binning of data

[rkdash]

Hello,

I have a ascii table that looks like this:

84.575 1.775
84.593 1.773
84.618 1.768
84.642 1.768
84.666 1.763
84.685 1.761
84.696 1.759

Th first column ranges from 0 to 120 and the second column from 0 to 4. I want to "bin" the first column (within 2 units) and calculate the average of the data from the second column that fall within the "bin".

For example, I want the bins (from first column) to be like [0-2], [2-4], [4-6]..., [118-120] with bin centers 1, 3, 5, ..., 119. And average of the data from second column falling under the bins.

So, my table should be like this:

1 ?
3 ?
5 ?
....
....
119 ?



Thanks

 

[p5wizard]

Try this:

awk '{
 tot[int($1/2)+1]+=$2
 num[int($1/2)+1]++
 }
END {
 for (i=1;i<=119;i++)
  print i, num[i] ? tot[i]/num[i] : 0
}' /path/to/your/file

HTH,

p5wizard

 

[vgersh99]

p5wizard,
your 'binning' is not exactly correct. Here's something better [I hope].
nawk -f rk.awk myFile

rk.awk:

{
 bin= (($1%2) < 1) ? int($1) + 1 : int($1)
 tot[bin]+=$2
 num[bin]++
 }
END {
 for (i=1;i<=119;i+=2)
  print i, num[i] ? tot[i]/num[i] : 0
}

vlad
+----------------------------+
| #include<disclaimer.h> |
+----------------------------+

 

[rkdash]

Thank you both !!