Textscan, dlmread, and importdata all not working

I have a very simple text file comprised of 1's and 0's. I have tried to read it into MATLAB with every function I can find, but nonetheless I get an error of some form. dlmread seems the most promising, but it throws a "Mismatch between file and format character vector." error. textscan did not work as it created a 1X1 cell with all of the data in it, and trying A = A{:} also did not work, similar error with importdata. Any help would be much appreciated.

[Adam Danz]

Could you attach the text file or a watered down version of the text file if it's huge? The line of code that calls dlmread() would be helpful, too.

[Skyler Castillo-Wilson]

y = dlmread('FakeData.txt');

[Star Strider]

‘... trying A = A{:} also did not work ...’

Try using the cell2mat function, or alternatively:

[A{:}]

[Skyler Castillo-Wilson]

tried the cell2mat function, all of the data ends up in one cell for some reason so it returns a [] matrix.

[Stephen23]

" textscan did not work as it created a 1X1 cell with all of the data in it"

Why do you think that having the data in one cell is a problem? What is important is how you called textcan and the content of the cell. But unless you give us this information we will just be guessing...

[Skyler Castillo-Wilson]

Sorry, the goal is to delineate the data so that each value of 1 or 0 is parsed into its own cell. I plan on for looping through each individual value, and I don't think thats possible if its all in one cell (if it is that would be amazing).

Answer 1

[Adam Danz]

See how this works for you,

fid = fopen('FakeData.txt'); 
t = textscan(fid, '%s'); 
fclose(fid)
t2 = cell2mat(cellfun(@str2double, t, 'uniformoutput', false));

[Skyler Castillo-Wilson]

BANG! Thank you! I did not realize that the file was being read in as a string, so my formatspec operator was set to look for numbers. Thanks for the help

[Paolo]

Why not simply

 t2 = str2double(t{:})

?

[Adam Danz]

Note that your first and last character in the FakeData text file has an additional strange character. When textscan reads this it replaces that strange character with a space. So you'll need to remove white space in 't' before converting it to double in 't2'. Otherwise you'll have NaN for those character.

[Adam Danz]

Yeah, that cell2mat() call can definitely be simplified as Star Strider and Paolo point out.

[Skyler Castillo-Wilson]

I see that NaN error, but the text file has no such weird character, merely 1's and 0's -- see attached text file

[Image Analyst]

If I click on it in the Firefox browser, it shows that the first and next to the last character are weird - not a 0 nor a 1.

[Adam Danz]

Yeah, my downloaded txt file also has those weird chars.

[Skyler Castillo-Wilson]

Update for those who helped: Naturally, the only issue was compatibility between what a number means to my PC keyboard vs. my MacBook Pro. That messed with the text file just enough to ruin all of those functions. Problem resolved!