# type/token ratio lemmat
cat $1| \
awk ' \
BEGIN {
        poc=0;ruz=0;
      }
{
 c=$2;
 p=match(c,/\>/);
 le=substr(c,p+1);
 
 if ($2!="") 
     { sl[le]++; poc++;}



}
END {
      for (a in sl) 
       {
#         printf "%s %i %f\n",a,sl[a],sl[a]/poc;
         ruz++;
       }
      printf "Pomer type/token na lemmatech %i %i %f \n",poc,ruz,poc/ruz;
      printf "Logaritmizovany pomer %f \n",log(poc)/(log(10)*log(ruz));
    }
'

