# Delete row from Matrix

1,471 views (last 30 days)
Nancy on 21 Jun 2012
I have a Matrix of 400 rows and 40 columns. I want to completely remove rows 3 and 9 to get a matrix with 398 rows. How can I do it in MATLAB.

Walter Roberson on 7 Sep 2016
Yuvraj Praveen Soni on 9 Dec 2019
if assume A is your matrix with 400 rows and 40 column,
To remove 3rd row
a(3,:)=[];
now your 9th row become 8th row;
a(8,:)=[];
and with this your 3rd and 9th row will be deleted, you can also check the size by
size(a);
Thank You

Jan on 22 Jun 2012
Edited: MathWorks Support Team on 9 Nov 2018
If you have a matrix A and want to delete the 3rd and 9th rows, you can use the command:
A([3,9],:) = [];

Show 1 older comment
Walter Roberson on 6 Sep 2018
databases have their own implementations and their own tradeoffs. For example,
deletes = [3, 9];
q = ['(', strjoin(sprintfc('%d', deletes), ','), ')'];
sqlquery = ['DELETE * FROM inventoryTable WHERE productNumber in ' q];
curs = exec(conn,sqlquery);
How the database implements this is up to the database. As is typical with databases, the efficiency of this can depend upon whether productNumber is an indexed key. But that has nothing to do with MATLAB.
Katarina on 18 Jun 2019
I used this in MATLAB R2017a and e.g.
A(3,:) = []
removes a row, but
A([3,5],:) = []
A([3,3],:) = []
A([k,k],:) = []
also removes only a row, but not a column as said above.
Is it that it doesn't work in my version of matlab of this answer needs updating?
Walter Roberson on 18 Jun 2019
Katarina, I am not clear as to where you are seeing anything about deleting columns in this Answer?
If you want to delete a column then name it in the second index:
A(:,k) = [];

Peter on 30 Nov 2012
"I have a Matrix of 400 rows and 40 columns.I want to completely remove rows 3 and 9 to get a matrix with 398 rows. How can I do it in MATLAB."
Matrix_2 = Matrix_1( [1:2,4:8,10:end] , : )
Best,
Pete

Dan W on 23 Jan 2015
I'm not sure if this is new syntax or not, but it works with R2012a and it's fast and simple.
x = rand(100);
tic;
x([3,9],:) = [];
toc; % Elapsed time is 0.000230 seconds.

Mehul Agrawal on 10 Jun 2016
Thanks, this works pretty easy. But I have a problem where the elements to remove are decided dynamically. So, I have a matrix m1 of size 100 X 100. And another matrix m2 of size 10X1. m2 has the row number to remove from m1 (they are not in any order). What is the best way to do this ?
Eg: m1 = rand(100); m2 = [1,6,4,8,10]; (this is the output of another function call).
Andrei Bobrov on 10 Jun 2016
Hi Mehul! It new question.
Andrei Bobrov on 10 Jun 2016
out = m1;
out(m2,:) = [];

Andrei Bobrov on 21 Jun 2012
m = m(setdiff(1:size(m,1),[3,9]),:);

Walter Roberson on 21 Jun 2012
It copies the rows of m that are _not_ row 3 or 9.
Jan on 22 Jun 2012
SETDIFF has a remarkable overhead. ISMEMBER is smarter and twice as fast for a 100x100 matrix:
m = m(~ismember(1:size(m, 1), [3,9]), :);
pradeep kumar on 30 Aug 2014
@ Andrei Bobrov , @ Walter Roberson,@ Jan Simson . how delete a particular row and column of a matrix by using "setdiff" . Say m= [1 2 3 4 ; 5 6 7 8; 9 10 11 12 ; 13 14 15 16 ]. i want to delete 1st row and 2nd column to obtain m=[5 7 8; 9 11 12;13 15 16]

Alireza Rezvani on 19 Jun 2016
sry, how i can Deleting individual columns of a matrix, any body know?

Muhammad Usman Saleem on 19 Jun 2016
Assume out is your matrix and you want to delete its first column, try this code,
out(:,1) = [];
surendra bala on 31 Mar 2018
Yes. This is the easiest way you can do.

Drew Closner on 10 Oct 2016
So I have to make a function that is able to delete a row in a matrix [I have 3 by 3]. I am not sure how to do this, please leave any help you can!

LISSA DUVVU on 29 Sep 2018
i want to delete all columns data where the 4th row contains number 7

#### 1 Comment

Jan on 7 Oct 2018
Please do not attach a new (and very vague) question in the section for answers of another questions. Such thread-hijacking is confusing only, because it is not longer clear, to which question an answer belong. Please open your own question and delete this pseudo-answer. Thanks.