Monday, December 12, 2011

Normalize your data so that all data reside between 0 and 1

Problem:
----------
Very often, I have to do this sort of Normalization of data where the data should reside between say 0 and 1.

How to?
----------
Let's answer that by an example where I have to normalize the following vector.
(vector is a matrix with one column or one row! Sometime people call it row vector or column vector :D )


>> myvector=[ 5 7 20 9 10 5]


myvector =


     5     7    20     9    10     5

So to normalize we have to divide each value by the sum of the vector.


normalized_myvector=myvector/sum(myvector)


normalized_myvector =


    0.0893    0.1250    0.3571    0.1607    0.1786    0.0893


The normalized vector will always sum up to 1. To check that

>> sum(normalized_myvector)


ans =


     1

No comments:

Post a Comment

Please ask if anything is not clear enough..........