/Data/corp/freq < infile > outfile
/Data/corp/freq /Data/corp/master.voc for 130M count
large count is grand.alpha
all input must be lower case, single column of words
program takes a while to load, give it time.
freq needs the vocab list to be sorted (how??)
defaults to large corpus count
How to make freq list??
pipe sort into countup
sort -n -r +1 < test > x; -n uses full numeric