Replication factors must be a row vector of integers or integer scalars.

Replication factors must be a row vector of integers or integer scalars.

meanX = repmat(double(meanX), [nFilters/16, 1]);

Can anyone help me?

Answer 1

[Walter Roberson]

If nFilters was not an integer divisible by 16, then nFilters/16 would not be a positive integer. The number of repetitions you specify must be positive integers. For example,

meanX = repmat(double(meanX), [ceil(nFilters/16), 1]);

would be more likely to be a positive integer (but would still fail if nFilters was 0 or negative.)

[Asghar Ali]

nFilter is 175.

[Asghar Ali]

Thanks Walter, it works fine, but I get following error now: Non-singleton dimensions of the two input arrays must match each other.

rawValues = bsxfun(@minus, (filtX-bsxfun(@times, meanX, filtMean)) ./ stdX, whitening);

[Walter Roberson]

I suspect that your code relies upon the number of filters being a multiple of 16, such as would be the case if you had 176 filters instead of 175.

Answer 2

[David J. Mack]

nFilters/16 is probably not an integer. Depending on your needs, try round(nFilters/16) instead or use a number of filters which is an integer multiple of 16.

Greetings, David