I know how to read in multiple Excel files, but am struggling to conduct the same analysis on all of those files. The analysis requires I average some values in different columns, then print those average values to a separate Excel sheet. I can do this with one Excel file, but have trouble figuring out how to print each average value in a different row in the output Excel file. Here is the code I have that works for one file (reads it, averages values in column 4, then prints to a separate Excel file):
data = xlsread('test_1.xlsx');
average_values_1 = data(:,4);
a = [average_values_1];
data_cells = num2cell(a);
column_header ={'Average Value 1'};
row_header(1,1) ={'File 1'}
output = [{' '} column_header; row_header data_cells];
xlswrite('Test Averages.xls', output);
How might I do this over and over again while printing values from each file in the output file as its own table? I suspect a nested loop is in my future.
Thanks in advance.
Here is working example of what you possibly want to do with xlswrite[1]:
filename = 'testdata.xlsx'; % Filename to save average values in
for k = 1:10 % Looping for 10 iterations
sheet = 2; % Selecting sheet2
Avg = randi([1 10],1,1); % Generating a random average each time the loop is run
xlRange = char(64+k); % 65 is the ASCII value of A
xlswrite(filename,Avg,sheet,xlRange); % Writing the excel file
end
This code gives the following output [2] :
Fig.1: Values are saved in a single row of excel file
If you want to get the output in a single column then use this xlRange = ['A',num2str(k)]; instead. It'll give you the following output [2] :
Fig.2: Values are saved in a single column of excel file
[1]: Read the documentation of xlswrite for more details.
[2]: Output values may vary since random integers are generated.
Related
I am having a simple problem while reading excel data which contains strings, long string, and numbers. Now I need to make each column (I have 11 here) to define separate variables of 1 column vector so that I can plot in MATLAB against each other or combination.
But the problem is the reading the file and creating 11 column vector. When I assign variable the header also comes.
Code:
%fid = fopen('Data_Link.xlsx');
[num,txt,raw] = xlsread('Data_Link.xlsx');
%fclose(fid);
% Extract data from readData
A = raw(:,1);
B = raw(:,2);
C = raw(:,6);
So I need the variables without header
Data file is truncated and given here.
Can anyone help me?
You can use readtable as ThP suggested. But if you want to use xlsread and you want your data without the header, you just need to remove the first row as in the below example:
%fid = fopen('Data_Link.xlsx');
[num,txt,raw] = xlsread('Data_Link.xlsx');
%fclose(fid);
% Extract data from readData
A = raw(2:end,1);
B = raw(2:end,2);
C = raw(2:end,6);
Note that each array will receive data from row 2 to last row.
You can use readtable instead of xlsread.
Using
T = readtable(‘Data_Link.xlsx’)
will result in a table with a variable for each column. For example T.Year would hold the values from the ‘Year’ column and T.Title would hold the values from the ‘Title’ column, etc.
I have a Matlab script with an output of a multidimensional array LCOE (3 dimensions) of size 16:12:34. This output needs to be written to Excel, therefore I use xlswrite.
I have tried this:
T = LCOE(:,:,1);
xlswrite('filename', T, 'sheetname', 'B2');
This does what it's supposed to, but only writes one table to excel, and I would like to write all 34 tables to excel underneath each other, spaced by `2 blank rows.
Then, I tried this:
for y = 1:34
T = LCOE(:,:,y)
xlswrite('filename', T, 'sheetname', strcat('B', num2str(2+(y-1)*18)));
This works, but is very slow, since matlab writes each table separately to excel. Is there a faster way to do this?
Instead of using xlswrite again and again. Dump all the values of a 3D matrix into a 2D matrix and add rows of NaNs so that when you write that to excel file, you get 2 blank rows.
Following code improves the execution time by the factor of more than 10.
LCOE = 100*rand(16,12,34); % Taking random values for LCOE
T = NaN(18*34-2 ,12); % 1. Pre-allocation 2. 16+2 = 18
% Following loop dumps all the values of 3D matrix into a 2-D followed by 2 rows of NaN
% to leave 2 blank rows in excel file.
for k = 1:34
T(18*(k-1)+[1:16], :) = LCOE(:,:,k);
end
xlswrite('filename', T, 'sheetname', 'B2'); % Writing the Excel file
In my system, my code takes about 1 second to execute while your code takes about 10.5 seconds. So that's a significant difference.
I have a .mat file which contains titles={'time','data'} and 2 column vectors:
time=[1;2;3;4;5] and data=[10;20;30;40;50].
I created a new cell called table={'time','data';time data} and i used:
xlswrite(filename,table);
However, when i open the xlsx file it shows me only the titles and not showing the numbers.
I saw that xlswrite will show empty cell in case im trying to export more than 1 number in a cell.
Is there anything i can do to export the whole vector instead of writing each value in it's cell?
The final result that i tried to get is like this:
time data
1 10
2 20
3 30
4 40
5 50
You have a couple options. Usually what I do is break it into two xlswrite calls, one for the header and one for the data.
titles = {'time','data'};
time = [1;2;3;4;5];
data = [10;20;30;40;50];
xlswrite('myfile.xlsx', titles, 'Sheet1', 'A1');
xlswrite('myfile.xlsx', [time, data], 'Sheet1', 'A2');
Alternatively, if you have R2013b or newer you can also use the table builtin, which has its own method for writing out data. With the same sample data:
mytable = table(time, data, 'VariableNames', titles);
writetable(mytable, 'myfile.xlsx');
I have a .mat file which contains titles={'time','data'} and 2 column vectors:
time=[1;2;3;4;5] and data=[10;20;30;40;50].
I created a new cell called table={'time','data';time data} and i used:
xlswrite(filename,table);
However, when i open the xlsx file it shows me only the titles and not showing the numbers.
I saw that xlswrite will show empty cell in case im trying to export more than 1 number in a cell.
Is there anything i can do to export the whole vector instead of writing each value in it's cell?
The final result that i tried to get is like this:
time data
1 10
2 20
3 30
4 40
5 50
You have a couple options. Usually what I do is break it into two xlswrite calls, one for the header and one for the data.
titles = {'time','data'};
time = [1;2;3;4;5];
data = [10;20;30;40;50];
xlswrite('myfile.xlsx', titles, 'Sheet1', 'A1');
xlswrite('myfile.xlsx', [time, data], 'Sheet1', 'A2');
Alternatively, if you have R2013b or newer you can also use the table builtin, which has its own method for writing out data. With the same sample data:
mytable = table(time, data, 'VariableNames', titles);
writetable(mytable, 'myfile.xlsx');
I have a directory on C drive containing a number of excel files of the same format. I would like to copy column H from each file into a new file using the following script I found online:
dirs=dir('C:\xxx\*.xlsx');
dircell=struct2cell(dirs);
filenames=dircell(1,:);
range = 'H:H';
n = (numel(filenames));
for i = 1:n;
Newfile(:,i) = xlsread(filenames{i},range);
end
This gives an error message of "Subscripted assignment dimension mismatch." with only one column extracted in the resulting file (Newfile).
I played around with the range and noticed that error occurs when xlsread reaches the end of the list of the first file and stops when the value is empty. My column H's have different number of filled values (i.e. file 1 has 20, file 2 has 100, file 3 has 3, etc.).
So, my question is whether it is possible to modify this script so that when it encounters an empty cell, either an empty cell or a NaN cell is extracted and most importantly that it will move on to the next column.
Thank you so much for your help in advance!
Not having Matlab at home I have to take it from the top of my head.
Since the column you read, H, has different number of valid entries you should not try to force them into the resulting array NewFile directly but rather use a temporary variable
dirs=dir('C:\xxx\*.xlsx');
dircell=struct2cell(dirs);
filenames=dircell(1,:);
range = 'H:H';
n = numel(filenames);
Newfile = NaN*ones(1, n);
for nf = 1:n;
tempVar = xlsread(filenames{nf},range);
r = size(NewFile,1); % get number of rows in NewFile
if length(tempVar) > r
% Make Newfile big enough to fit column nf
Newfile = [Newfile;NaN*ones(length(tempVar)-r,n)];
end
Newfile(:,nf) = tempVar;
end