Counting the number of events and total number of blocks from a file.

I have results of my experiment on a .dat file (.text version is attached here with). It is obtained after doing experiments for several cycles (each cycle is named 'injection#1,injection#2...). I have a program which reads the data eliminating the 'injection' headers and other texts.

 PATH = 'E:\MyData\Experiments';
 fid=fopen(fullfile(PATH,'Test'));
 fgetl(fid) ;                                  
 % Skip 1st line (text describing the experiment).
 buffer = fread(fid, [1,Inf], '*char') ;
 fclose(fid) ; 
 % Eliminate 'injection' headers.
 buffer = regexprep( buffer, 'i.*?\n', '\n') ; 
 % Convert to numeric type.
 data = textscan(buffer, '%f %f %f') ;

 x = data{1} ;    
 y = data{2} ;
 time = data{3} ;
 data = [data{:}] ;

In this data, each row of number is an 'event'. How can I get the total number of injections and number of events in some injections (for example, number of events in injection 5, or total number of events between injection 5 to 20). I think, this should be done before eliminating the 'injection' headers in the program.

To make more clear, in the attached file there are 94 injections (total injections) and there are 29 events from injection#1 to injection#4 (injection#1: 12 events, injection#2: 6 events, injection#3: 4 events, injection#4: 7 events). As I have a lot of files with a lot of injections, it is difficult to look manually into each file. So, the program should read and give me the total number of injections and number of events between two specified injections.

Answer 1

[Cedric]

Hello Aneps, when I run the following code on your text document:

 content = fileread( 'Test.txt' ) ;
 blocks  = regexp( content, '#:\d+\s*([^i]+)', 'tokens' ) ;
 data    = cellfun( @(c) reshape(sscanf(c{1}, '%f'), 3, []).', blocks, ...
                    'UniformOutput', false ) ;
 counts  = cellfun( @(d) size(d, 1), data ) ;

I get a cell array of data, e.g.:

 >> data{2}
 ans =
         405         160      220515
         304         144      653785
         445         224      688895
         429         272      743333
         343         368      753373
         306         352      869463

which are all events associated with injection #2, and a numerical array of counts, which are the number of events for the 94 injections:

 >> counts
 counts =
  Columns 1 through 16
  12     6     4     7     7     7     7    12     8    12    11     7     9     7     7     5
  Columns 17 through 32
  10     6     8     9    16    13    14    14     9    14     8     5    10     9    17     9
  Columns 33 through 48
  6    11    11     9    11     5     9    12     8    14     8     8     7     9    11    10
  Columns 49 through 64
  7     6     9     5    10     8     4     8    10     8     9    10     8     6     8    13
  Columns 65 through 80
  11     4     9    11     7     6    13     7     5     3     7     3     9     8     9     4
  Columns 81 through 94
  5     8     6     9     7    11     5     7     7     7     6     8     8     5

i.e.

 >> counts(2)
 ans =
     6

indicates that there were 6 events associated with injection #2 (displayed above). I think that it is what you are looking for (?)

Note 1: this does the whole processing in a simpler way than my solution from 2013, so you don't need the former code anymore.

Note 2: you can concatenate the cell array data if you still want to have a large array with all data blocks merged, e.g. this way:

 >> allData = vertcat( data{:} ) ;

[aneps]

Thank you.