How to store a string data into an array

2018 12 9
1 个回答
更新时间：2018 12 9
9 次查看（30 天）

0 个投票

I have a an array like this:
trainLabels = zeros(315,1);
And I would like to put a string for example 'bass' on every element of that array. I tried using
fold = 'bass';
testLabels(testctr,1) = fold;
but its giving me this error:
Unable to perform assignment because the size of the left side is 1-by-1 and the size of the right side is 1-by-4.

请先登录，再回答此问题。

回答（1 个）

madhan ravi
madhan ravi 2018-12-9

0 个投票

trainLabels = cell(315,1);
trainLabels(:)={'bass'}

15 个评论
显示 13更早的评论 隐藏 13更早的评论

Janrex Pensader
Janrex Pensader 2018-12-9
the fold variable is changing every time because it is in a loop and i need to store those into the array.
madhan ravi
madhan ravi 2018-12-9
ah then it would be
trainLabels{i} = {your variable} % i is the loop iterator
Janrex Pensader
Janrex Pensader 2018-12-9
It's showing this error now
Unable to perform assignment because brace indexing is not supported for variables of this type.
madhan ravi
madhan ravi 2018-12-9
Ah upload your datas as a .mat file and upload the part of the code your trying to store datas
Janrex Pensader
Janrex Pensader 2018-12-9
the data that I need to store are foldernames ang here is my code:
files = dir('C:\Program Files\MATLAB\R2018a\bin\vision\data\');
files(ismember( {files.name}, {'.', '..'})) = [];
dirFlags = [files.isdir];
% Extract only those that are directories.
subFolders = files(dirFlags);
% Print folder names to command window.
trainctr =1;
testctr =1;
trainLabels = zeros(315,1);
testLabels = zeros(336,1);
trainingImages = zeros(32,32,3, 315);%315
testingImages = zeros(32,32,3, 336);%336
%subfolder loop
for k = 1 : length(subFolders)
%FOR DEBUGGING: Print all subfolders
%fprintf('Sub folder #%d = %s\n', k, subFolders(k).nametest(:,:,1,1));
direc = 'C:\Program Files\MATLAB\R2018a\bin\vision\data\';
newDir = strcat(direc,subFolders(k).name);
subname = subFolders(k).name;
%FOR DEBUGGING: Print new Directory
%fprintf('New Dir: %s\n', newDir);
scanDir = dir(newDir);
scanDir(ismember( {scanDir.name}, {'.', '..'})) = [];
scanFlags = [scanDir.isdir];
scanSubFolders = scanDir(scanFlags);
%note category loop
for i = 1 : length(scanSubFolders)
newDir2 = strcat(newDir,'\');
newDir2 = strcat(newDir2,scanSubFolders(i).name);
newDir3 = newDir2;
newDir2 = strcat(newDir2,'\*.png');
imagefiles = dir(newDir2);
nfiles = length(imagefiles);
fprintf('%s',subFolders(k).name);
fprintf('/');
fprintf('%s\n',scanSubFolders(i).name);
fold = scanSubFolders(i).name;
%image loop
for ii=1:nfiles
fprintf('%s\n',imagefiles(ii).name);
%
photo=imagefiles(ii).name;
filename = fullfile(newDir3,photo);
Image = imread(filename);
folder = strcmp('testing',subname);
if folder == 1
test(:,:,:,testctr) = Image;
testLabels{testctr} = {fold};
testctr = testctr+1;
else
train(:,:,:,trainctr) = Image;
trainLabels{trainctr} = {fold};
trainctr = trainctr+1;
end
end
fprintf('\n');
end
fprintf('\n');
end
madhan ravi
madhan ravi 2018-12-9
You didn't look into my answer properly! You forgot to add this line before the loop
trainLabels = cell(315,1);
Janrex Pensader
Janrex Pensader 2018-12-9
owhh sorry can you explain what that line will do?
Janrex Pensader
Janrex Pensader 2018-12-9
I added trainLabels = cell(315,1); at the top of the code and the error is gone but when i checked the elements of the array it shows this.
testLabels
testLabels =
336×1 cell array
{1×1 cell}
{1×1 cell}
{1×1 cell}
{1×1 cell}...
madhan ravi
madhan ravi 2018-12-9
It's called preallocation it ensures that the variable is a cell array.
madhan ravi
madhan ravi 2018-12-9
celldisp(testLabels) %to view the contents of the cell
If my answered solved your question make sure to accept the answer and give a vote.
Janrex Pensader
Janrex Pensader 2018-12-9
are there any other ways to do this? because I'm going to connect this to a code that needs an ordinary array of names.
madhan ravi
madhan ravi 2018-12-9
编辑：madhan ravi 2018-12-9
"I'm going to connect this to a code that needs an ordinary array of names."
why not ? you can easily connect it
Janrex Pensader
Janrex Pensader 2018-12-9
The code I'm going to connect it to is the object detection sample code from mathworks.
https://www.mathworks.com/help/vision/examples/object-detection-using-deep-learning.html
My code will replace the loading of cifar data to the cnn. the trainLabels will replace the array trainingLabels, and testLabels will replace testingLabels. if i sue your solution for my probem will this code accept my array?
madhan ravi
madhan ravi 2018-12-9
To be frank and honest I don't have any experience with that field but in the link you suggested cell arrays ahd been used , why not try it and find it out?
Janrex Pensader
Janrex Pensader 2018-12-9
It didnt work it cause an error it shows "Invalid training data. Y must be a vector of categorical responses." where y is the trainingLabels

请先登录，再进行评论。

类别

在 帮助中心 和 File Exchange 中查找有关 File Operations 的更多信息

标签

Community Treasure Hunt

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

Start Hunting!

Translated by