A question about ismember function

Suppose I have a matrix

M=[1 1 1; 1 1 1; 0 1 0; 5 5 5; 9 1 8; 5 5 5]; D=[1 1 1; 5 5 5; 9 1 8];
I=ismember(M,D,'rows');

I want to assign a number 1 to rows in M that are in D and 0 to rows in M that are not in D then, remove from M all the rows that crossponds to 1.

The answer is

I=   [1 1 0 1 1 1]'
     

so I want to remove row in M that crossponds to 1 so I type

M(I,:)=[]; 

It gives M=[0 1 0]; which is the required.

My question is this correct for all problems of this kind?

what is the meaning of M(I,:) ? is not I a vector, how can we type M(vector,:) ? I mean I understand the command M(2,:) for example it means get me the second row in M but what is the meaning of M(vector,:)?

Answer 1

[Dave B]

Check out the documentation page here: https://www.mathworks.com/help/matlab/math/array-indexing.html (Indexing with Logical Values is about halfway down)

There's also a nice article on indexing here: https://www.mathworks.com/company/newsletters/articles/matrix-indexing-in-matlab.html

What this boils down to is you can index by row/column numbers: M(3,:) M([1 3 5],:) or by logicals (M([true;false;true;false;true],:))

M=magic(5)

M = 5×5

17 24 1 8 15 23 5 7 14 16 4 6 13 20 22 10 12 19 21 3 11 18 25 2 9

M(3,:)

ans = 1×5

4 6 13 20 22

M([1 3 5],:) 

ans = 3×5

17 24 1 8 15 4 6 13 20 22 11 18 25 2 9

M([true;false;true;false;true],:)

ans = 3×5

17 24 1 8 15 4 6 13 20 22 11 18 25 2 9

Note that logical indexing is really really useful!

M(M(:,1)>10,:) % Rows of M where the first column is greater than 10

ans = 3×5

17 24 1 8 15 23 5 7 14 16 11 18 25 2 9

M(mod(M,2)==0)' % Even values in M

ans = 1×12

4 10 24 6 12 18 8 14 20 2 16 22

load sonnetsTable.mat
tbl.Word(tbl.Count>20)' % Words that occurred more than 20 times

ans = 1×22 cell array
    {'Which'}    {'beauty'}    {'being'}    {'every'}    {'heart'}    {'love'}    {'might'}    {'night'}    {'praise'}    {'shall'}    {'should'}    {'still'}    {'sweet'}    {'thee'}    {'their'}    {'thine'}    {'those'}    {'though'}    {'where'}    {'which'}    {'world'}    {'would'}

load patients

% mean of smokers, non smokers, and smokers with weight above the median
[mean(Diastolic(Smoker)) mean(Diastolic(~Smoker)) mean(Diastolic(Smoker & Weight>median(Weight)))]

ans = 1×3

89.9118 79.3788 90.3182