# How to remove the first element and last element based on First column

5 views (last 30 days)
Smithy on 29 Apr 2022
Commented: Voss on 29 Apr 2022
I have a matrix 23 by 2 matrix. and it looks as below.
First Column starts from 1 to 5 and second column is the value.
I would like to removes the beginning(i.e the first element) and removes the end(i.e the last element) based on first column.
For example, for 1 from first column, I hope to remove first element [1,293] and last element [1,348]
and for 2 from first column, I hope to remove first element [2,237] and last element [2,326]
and for 3 from first column, I hope to remove first element [3,186] and last element [3,327]
until for 5 from first column.
Is there a way for it?
x1 = [1,293;...
1,274;...
1,329;...
1,348;...
2,237;...
2,374;...
2,326;...
3,186;...
3,237;...
3,274;...
3,327;...
4,186;...
4,235;...
4,274;...
4,326;...
4,359;...
5,161;...
5,164;...
5,186;...
5,235;...
5,274;...
5,326;...
5,358];

Voss on 29 Apr 2022
Here's one way
x1 = [1,293;...
1,274;...
1,329;...
1,348;...
2,237;...
2,374;...
2,326;...
3,186;...
3,237;...
3,274;...
3,327;...
4,186;...
4,235;...
4,274;...
4,326;...
4,359;...
5,161;...
5,164;...
5,186;...
5,235;...
5,274;...
5,326;...
5,358];
idx = find(diff(x1(:,1))) % indexes where the first column of x1 changes
idx = 4×1
4 7 11 16
idx = idx+[0 1] % those indexes and the one after
idx = 4×2
4 5 7 8 11 12 16 17
idx = [idx(:); 1; size(x1,1)] % include the first and last rows of x1
idx = 10×1
4 7 11 16 5 8 12 17 1 23
x1(idx,:) = []; % remove those rows
disp(x1);
1 274 1 329 2 374 3 237 3 274 4 235 4 274 4 326 5 164 5 186 5 235 5 274 5 326
##### 2 CommentsShowHide 1 older comment
Voss on 29 Apr 2022
You're welcome!

dpb on 29 Apr 2022
Edited: dpb on 29 Apr 2022
Of course there is. :)
But, specifically how to do so is dependent on answer to several other questions regarding what is really wanted.
First, how were the elements above determined to want to be removed? The best way to do something of this sort is to have an algorithm that determines which values are those in Q? and thus avoid having to make up lists of specific hard-coded values.
Secondly, however those values are determined, are there
1. Possibility of more than one element of same value, and
2. Are the values exactly those to machine precision (testing for equality with floating point values is an iffy proposition)
3. Is there a need to remove these in pairs as given and do something else with the remainder before moving on to the next pair or can they all just be deleted in "one swell foop!"
?
With answers to those questions, there's at least enough to begin to posit an answer...

R2021a

### Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!