CODEWITHSUNDEEP
rmatrixrow
user2928318
user2928318

Reputation: 549

Normalize rows of a matrix within range 0 and 1

I am trying to normalize all rows of my matrix data at once within range 0 and 1. But I don't know how to do it.. For example, I want to normalize each "obs1", "obs2", "obs3". Thus, minimum, maximum, and sum of each "obs1", "obs2", "obs3" will be used. My data format is,

`mydata

             a        b        c         d         e
obs1   8.15609  11.5379  11.1401   8.95186   7.95722
obs2 339.89800 856.3470 691.3490 590.28600 543.67200
obs3   2.12776  46.4561 136.8860 118.09100 119.86400

`

Also, When I searched to perform this, people used "function()". When/for what does this used?

Thank you very much for your help in advance! :)

Upvotes: 22

Views: 39997

Answers (3)

CsBence
CsBence

Reputation: 161

You could use the apply with rescale as the following:

apply(mydata, 1, rescale)

where the second argument 1 tells apply to work with rows.

The default range is [0, 1] but a custom range can be specified with the to argument that will be forwarded to the rescale function:

apply(mydata, 1, rescale, to=c(1,2))

Dependecy:

if(!require(scales)){
    install.packages("scales", dependencies=TRUE)
    library(scales)
}

Upvotes: 5

Sushant Kulkarni
Sushant Kulkarni

Reputation: 467

for(i in 2:length(mydata[1,])){

    mydata[,i] <- prop.table(mydata[,i])

}

Normalized matrix will be updated in mydata

Upvotes: 0

user1981275
user1981275

Reputation: 13372

To normalize for each row, you can use apply and then subtract the minimum from each column and divide by the difference between maximum and minimum:

t(apply(mydata, 1, function(x)(x-min(x))/(max(x)-min(x))))

gives you 

              a         b         c         d         e
obs1 0.05553973 1.0000000 0.8889038 0.2777796 0.0000000
obs2 0.00000000 1.0000000 0.6805144 0.4848262 0.3945675
obs3 0.00000000 0.3289472 1.0000000 0.8605280 0.8736849

What happens is that you apply the function 

function(x){
   (x-min(x))/(max(x)-min(x))
}

to each row of your data frame.

Upvotes: 30

Related Questions