0×1 empty double column vector
Ältere Kommentare anzeigen
Hi,
I have an array of 19x2 double (r_F), contains 19 integers in column 1 and 2. Now I want to do a vlookup of integers that are stored in the array auswahl; the size is 100x1. If the integer of auswahl were find in column 1 of r_F then write the integer of column 2 in the array R_X. The following code do that for me.
R_X = [];
for i = 1:size(auswahl, 1)
index = find(r_F(:, 1) == auswahl(i));
content = r_F(index, 2);
disp(i + ": " + index + " " + content)
R_X = [R_X; content];
end
Strangely enough contains the result on index 23 and 74 0x1 empty double column vector?! Has anyone an idea why this is happen?
Kind regards,
Patrick
2 Kommentare
Guillaume
am 12 Feb. 2020
Supplement: I fill the vector auswahl with the following command:
auswahl = round(rand(100,1)*19,0);
It seems, that auswahl contains not only integers.
Shubhanshi Mishra
am 3 Jul. 2021
Hello;
I have a matrix(100*4). I want to get the index of a particular value in 4th column. I am using..
idx = find(matrix(:,4)==desiredvalue)
for the values of indices 2 to 50, it is coming as 0*1 empty double column vector and for indices 51 to 100, it is coming correct.
I am not able to understand , why it is so.
please help.
Antworten (1)
Guillaume
am 12 Feb. 2020
a) If you want to generate integers between 0 and 19 use:
auswahl = randi([0 19], 100, 1); %use randi to generate random integers
b) Despite what you may have been taught, loops are rarely needed in matlab and often complicates everything. To keep the elements in the second column of r_F for which the first column is found in auswahl:
R_X = r_F(ismember(r_F(:, 1), auswahl), 2); %keep elements of column 2 of r_F for which column 1 is a member of auswahl
8 Kommentare
Patrick Hartzsch
am 12 Feb. 2020
Guillaume
am 12 Feb. 2020
Ah, ok. then it's:
[found, where] = ismember(auswahl, r_F(:, 1)); %find which elements of auswahl are in first column of r_F and in which row
R_X = r_F(where(found), 2); %get matching second column of r_F
Patrick Hartzsch
am 13 Feb. 2020
Patrick Hartzsch
am 13 Feb. 2020
Guillaume
am 13 Feb. 2020
"Ok, the variables found and where were 100x1 logical and 100x1 double. But the result R_X is an 86x1 double"
Nothing surprising about that. Indeed found and where should be the same size as auswahl. R_X is going to have as many elements as there are true (1) values in found, so it can be 86x1 if nnz(found) is 86.
"when I create the vector [with randi] the behavior described above occurs. If I create the vector manually and paste the integers in it, the code were running fine.
randi creates integers. I suspect that the values in r_F (the first column) are not exact integers (maybe of by 2e-16 from the integer value) and what you copy manually are also not exact integers. Attach a mat file containing r_F so we can check.
or test it yourself with:
rows = find(mod(r_F(:, 1), 1) ~= 0);
assert(isempty(rows), 'row(s) %s of first column of r_F contain non-integers', strjoin(compose('%d', rows), ', '))
Patrick Hartzsch
am 14 Feb. 2020
I'm afraid I can't see anything wrong with the results produced by your script. Both auswahl and the first column of r_F are guaranteed to contain integers due to the way you construct them.
edit: However note that auswahl contains integers from 0 to 19 included, while r_F contains integers from 1 to 19. Perhaps, you meant to have integers from 1 to 19 in auswahl all along, in which case:
- I did write "If you want to generate integers between 0 and 19 use"
- That's what your original round(rand(100,1)*19,0) did
If you want integers between 1 and 19 (included) then:
auswahl = randi(19, n, 1)
Patrick Hartzsch
am 15 Feb. 2020
Kategorien
Mehr zu Logical finden Sie in Hilfe-Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!