange(1,7).reshape ((3, 2))

% Analog for numpy, a popular numeric library for Python

 * .m exports only the first function from the module, the rest remain local to the module.

The name of this function must match the name of the * .m file without extension.

% myRand.m file

% Function with args. this function can be called from outside

function [rnd, s] = myRand (low, high)

rnd = low + rand (3,4) * (high-low);

s = sumAll (rnd);

end

% Subfunction with args, available only in this module

function sm = sumAll (x)

sm = sum (x (:));

end

These are variables within a function that persist internally between intermediate values ​​of the function, much like static in C

persistent x;

Perform initial initialization using isempty () checking for an empty 0-by-0 matrix.

function a = testMe ()

   persistent x;

   if isempty (x)

       x = 0;

   end

   x = x + 1;

   a = x;

end

If you want to reset persistent variables, then you need to call from outside clear with the name of the function.

clear testMe

There are strings in Matlab that store chars of the ASCII type

a = 'qwerty';

length (a);

b = a (2);

a (3) = '!';

display (a);

% Convert to an array of chars

double(a)

% String replace

strrep('a b c', 'b', 'replace')

Creation of a structure and immediately filling in its key, value. 

a = struct ('a', 123, 'b', 456)

After that similar to Python, it's possible to append argument just via explicit writing to attributes that did not exist before writing.

a.c = 1

There is limited support, they are called cells. 

They are really needed while storing an array of strings. Suitable for storing objects of different types in an array.

b{1} = 123; % Create 1x1 cell

b{2} = '123'; % Create 1x1 cell

a = {1,2};

for = i 1: length (a)

end

m = cell (2,3); % Create 2x3 cell

Declare a variable in a function that is global, not local

View help for exist command. Similar to Python's help(command)

Same as <x, y = 1,1;> Identical construct is available in Python.

In Python for this construct, the boxing/unboxing of the value occurs.

In Python, as in Matlab, a complex type variable is built into the language, unlike C ++. 

Complex types as I know are represented there as elements of the standard library

Append value to the vector

Vector filtering - only obtain items from the vector which satisfies the constraint.

In the version for Matlab, not only scalar 2 can be used, but also a vector of values ​​of a suitable size for pairwise comparison.

In Matlab terms, it is called logical indexing

In Python terms, it is called list comprehensions

If it is necessary to read the elements of a matrix for this operation, the matrix is ​​converted into a vector with a column-major order.

If it is necessary to write, then the target matrix is not converted into a vector, but viewing the elements of the matrix is ​​like occurs in column-major order.

 Set (update) values ​​of part of vector elements

 View variable type

The creation of a structure does not require a prior declaration of the structure type. In Matlab, you don't even need to write  a = Empty ()

Extra functions for working with structures: