How can I solve a dlmread error on A2016a?

I have been having an issue with producing a set of CTD data on Matlab for a few days and don't seem to understand how to solve this. I feel as though it may be an issue with importing data but I'm not 100% sure.

data=dlmread([name,'.cnv'],'',96,0); 
Error using dlmread (line 147)
Mismatch between file and format character vector.
Trouble reading 'Numeric' field from file (row number 1, field number 1) ==> #   </sensor>\n

Error in ctd1 (line 4)
data=dlmread([name,'.cnv'],'',96,0); 

These are the error suggestions that come up, but not sure what they're suggesting as I'm quite new to Matlab and my understanding is still limited.

How can I solve the dlmread issue?

Is this issue related to importing data into Matlab?

Answer 1

[dpb]

dlmread will read ONLY numeric data; the message is telling you there's alphanumeric data in the section of the file you're trying to read. So, "Yes, Virginia, there is a problem in importing data into Matlab", but the problem is you've chosen the wrong tool for the job here.

There are many tools for the purpose in Matlab; granted, the "veritable plethora" of features is one of the things that makes initial starting difficult because it is hard to know which to choose.

Here, probably either readtable or textscan will be the choice; which, precisely we'd need to see the data file content to be able to say with certainty.

Attach a file w/ the paperclip icon so somebody here can have a look (and then tell us what you're trying to import from the file if not the whole thing as it appears).

[Walter Roberson]

dlmread will read ONLY numeric data

By the time of the R2016a release that the user is using, dlmread() is able to handle text existing in header rows or header columns.

the message is telling you there's alphanumeric data in the section of the file you're trying to read

The key there is "section of hte file you're trying to read". The 96 lines that the user requested to be skipped do not count towards being portions of the files that the user is trying to read.

[Willem Swann]

% SCRIPT 1:Reading, processing and presenting CTD data                   
clear all;close all;clc;
name='ST1'
data=dlmread([name,'.cnv'],'',96,0);
 
depth=data(:,1);                % depth vector 
temp=data(:,2);                 % temperature vector 
salt=data(:,3);                 % salinity vector 

figure;
plot(temp,depth)
set(gca,'Ydir','reverse')
xlabel('Temperature (deg C)')
ylabel('Depth(m)')
 
bin=0.5;           % thickness of the averaging bin (m)
Dbins=max(depth); % definition of the maximum depth
bins=[0.5:bin:Dbins];     % setting the bin’s vector
 
for i=1:length(bins)-1
   ind=find(depth>=bins(i)&depth<bins(i+1));
   % returns the vectors ind with the numbers of rows  
   % within the depth range from bins(i) to bins(i+1)
   Toutput(i)=mean(temp(ind));
   % calculates an average temperature Toutput(i) 
                              % in the i-th bin
end
 
depthbin=[0.5+bin/2:bin:Dbins-bin/2];
hold on    
plot(Toutput,depthbin,'r')
xlabel('Temperature(grad C)')
ylabel('Depth(m)')
 
legend('Temperature: Original data','Temperature: Smoothed data','location','southwest')

Here is the script I am trying to run with the error that has come up as seen before.

[Willem Swann]

The data I am trying to import is CTD (conductivity, temperature, salinity) taken from a local tidal estuary. This is under st1 and through to st7.

[dpb]

Attach the file...we can only guess.

Walter had the very good thought that the line count is off by one row...

Alternatively, in lieu of receiving the data we need to solve the problem, try

opt=detectImportOptions('yourfilenamehere');
data=readtable('yourfilenamehere',opt);

and see if joy ensues. If the file structure isn't quite regular before it gets to your data, however, it'll probably require human intervention. But the humans here other than you don't have the info they require...

Answer 2

[Walter Roberson]

You need to increase your 96 to (at least) 97.