# example:	gawk -f columnEntropy.awk Columns="-3-" data/subsets/coc81_all.csv.converted
# or		gawk -f columnEntropy.awk I=3 data/subsets/coc81_all.csv.converted

BEGIN {
	FS=OFS=","
	I=1;
	Columns="";
}

NR==1 {
	for(x=1; x<=NF; x++) {
		target="-"x"-";
		if (index(Columns, target)!=0) { I=x; }
	}
}

{
	Frequency[$I]++;
}
	
END {
	entropy=0;
	for (prob in Frequency) {
		p=Frequency[prob]/NR;
		entropy-=(p*log(p));
	}
	print entropy;
}