cat input.txt
sun clo clo sun clo clo sun sun sun clo sun clo
####
analyse_text.pl --input-corpus input.txt --ngram-length 2 --output-state mystate --separator ' '
####
{
"counts" => { "clo|clo" => 2, "clo|sun" => 3, "sun|clo" => 4, "sun|sun" => 2 },
"cum-twisted-dist" => {
clo => ["clo", 0.4, "sun", 1],
sun => ["clo", 0.666666666666667, "sun", 1],
},
"dist" => {
"clo|clo" => 0.181818181818182,
"clo|sun" => 0.272727272727273,
"sun|clo" => 0.363636363636364,
"sun|sun" => 0.181818181818182,
},
"N" => 2,
}
####
predict_text.pl --input-state mystate --separator ' ' --num-iterations 30 --seed 'clo'
####
/home/andreas/usr/bin/predict_text.pl : starting with seed 'clo' ...
clo sun clo clo sun clo sun clo clo clo clo sun
sun sun clo clo sun clo sun clo sun sun clo sun
sun sun clo clo sun clo sun sun sun clo sun clo
sun clo clo sun clo clo clo sun clo clo sun sun
sun clo clo sun clo clo sun clo sun clo clo clo
clo sun sun clo sun sun clo sun clo sun sun clo
sun sun clo sun sun clo sun clo sun sun clo sun
clo sun sun sun sun clo sun clo clo sun clo clo
clo sun sun clo sun