how to extract this data?

Anyone,

I want to extract the data in the end row.

The meaningful data comes every "0" changes in the first row...

Best,

Pen

[dpb]

Paste the data as text rather than as an image if you expect somebody to actually do anything other than "air code"

[Pengju]

I am sorry for that.

Answer 1

[Star Strider]

One option:

fidi = fopen('your_text_file_name.txt','r');
Data = textscan(fidi, '%f%f%f', 'HeaderLines',4, 'CollectOutput',1);
fclose(fidi);

You might have to add an 'EndOfLine' argument if you only get a (1x1) cell from the previous code:

fidi = fopen('your_text_file_name.txt','r');
Data = textscan(fidi, '%f%f%f', 'HeaderLines',4, 'CollectOutput',1, 'EndOfLine','\r\n');
fclose(fidi);

You can then parse the columns as you wish. Use a cell array. Find the zeros in the first column using the find function:

DataD = cell2mat(Data);
Zeros1 = [find(DataD(:,1) == 0); size(DataD,1)];
for k1 = 1:length(Zeros1)
    Record{k1} = DataD(Zeros1(k1):Zeros1(k1+1)-1,:);
end

Note — This is UNTESTED CODE, since I do not have your file to test it with.

[Pengju]

Thanks a lot.

But: ??? Attempted to access Zeros1(100002); index out of bounds because numel(Zeros1)=100001.

Error in ==> read_20160204_2 at 12 Record{k1} = DataD(Zeros1(k1):Zeros1(k1+1)-1,:);

[Star Strider]

My pleasure.

The loop should be:

for k1 = 1:length(Zeros1)-1
    Record{k1} = DataD(Zeros1(k1):Zeros1(k1+1)-1,:);
end

Error on my part. This new version should work.

[Pengju]

Thanks a lot,

I may give you the data file. I want put all the data in the red squares in one array...

[Pengju]

[Pengju]

[Star Strider]

This code works correctly with your file:

fidi = fopen('test.txt','r');
Data = textscan(fidi, '%f%f%f', 'HeaderLines',4, 'CollectOutput',1);
fclose(fidi);
DataD = cell2mat(Data);
Zeros1 = unique([find(DataD(:,1) == 0); size(DataD,1)]);
for k1 = 1:length(Zeros1)-1
    Record{k1} = DataD(Zeros1(k1):Zeros1(k1+1),:);
end
Output = [Record{1}(2:end-1,3); Record{2}(2:end-1,3); Record{3}(2:end-1,3)];

I had to work to edit your post so I could see your image, but I managed to do it.

The ‘Output’ array has the data in the red squares.

[Pengju]

Almost final. For the data in the red square, I have 100000 cells, the same number of k1.

So, actually, I want read all of those data into the same output array.

[Star Strider]

I only have the data you provided, and I have demonstrated how to read your file and extract the data you want.

I leave it to you to do the rest.

[Pengju]

Thanks a lot

[Star Strider]

My pleasure.

If my Answer solved your problem as you stated it in your Question, please Accept it.

Answer 2

[dpb]

As noted above, w/o a file to test, something like

fid=fopen('yourfile');    % open file
d=cell2mat(textscanf(fid,'','headerlines',4)); % read numeric data (Nx3 array)
fid=fclose(fid);                   % done w/ file
ix=[find(d(:,1)==0);length(d)+1];  % locations of segment breaks in data plus end
for i=1:length(ix)-1          % over the number of breaks
  data{i}=d(ix(i):ix(i+1)),3); % save each group as a cell array
end

Do whatever with the contents of data for each cell as desired.

[Pengju]

??? Index exceeds matrix dimensions.

Error in ==> read_20160204_1 at 11 data{i}=d(ix(i):ix(i+1),3); % save each group as a cell array