Do not use the for loop to calculate the first number greater than 0 in each line

1 visualizzazione (ultimi 30 giorni)
x=randn(10,10)
...
  2 Commenti
Jan
Jan il 19 Nov 2022
This sounds like a homework question. So please show, what you have tried so far and ask a specific question about Matlab. The forum will not solve your homework, but assist you in case of questions concerning Matlab.
slevin Lee
slevin Lee il 19 Nov 2022
Modificato: Jan il 19 Nov 2022
x=randn(10,10)
x1=x>0
x2=cumsum(x1,2)
x3=x2==0
x4=sum(x3,2)+1
x5=x(:,x4).*eye(10,10)
answer=sum(x5,2)
sorry!
I just want to see if there's a better way

Accedi per commentare.

Risposta accettata

Jan
Jan il 19 Nov 2022
Modificato: Jan il 19 Nov 2022
x = randn(10, 10)
y = x.'; % Required for the last step y(y3==1)
y1 = y > 0
y2 = cumsum(y1)
% Now a row can contain multiple 1's. Another CUMSUM helps:
y3 = cumsum(y2)
y(y3 == 1)
Another way:
x = randn(10, 10)
m = x > 0
y = cumsum(x .* m, 2)
y(~m) = Inf % Mask leading zeros
min(y, [], 2)
Or with beeing nitpicking for the formulation "no for loop":
result = zeros(1, 10);
m = x > 0;
k = 1;
while k <= 10
row = x(k, :);
result(k) = row(find(row(m(k, :)), 1));
k = k + 1;
end
And another method:
[row, col] = find(x > 0);
first = splitapply(@min, col, row);
x(sub2ind(size(x), 1:10, first.'))

Più risposte (1)

DGM
DGM il 19 Nov 2022
Modificato: DGM il 19 Nov 2022
Here's one way.
x = randn(10,10)
x = 10×10
0.5355 -1.4904 -0.9662 -0.7693 0.6789 0.6031 0.2375 -1.0743 0.6916 -0.8738 -0.0884 0.5313 -0.2913 1.0506 -1.1930 0.1070 0.2047 -0.2512 2.0512 0.1435 -0.1235 0.5840 1.3542 -0.3692 1.1048 -0.4211 -2.1762 -0.3553 0.8296 -2.1670 -0.6151 0.7344 -1.1325 -1.9521 -1.5595 0.8968 -0.8484 -2.5253 -0.1668 1.8909 1.0192 -0.4541 0.9363 -0.1635 -1.7753 -0.2266 1.1185 2.0062 -1.3746 -0.3283 -0.2484 -1.0466 1.2285 0.4551 -1.1110 -1.1021 0.3378 0.2945 -0.2886 1.0569 1.4228 -1.4178 0.5278 0.1037 0.7001 -0.3782 0.3081 1.4122 0.6484 -1.0982 -0.0277 -1.6063 -0.8334 -0.2916 -1.0505 -0.6081 -0.4252 -1.0511 0.4218 0.2551 -0.2180 0.5246 0.2145 1.4641 2.2411 -1.4396 -0.2848 -0.4727 2.5402 2.0645 0.5302 0.6045 -0.7767 1.2057 -0.9240 0.0234 1.1590 -0.0401 1.3279 0.5901
[~,idx] = min(x<=0,[],2) % note the logical inversion
idx = 10×1
1 2 2 2 1 3 1 9 2 1
firstnz = x(sub2ind(size(x),(1:size(x,1))',idx)) % get the values from the array
firstnz = 10×1
0.5355 0.5313 0.5840 0.7344 1.0192 1.2285 1.4228 0.4218 0.5246 0.5302

Tag

Community Treasure Hunt

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

Start Hunting!

Translated by