Question

I have my sparse features stored in a text file in the following format (ArrayIndex: Value). Currently I am parsing the text using reg-expressions and converting this to a matlab array. What I wanted to know was, is there a faster/better more MATLAB-ish approach to convert this format of data into a matlab array.

 2402:0.099061 2404:0.136546 2406:0.447161 2407:0.126333 2408:0.213803 2411:0.068189 2416:0.223526 2417:0.090420

Answer 1

EDIT:

You can read and parse the file using TEXTSCAN, then the build sparse matrix from those values:

fid = fopen('input.txt');
C = textscan(fid, '%f:%f');
fclose(fid);
C = sparse(1,C{1},C{2});

result:

>> C
C =
   (1,2402)     0.099061
   (1,2404)      0.13655
   (1,2406)      0.44716
   (1,2407)      0.12633
   (1,2408)       0.2138
   (1,2411)     0.068189
   (1,2416)      0.22353
   (1,2417)      0.09042