How delete duplicate strings in cell array
显示 更早的评论
I have a matrix with some cell arrays of chars representing numbers and I need to eliminate duplicate elements of each columns. I tried using unique but when I have elements with diferents lengths unique do not solve the problem.
My data is :
M= {'100' '500' '2 3 4' '50000'; '100' '500' '2' '0';'100' '500' '3' '0';'100' '500' '4' '0';'100' '500' '2 3 4' '20000'};
M = 5x4 cell of chars
unique(M(:,1)) %gives me '100'
unique(M(:,2)) %gives me '500'
both are corrects but
unique(M(:,3)) %gives me '2' '2 3 4' '3' '4'
unique(M(:,4)) %gives me '0' '0 20000' '20000' '50000'
and I need only one result: '2 3 4' for column 3 and '0 20000 50000' for column 4
Matrix M is size N x 4
The result I need is R = {'100' '500' '2 3 4' '0 20000 50000' }
My code is strange and I need a more nice solution, I am using Matlab version 2017b.
for i=1:4
mm = M(:,i);
tmp=unique(mm);
if size(tmp,1) == 1
R{i} = char(tmp);
else
R{i}=char(cellstr(strjoin(unique(split(string(strjoin(mm,' ')))))));
end
end
采纳的回答
You could easily hide the loop inside cellfun, e.g.:
>> fun = @(c)sprintf(' %u',unique(sscanf(sprintf(' %s',c{:}),'%f')));
>> out = cellfun(fun,num2cell(M,1),'uni',0)
out =
' 100' ' 500' ' 2 3 4' ' 0 20000 50000'
Of course with R2017b you can probably replace the outer sprintf with strjoin.
更多回答（0 个）
类别
在 帮助中心 和 File Exchange 中查找有关 Characters and Strings 的更多信息
Translated by 
