selection of value in particular column

     79.0000   88.0000   1.0000   0   94.9457
    80.0000    87.0000   0        0    94.9107
    81.0000    99.0000   1.0000   0    94.9457
    82.0000    100.0000   0       0    94.9107
    83.0000    88.0000   1.0000   0    94.4731
    84.0000    98.0000   1.0000   0    94.9457
    85.0000    88.0000    1.0000  0    94.7606
    86.0000    88.0000     0      0    94.9107
    95.0000    97.0000    1.0000  0    92.0663
    96.0000   98.0000    1.0000   0     93.7677
    97.0000   100.0000    0       0     94.9107
    98.0000    100.0000   0      0      92.9658
    99.0000    100.0000  1.0000   0      90.4313

    i want to round off fifth column for example 90.43 as 90or 91 and i want to select values such that i may have more (90s,912s,94s,95s) ,i want t5he result such that those values (90s,912s,94s,95s)must not be more that 2 for ex(only 2 90s,91s)
    can u help please

Answer 1

[Daniel Shub]

The first part is easy

x = [
    79.0000   88.0000   1.0000   0   94.9457
    80.0000    87.0000   0        0    94.9107
    81.0000    99.0000   1.0000   0    94.9457
    82.0000    100.0000   0       0    94.9107
    83.0000    88.0000   1.0000   0    94.4731
    84.0000    98.0000   1.0000   0    94.9457
    85.0000    88.0000    1.0000  0    94.7606
    86.0000    88.0000     0      0    94.9107
    95.0000    97.0000    1.0000  0    92.0663
    96.0000   98.0000    1.0000   0     93.7677
    97.0000   100.0000    0       0     94.9107
    98.0000    100.0000   0      0      92.9658
    99.0000    100.0000  1.0000   0      90.4313
  ];

x(:, 5) = round(x(:, 5))

I have no idea what you want for the second part. What do you want to do with all the extra 95's?

Base on your comment you can discard all but the first two occurrences of each rounded number with:

y = cell2mat(arrayfun(@(c)x(find(x(:, 5) == c, 2, 'first'), :), unique(x(:, 5)), 'UniformOutput', false));

[FIR]

i want to ignore all those i needjust 2 values in each

[Daniel Shub]

See edit which ignores triplicates and beyond.