def __init__(self):
# Heuristic object
self.heuristic = QueenHeuristic()
# States creator object
self.states_creator = StatesCreator()
# Max. number of iterations for searching a solution.
self.max_iter_count = 500
def __init__(self):
self.qh = QueenHeuristic()
class StatesCreator(object):
def __init__(self):
self.qh = QueenHeuristic()
def generate_random_start_state(self):
# Default state
state = [
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,0,0],
]
# Loop until a conflict is reached.
conf_found = False
while not conf_found:
# Populate columns
for j in range(0, 8):
rand_i = randint(0, 7) # Select random row in the column.
state[rand_i][j] = 1 # Place queen there.
# Check if there are any conflicts.
conf_c = self.qh.count_total_conflicts(state)
if conf_c > 0:
return state
def get_sample_start_state(self, num):
# lecture example 1
if num == 0:
state = [
[0,0,0,0,1,0,0,0],
[1,0,0,0,0,0,0,0],
[0,0,0,0,0,1,0,0],
[0,0,0,1,0,0,0,0],
[0,1,0,0,0,0,0,0],
[0,0,0,0,0,0,1,0],
[0,0,1,0,0,0,0,0],
[0,0,0,0,0,0,0,1],
]
conflicts_n = 1
# lecture example 2
elif num == 1:
state = [
[1,0,0,0,0,0,0,0],
[0,1,0,0,0,0,0,0],
[0,0,1,0,0,0,0,0],
[0,0,0,1,0,0,0,0],
[0,0,0,0,1,0,0,0],
[0,0,0,0,0,1,0,0],
[0,0,0,0,0,0,1,0],
[0,0,0,0,0,0,0,1],
]
conflicts_n = 28
# https://thewalnut.io/app/release/14/#time=9
elif num == 2:
state = [
[0,0,0,0,0,1,0,0],
[0,0,0,1,0,0,0,0],
[0,1,0,0,0,0,0,0],
[0,0,0,0,0,0,1,0],
[0,0,1,0,1,0,0,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,0,1],
[1,0,0,0,0,0,0,0],
]
conflicts_n = 1
# http://www.aiai.ed.ac.uk/~gwickler/eightqueens.html
elif num == 3:
state = [
[1,0,0,0,0,0,0,0],
[0,0,0,0,1,0,0,0],
[0,1,0,0,0,0,0,0],
[0,0,0,0,0,1,0,0],
[0,0,1,0,0,0,0,0],
[0,0,0,0,0,0,1,0],
[0,0,0,1,0,0,0,0],
[0,0,0,0,0,0,0,1],
]
conflicts_n = 1
elif num == 4:
state = [
[1,1,0,0,0,0,0,0],
[0,0,1,0,0,0,0,0],
[0,0,0,1,0,0,0,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,0,0,1,0],
[0,0,0,0,0,0,0,0],
[0,0,0,0,1,1,0,0],
[0,0,0,0,0,0,0,1],
]
conflicts_n = 1
return (state, conflicts_n)
def get_sample_final_state(self, number):
conflicts_n = 0
if number == 1:
state = [
[1,0,0,0,0,0,0,0],
[0,0,0,0,1,0,0,0],
[0,0,0,0,0,0,0,1],
[0,0,0,0,0,1,0,0],
[0,0,1,0,0,0,0,0],
[0,0,0,0,0,0,1,0],
[0,1,0,0,0,0,0,0],
[0,0,0,1,0,0,0,0],
]
elif number == 2:
state = [
[0,0,0,0,1,0,0,0],
[0,1,0,0,0,0,0,0],
[0,0,0,1,0,0,0,0],
[0,0,0,0,0,0,1,0],
[0,0,1,0,0,0,0,0],
[0,0,0,0,0,0,0,1],
[0,0,0,0,0,1,0,0],
[1,0,0,0,0,0,0,0],
]
elif number == 3:
state = [
[0,0,0,1,0,0,0,0],
[0,0,0,0,0,0,1,0],
[0,0,0,0,1,0,0,0],
[0,0,1,0,0,1,0,0],
[1,0,0,0,0,0,0,0],
[0,0,0,0,0,1,0,0],
[0,0,0,0,0,0,0,1],
[0,1,0,1,0,0,0,0],
]
return (state, conflicts_n)
def get_sample_inter_state(self, number):
if number == 1:
state = [
[0,0,0,0,1,0,0,0],
[0,1,0,0,0,0,0,0],
[0,0,0,1,0,0,0,1],
[0,0,0,0,0,0,1,0],
[0,0,1,0,0,0,0,0],
[0,0,0,0,0,0,0,1],
[0,0,0,0,0,1,0,0],
[1,0,0,0,0,0,0,0],
]
conflicts_n = 3
elif number == 2:
state = [
[1, 0, 0, 0, 1, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 1, 0, 0],
[0, 0, 0, 1, 0, 0, 0, 0],
[0, 1, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 1, 0],
[0, 0, 1, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 1]
]
conflicts_n = 3
return (state, conflicts_n)
class ChessQueenWorld(object):
def __init__(self):
# Heuristic object
self.heuristic = QueenHeuristic()
# States creator object
self.states_creator = StatesCreator()
# Max. number of iterations for searching a solution.
self.max_iter_count = 500
def solve_random_board(self):
gen_state = self.states_creator.generate_random_start_state()
(sol_path, total_steps) = self.solve_given_board(gen_state)
solution = sol_path if sol_path else False
return solution, gen_state, total_steps
def solve_sample_board(self, num):
gen_state = self.states_creator.get_sample_start_state(num)[0]
(sol_path, total_steps) = self.solve_given_board(gen_state)
#self._show_solution(sol_path)
def solve_given_board(self, start_state):
# Prepare variables
solution_path = []
current_state = start_state
conf_count = self.heuristic.count_total_conflicts(current_state)
solution_path.append(current_state)
total_steps = 0
# Show board
self._show_board(start_state)
print('Number of conflicts: %d') % conf_count
# Repeat until there are no conflicts.
iter_n = 0
while conf_count > 0:
# Check n. of performed iterations
if iter_n > self.max_iter_count:
return False, 0
iter_n += 1
#print ('========iter. %d========') % iter_n
# Create a new state from current state.
(new_state, performed_steps) = self.heuristic.choose_min_conflict_positions(current_state)
# Check if the new state is alredy in solution path.
if new_state in solution_path:
continue # If yes, re-run the method.
# If not, it's part of solution.
conf_count = self.heuristic.count_total_conflicts(new_state)
solution_path.append(new_state)
current_state = new_state
total_steps += len(performed_steps)
# Show solution
print # blank line
if solution_path:
print('>>>Solved in %d steps.') % total_steps
self._show_board(solution_path[-1])
else:
print('<<<NOT solved after %d iterations.') % self.max_iter_count
# result
return solution_path, total_steps
def bulk_solve(self, boards_n, csv = False):
# Prepare variables for statistics
steps_sum = 0.0
solved_n = 0
unsolved_boards = {}
solved_stats = {} # key is number of conflicts in start board
# Generate and solve some boards
for i in range(1, boards_n+1):
print('======Board n. %d======') % i
# Get and solve random board.
(sol_path, start_state, total_steps) = self.solve_random_board()
conf_count = self.heuristic.count_total_conflicts(start_state)
#self._show_board(start_state)
# Check if a solution was found.
if sol_path:
# Increment common values.
solved_n += 1
steps_sum += total_steps
# Save values for initial conflicts count.
if conf_count in solved_stats:
solved_stats[conf_count][0] += 1
solved_stats[conf_count][1] += total_steps
else:
solved_stats[conf_count] = []
solved_stats[conf_count].append(1.0)
solved_stats[conf_count].append(total_steps)
else:
if conf_count in unsolved_boards:
unsolved_boards[conf_count].append(start_state)
else:
unsolved_boards[conf_count] = [start_state]
# blank line
print
# Show basic statistics
avg_steps_c = round(steps_sum / boards_n, 2)
percent_solved = (float(solved_n) / boards_n) * 100
print('======BASIC STATS======')
print('Number of generated boards: %d') % boards_n
print('Number of solved boards: %d ... ' + str(round(percent_solved, 2)) + ' %%') % solved_n
print('Average number of steps for solution: ' + str(avg_steps_c))
# Show detailed statistics
print('======DETAILED STATS======')
for conf_count, (solved_count, steps_sum) in sorted(solved_stats.iteritems()):
avg_steps_c = round(steps_sum / solved_count, 2)
total_boards = solved_count + len(unsolved_boards.get(solved_count, []))
percent_solved = (float(solved_count) / total_boards) * 100
if csv:
print str(conf_count)+','+str(round(percent_solved, 2))+','+str(avg_steps_c)
else:
print('===%d conflicts===') % conf_count
print('Generated boards: %d') % total_boards
print('Solved boards: %d ... ' + str(round(percent_solved, 2)) + ' %%') % solved_count
print('Average steps: ' + str(avg_steps_c))
def _show_board(self, state):
for row in state:
print row
def _show_solution(self, solution_path):
print('>>>>Solution<<<<')
if not solution_path:
exit('No solution found.')
print('Number of iterations: %d') % (len(solution_path) - 1)
print('Final state: ')
self._show_board(solution_path[-1])
print('{n. of conflists: %d}') % self.heuristic.count_total_conflicts(solution_path[-1])
