Prev: Fitting to Gaussian Distribution
Next: DAQ parallel processing
From: ronnie on 17 Jun 2010 09:16
Hello, I have a problem with grouping text data together.

I have a text file as such:

date time n numval numval numval numval
-, -, 8, 1.123 1, 23, .214
-, -, 8, 1.512 1, 45, .435
-, -, 8, 1.133 1, 32, .522
-, -, 14, 6.421 0, 0, 0
-, -, 8, 1.233 1, 43, .456
-, -, 8, 1.455 1, 47, .656
-, -, 6, 4.512 0, 0, 0
-, -, 8, 1.552 1, 31, .253
-, -, 8, 1.488 1, 45, .349

I only want to take the rows that have 8's for the n values. (How do I do this in matlab)?

The real data looks similar to this but it is many rows longer. Also the dates are actual dates and the times are actual times I just didn't write each date and time that they actually were.

Help is much appreciated.
From: us on 17 Jun 2010 13:30
ronnie <asdfessssy(a)aol.com> wrote in message <1689549364.362686.1276795012661.JavaMail.root(a)gallium.mathforum.org>...
> Hello, I have a problem with grouping text data together.
>
> I have a text file as such:
>
> date time n numval numval numval numval
> -, -, 8, 1.123 1, 23, .214
> -, -, 8, 1.512 1, 45, .435
> -, -, 8, 1.133 1, 32, .522
> -, -, 14, 6.421 0, 0, 0
> -, -, 8, 1.233 1, 43, .456
> -, -, 8, 1.455 1, 47, .656
> -, -, 6, 4.512 0, 0, 0
> -, -, 8, 1.552 1, 31, .253
> -, -, 8, 1.488 1, 45, .349
>
> I only want to take the rows that have 8's for the n values. (How do I do this in matlab)?
>
> The real data looks similar to this but it is many rows longer. Also the dates are actual dates and the times are actual times I just didn't write each date and time that they actually were...

....but you must - otherwise people end up in a lot of tedious, time-consuming second guessing...
now, show a couple of complete lines...

us
From: dpb on 17 Jun 2010 13:31
ronnie wrote:
....
> I have a text file as such:
>
> date time n numval numval numval numval
> -, -, 8, 1.123 1, 23, .214
> -, -, 8, 1.512 1, 45, .435
> -, -, 8, 1.133 1, 32, .522
> -, -, 14, 6.421 0, 0, 0
> -, -, 8, 1.233 1, 43, .456
> -, -, 8, 1.455 1, 47, .656
> -, -, 6, 4.512 0, 0, 0
> -, -, 8, 1.552 1, 31, .253
> -, -, 8, 1.488 1, 45, .349
>
> I only want to take the rows that have 8's for the n values. (How do
> I do this in matlab)?
....

read the data into an array (let's call it x) w/ textscan()

to use data for those points where n (column 3) == 8

y=x(x(:,3)==8,:);

would return an array y with only such lines.

or, alternatively, you could compress x by

x(x(:,3)==8,:) = [];

or, you could use find() to save the row indices amongst the various
possibilities...

read the "Getting Started" section on data manipulation, arrays, etc., ...

--
From: ronnie on 17 Jun 2010 11:53
Lets refine the question and say that there is text data on top of the column headers of the text file. What I mean is: there is the data that I posted and then there is test data on top of that for 3 lines in the text file.
It says:

Peter Smith
revolutions and rotations
degrees to the normal; file

Then it has the data that I posted. (I tried making the code work when I copied all of the data into another text file.) It worked but I can't figure out how to tell the program to exclude all of the text data at the top of the file.

Thanks in advance for help,
Ronnie C
From: dpb on 17 Jun 2010 16:15
ronnie wrote:
> Lets refine the question and say that there is text data on top of
> the column headers of the text file. ...
> Then it has the data that I posted. (I tried making the code work
> when I copied all of the data into another text file.) It worked but
> I can't figure out how to tell the program to exclude all of the text
> data at the top of the file.
....

Use the optional 'headerlines' keyword in textscan() or textread() when
reading the data

doc textscan

--
 | 
Pages: 1
Prev: Fitting to Gaussian Distribution
Next: DAQ parallel processing