/* 

  Animal population in Picat.

  Cf my Gamble model gamble_animal_population.rkt

  This program was created by Hakan Kjellerstrand, hakank@gmail.com
  See also my Picat page: http://www.hakank.org/picat/

*/

import ppl_distributions, ppl_utils.
import util.

main => go.

/*
  Model 1

  var : mice
  Probabilities (truncated):
  5: 0.2569000000000000
  6: 0.2031000000000000
  4: 0.1992000000000000
  3: 0.1232000000000000
  .........
  1: 0.0105000000000000
  9: 0.0100000000000000
  10: 0.0008000000000000
  0: 0.0008000000000000
  mean = 4.9884

  var : p
  Probabilities:
  false: 0.9320000000000001
  true: 0.0680000000000000
  mean = [false = 0.932,true = 0.068]

  var : rabbits
  Probabilities:
  2: 0.3041000000000000
  1: 0.2738000000000000
  3: 0.1944000000000000
  0: 0.1063000000000000
  4: 0.0886000000000000
  5: 0.0272000000000000
  6: 0.0048000000000000
  7: 0.0007000000000000
  8: 0.0001000000000000
  mean = 1.9901

  var : voles
  Probabilities:
  3: 0.2714000000000000
  2: 0.2231000000000000
  4: 0.2040000000000000
  1: 0.1179000000000000
  5: 0.1058000000000000
  6: 0.0387000000000000
  0: 0.0299000000000000
  7: 0.0079000000000000
  8: 0.0010000000000000
  9: 0.0003000000000000
  mean = 3.0215

*/
go ?=>
  reset_store,
  run_model(10_000,$model,[show_probs_trunc,mean]),
  nl,
  % show_store_lengths,nl,
  % fail,
  nl.
go => true.

model() =>
  AllAnimals = ones(30,voles) ++ ones(50,mice) ++ ones(20,rabbits),

  Animals = resample(10,AllAnimals),

  Voles = count_occurrences(voles,Animals),
  Mice = count_occurrences(mice,Animals),
  Rabbits = count_occurrences(rabbits,Animals),  

  P = check( (Voles == 4, Mice == 5, Rabbits == 1)),
  
  add("voles",Voles),
  add("mice",Mice),
  add("rabbits",Rabbits),
  add("p",P).


/*
  Model 2:
  Same as model 1 but using categorical_n/3 instead of resample, and
  using get_freq/1 for the occurrences.

  var : mice
  Probabilities (truncated):
  5: 0.2444000000000000
  6: 0.2049000000000000
  4: 0.2011000000000000
  3: 0.1244000000000000
  .........
  9: 0.0097000000000000
  1: 0.0097000000000000
  0: 0.0011000000000000
  10: 0.0006000000000000
  mean = 4.9798

  var : p
  Probabilities:
  false: 0.9357000000000000
  true: 0.0643000000000000
  mean = [false = 0.9357,true = 0.0643]

  var : rabbits
  Probabilities:
  2: 0.3012000000000000
  1: 0.2713000000000000
  3: 0.1964000000000000
  0: 0.1087000000000000
  4: 0.0884000000000000
  5: 0.0280000000000000
  6: 0.0054000000000000
  7: 0.0006000000000000
  mean = 1.9931

  var : voles
  Probabilities:
  3: 0.2633000000000000
  2: 0.2260000000000000
  4: 0.2057000000000000
  1: 0.1200000000000000
  5: 0.1079000000000000
  6: 0.0378000000000000
  0: 0.0286000000000000
  7: 0.0096000000000000
  8: 0.0010000000000000
  9: 0.0001000000000000
  mean = 3.0271


*/ 
go2 ?=>
  reset_store,
  run_model(10_000,$model2,[show_probs_trunc,mean]),
  nl,
  % show_store_lengths,nl,
  % fail,
  nl.
go2 => true.

  
model2() =>
  Animals = categorical_n([30,50,20],[voles,mice,rabbits],10),
  Freq    = Animals.get_freq(),
  Voles   = Freq.get(voles,0),
  Mice    = Freq.get(mice,0), 
  Rabbits = Freq.get(rabbits,0),

  P = check( (Voles == 4, Mice == 5, Rabbits == 1)),
  
  add("voles",Voles),
  add("mice",Mice),
  add("rabbits",Rabbits),
  add("p",P).