I want my code to read from a csv file, the csv file (dee1) has 2 columns which the first one is Time_stamps and second one is load value . but the output of the function only read time_stamps. what should I do to fix it ?

function [load_time_stamp, load_value]=get_the_actual_demand_data()
filename = 'dee1.csv';
delimiter = ',';
startRow = 2;
formatSpec = '%{MM-dd-yy HH:mm}D%*s%[^\n\r]';
fileID = fopen(filename,'r');
dataArray = textscan(fileID, formatSpec, 'Delimiter', delimiter, 'HeaderLines' ,startRow-2);
a = datestr(dataArray{1,1},'yyyy-mm-dd HH:MM');
load_time_stamp = cellstr(a);
fclose(fileID);
formatSpec = '%*s%f%[^\n\r]';
fileID = fopen(filename,'r');
dataArray = textscan(fileID, formatSpec, 'Delimiter', delimiter, 'HeaderLines' ,startRow-2);
load_value = dataArray{:, 1};
fclose(fileID);
end

[Stephen23]

@javad amoli: today I formatted your code correctly for you. In future you can do it yourself: seledt the code text, click the {} Code button, and add a trailing empty line.

Why do you fopen and fclose the same file twice, each time with a different format string? Surely it would be a lot simpler and more efficient to read the file just once, and get the results from the cell array using indexing.

It would be much easier for us (and faster for you to get help) if you uploaded a sample file to work with: click the paperclip button.

[javad amoli]

thank you for your edit. I uploaded the file too. first column is time_stamp and it's string second one is load value and number.

[Stephen23]

"I uploaded the file too."

Nope. You uploaded a screen shot. A screen shot is useless to us, because we cannot load it. Please upload a sample file, as requested.

[javad amoli]

I have uploaded mfile before. now I uploaded CSV file too.

Answer 1

[dpb]

The format string doesn't match the record; particularly the datetime format...

>> fmt = '%{M/d/yyyy HH:mm}D %f';
>> dat = textscan(fid,fmt, 'Delimiter',',','HeaderLines',1);
>> dat
dat =
1×2 cell array
  [76660×1 datetime]    [76660×1 double]
>> dat{1}(1:4)
ans = 
4×1 datetime array
 1/23/2015 08:00
 1/24/2015 08:00
 1/25/2015 08:00
 1/26/2015 08:00
>>

You must pay attention to detail...

[dpb]

Just showed you what needs changing; doesn't help the learning process to rely on somebody else to do what you can do for yourself...what you do in the function specifically will be dependent upon how you want to treat the data later; if it were me I'd probably return a table but that would also depend on what is needed to be done with the data going forward.

>> t=table(dat{1},dat{2},'VariableNames',{'Time','Demand'});
>> whos t
Name          Size              Bytes  Class    Attributes
t         76660x2             1840998  table              
>> t(1:4,:)
ans = 
       Time          Demand
  _______________    ______
  1/23/2015 08:00    254.4 
  1/24/2015 08:00    259.2 
  1/25/2015 08:00    265.6 
  1/26/2015 08:00    267.2 
>>

Alternatively, return just the datetime and associated demand data as arrays of native types...

>> Time=dat{1};
   Demand=dat{2};
>> whos dt dmnd
  Name          Size              Bytes  Class       Attributes

    Demand    76660x1              613280  double                
    Time      76660x1             1226589  datetime              

>> Time(1:4)
Time = 
  10×1 datetime array
   1/23/2015 08:00
   1/24/2015 08:00
   1/25/2015 08:00
   1/26/2015 08:00
>> Demand(1:4)
Demand =
  254.4000
  259.2000
  265.6000
  267.2000
>>

ADDENDUM

And if decide to use table (highly recommended in general) then it's even simpler with start with that goal in mind--

>> t=readtable(filename,'ReadVariableNames',false,'Format',fmt);
>> t.Properties.VariableNames={'Time','Demand'};
>> whos t
Name          Size              Bytes  Class    Attributes
t         76661x2             1841022  table              
>> t(1:4,:)
ans = 
       Time          Demand
  _______________    ______
  1/22/2015 08:00    257.6 
  1/23/2015 08:00    254.4 
  1/24/2015 08:00    259.2 
  1/25/2015 08:00    265.6 
>>

This is simple enough really don't need a function; just including the few (four or five, roughly) lines of code in line in the main routine isn't much distraction but, if you do decide to do so, do NOT hardcode the name of the file inside the routine...putting such "magic numbers" or constants inside code defeats the entire purpose of having arguments available; you have to actually change the code itself to use a different file. Separate data from code and the input file is data.