MATLAB: Issue with find()

arraydoublefindround

Hello,

I have something that seems to be quite a fundamental issue yet I am having issues fixing it:

I have an array containing various values, and I wish to find where in the array matches a value, which is caclulated eslewhere in the script.

Yet when using find() I'm getting some funny behaviour, as say I know one of the values is 0.65, and the variable containing 0.65 is called ratio_x, using

x = find(array == ratio_x)

returns an empty matrix, yet when using the value itself such as

x = find(array == 0.65)

it works fine.

Both array and ratio_x are saved as double, and array is rounded to 2 decimal places.

Any help would be appreciated.

Best Answer

It is not advised to compare floating point numbers using ==. You have to fix a small number (tolerance) say, eps = 10^-5 and check for is the absolute difference less then that. If so then the floating point numbers are close enough. '

Related Solutions

MATLAB: R2018b / R2019a runs very slow on macOS 10.14

There were instances of similar behavior in R2018b caused by the Magnet app. Turning it off should resolve the issue. Alternatively, you should be able to use keyboard shortcuts with no delays.

MATLAB: Issue with find() function

It’s called ‘floating point approximation error’. The concept is similar to the decimal approximation of 1/3 = 0.33333.... To get around it, you need to state the conditions in your find calls to allow for some imprecision.

Run these to get an idea of how the concept applies to your problem:

x = 0.1:0.05:0.6;
ix11 = find(x  <= 0.15, 1, 'last')
ix12 = find(x  >= 0.15, 1, 'first')
x1 = x([ix11  ix12])
ix21 = find(x <= 0.40, 1, 'last')
ix22 = find(x >= 0.40, 1, 'first')
x2 = x([ix21 ix22])

Best Answer

Related Solutions

MATLAB: R2018b / R2019a runs very slow on macOS 10.14

MATLAB: Issue with find() function

Related Question