How to exclude the characters row in a text file?

In the attached file i want to read only the numeric data of the file so that i can plot X Y. But i am facing problem that how to offset the row having strings. Someone help !

[Walter Roberson]

In the case of that particualar file, the data appears to be duplicated between the two sections, so it would appear to be fine to stop reading after the first set of numbers. Is that the general case, or is it necessary to read all of the number groups? If it is needed to read all of the number groups, should they be stored independently or as one matrix ?

[zainab malik]

Thank you for the comment. Yes its necessary to read all of the number grounds they might be different if I add more results. It's an example problem. They should be stored independently.. As, they are results from different jobs. Kindly help!

[zainab malik]

I am trying to use dlmread command it do read the values of first job but for second values(results) its giving error because of string at 84th row

Answer 1

[Walter Roberson]

The below code does not assume particular headers and does not assume that each group of numbers has the same number of columns, and does not assume empty space between groups, and does not assume any particular number of rows per group. It does, however, assume that each group has the same number of numeric columns within itself.

Output is a cell group_data with one entry per grouping, with the entry being a numeric array with multiple columns. Also output is a cell group_headers with one entry per grouping, with each entry being a cell array of column headers (blank delimited assumed) pulled from before the group. Under the circumstance that the file launches directly into numeric data, the header '(missing header)' will be used.

filename = 'abaqus.txt';

S = fileread(filename);
Slines = regexp(S, '(\r?\n)+', 'split');
mask = cellfun(@isempty, regexp(Slines, '^\s*\d', 'once'));
starts = strfind([1 mask], [1 0]);
stops = strfind([mask 1], [0 1]);
numgroups = length(starts);
group_headers = cell(numgroups, 1);
group_data = cell(numgroups, 1);
for G = 1 : numgroups
    grouplines = Slines(starts(G):stops(G));
    group_cols_cell = regexp( strtrim(grouplines), '\s+', 'split' );
    group_data{G} = str2double( vertcat(group_cols_cell{:}) );
    numcol = size(group_data{G},2);
    if starts(G) == 1
        group_headers{G} = {'(missing header)'};  %just in case no header
    else
        group_headers{G} = regexp(strtrim(Slines{starts(G)-1}), '\s+', 'split');
    end
end

[zainab malik]

Thankyou so much. It's working perfectly fine :)