how to separate string matrix by zeros

4 visualizaciones (últimos 30 días)
Josh
Josh el 2 de Jun. de 2021
Comentada: Josh el 2 de Jun. de 2021
I imported some tables into matlab as string matrices. Each row contains 16 values. I would like to separte them by a group of zeros but don't know how to do it. For example, as highlighted on the imageI'd like to extract those three sections from the matrix, but don't know how to code the loop
k=1;
for i=1:size(m,1)
x = m(i,:);
str = sprintf('%s,', x{:});
num = sscanf(str, '%g,', [16, inf]);
val(:,k)=num;
if sum(val(:,k))~=0
% read more rows in until sum(num)==0
k=k+1;
else
end
end
  1 comentario
Rik
Rik el 2 de Jun. de 2021
What is your intended output? A cell vector with one section in each cell element?

Iniciar sesión para comentar.

Respuesta aceptada

Stephen23
Stephen23 el 2 de Jun. de 2021
Editada: Stephen23 el 2 de Jun. de 2021
This would have been easier if you had imported the data as numeric, rather than as string.
Rather than relying on low-level loops and inefficient array expansion the MATLAB approach would be to use indexing and handle the entire matrix or "non-zero blocks" at once. For example:
S = load('matlab.mat')
S = struct with fields:
m: [7195×1 string]
M = sscanf(sprintf('%s,',S.m),'%f,',[16,Inf]).' % convert to numeric
M = 7195×16
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 7 5 1 7 5 0 0 8 4 3 5 2 6 1 1 0 7 5 1 7 5 0 0 8 4 3 5 2 6 1 7 2 6 0 7 7 1 1 8 0 3 0 5 4 7 7 7 2 6 0 7 7 2 1 8 0 3 0 5 4 7 7 7 7 10 13 13 13 14 14 16 9 10 10 12 13 12 13 7 7 10 13 13 13 14 14 16 9 10 10 13 12 13 13 15 9 9 9 10 10 10 9 15 14 15 15 15 14 14 13
Your question is unclear what you expect for the output. If you want one numeric matrix with all "non-zero" rows in it, then just use basic MATLAB indexing:
X = any(M~=0,2);
Z = M(X,:)
Z = 6199×16
1 0 7 5 1 7 5 0 0 8 4 3 5 2 6 1 1 0 7 5 1 7 5 0 0 8 4 3 5 2 6 1 7 2 6 0 7 7 1 1 8 0 3 0 5 4 7 7 7 2 6 0 7 7 2 1 8 0 3 0 5 4 7 7 7 7 10 13 13 13 14 14 16 9 10 10 12 13 12 13 7 7 10 13 13 13 14 14 16 9 10 10 13 12 13 13 15 9 9 9 10 10 10 9 15 14 15 15 15 14 14 13 15 9 9 9 10 10 10 9 15 14 15 15 15 14 14 13 13 12 13 14 14 14 13 13 13 13 12 11 13 13 12 13 13 12 13 14 14 14 13 13 13 13 12 11 13 13 12 13
If you want to keep those "non-zero groups" separate then you should split them into a cell array:
D = diff([false;X;false]);
Y = find(D<0)-find(D>0);
C = mat2cell(M(X,:),Y,16);
Checking the first three matrices:
C{1:3}
ans = 12×16
1 0 7 5 1 7 5 0 0 8 4 3 5 2 6 1 1 0 7 5 1 7 5 0 0 8 4 3 5 2 6 1 7 2 6 0 7 7 1 1 8 0 3 0 5 4 7 7 7 2 6 0 7 7 2 1 8 0 3 0 5 4 7 7 7 7 10 13 13 13 14 14 16 9 10 10 12 13 12 13 7 7 10 13 13 13 14 14 16 9 10 10 13 12 13 13 15 9 9 9 10 10 10 9 15 14 15 15 15 14 14 13 15 9 9 9 10 10 10 9 15 14 15 15 15 14 14 13 13 12 13 14 14 14 13 13 13 13 12 11 13 13 12 13 13 12 13 14 14 14 13 13 13 13 12 11 13 13 12 13
ans = 13×16
0 0 0 0 0 0 0 0 0 0 0 1 7 13 14 13 0 0 0 0 0 0 0 0 0 0 0 2 10 13 14 13 14 11 6 6 7 7 6 7 4 12 12 13 14 13 15 14 14 11 6 6 7 7 6 7 4 12 12 13 14 13 15 14 13 13 14 14 14 14 13 12 13 14 13 14 14 13 12 12 13 13 14 14 14 14 13 12 13 14 13 14 14 13 12 12 12 12 12 9 11 7 12 14 15 15 13 9 9 9 9 9 12 12 12 9 11 7 12 14 15 15 13 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 14 13 13 9 9 9 9 9 9 9 9 9 9 9 9 9 14 13 13
ans = 58×16
1 7 12 13 11 12 6 2 8 6 4 2 6 2 6 0 1 7 12 13 11 12 6 2 8 6 4 2 6 2 6 0 0 0 0 0 7 7 0 2 0 3 4 0 7 5 1 0 1 0 0 0 7 7 1 0 5 0 4 0 5 5 1 0 7 5 2 7 7 0 1 4 4 4 1 7 7 1 0 0 7 5 2 7 7 0 1 4 4 4 1 7 7 1 0 0 6 5 1 6 7 2 0 6 4 1 7 7 2 1 7 4 6 5 1 6 7 2 0 6 4 1 7 7 2 1 7 4 5 0 3 7 1 0 7 4 1 2 7 7 1 1 8 2 5 0 3 7 1 3 7 6 0 7 5 7 1 0 6 3

Más respuestas (1)

KSSV
KSSV el 2 de Jun. de 2021
id = zeros([],1) ; % indices of required strings
count = 0;
for i = 1:size(m,1)
t = str2num(m(i,:)) ;
if any(t)
count = count+1 ;
id(count) = i ;
end
end
iwant = m(id,:)

Categorías

Más información sobre Numeric Types en Help Center y File Exchange.

Community Treasure Hunt

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

Start Hunting!

Translated by