def compare_rows_or_cols(scores, init_set, solution_set, solutions):
abc_pixel_count = algorithm.fill_ratio(init_set[0]) + algorithm.fill_ratio(init_set[1]) \
+ algorithm.fill_ratio(init_set[2])
abc_shape_count = algorithm.find_regions(init_set[0]) + algorithm.find_regions(init_set[1]) \
+ algorithm.find_regions(init_set[2])
gh_pixel_count = algorithm.fill_ratio(solution_set[0]) + algorithm.fill_ratio(solution_set[1])
gh_shape_count = algorithm.find_regions(solution_set[0]) + algorithm.find_regions(solution_set[1])
possible_answers = []
for i, score in enumerate(scores):
if score != 0.0:
possible_answers.append( (i, solutions[i]))
comparisons = []
for answer in possible_answers:
x = (gh_pixel_count + algorithm.fill_ratio(answer[1]), len(gh_shape_count) + len(algorithm.find_regions(answer[1])) )
comparisons.append(x)
x = (abc_pixel_count, len(abc_shape_count))
closest = utility.closest_node(x, comparisons)
return (possible_answers[closest][0], possible_answers[closest][1])
def compare_top_bottom(scores, figures, solutions, problem):
if not scores:
scores = [.125, .125, .125, .125, .125, .125, .125, .125]
g_blobs = algorithm.get_blobs(algorithm.find_regions(figures[6]))
h_blobs = algorithm.get_blobs(algorithm.find_regions(figures[7]))
top_g = g_blobs[algorithm.get_top(g_blobs)]
bottom_h = h_blobs[algorithm.get_bottom(h_blobs)]
possible_answers = []
for i, score in enumerate(scores):
if score != 0.0:
possible_answers.append((i, solutions[i]))
comparisons = []
for answer in possible_answers:
solution_blobs = algorithm.get_blobs(algorithm.find_regions(answer[1]))
top_solution = solution_blobs[algorithm.get_top(solution_blobs)]
bottom_solution = solution_blobs[algorithm.get_bottom(solution_blobs)]
x = (answer[0], algorithm.calc_rms(top_g, top_solution) +
algorithm.calc_rms(bottom_h, bottom_solution))
comparisons.append(x)
m = min(comparisons, key=lambda t: t[1])
return m
def compare_top_bottom(scores, figures, solutions, problem):
if not scores:
scores = [.125, .125, .125, .125, .125, .125, .125, .125]
g_blobs = algorithm.get_blobs(algorithm.find_regions(figures[6]))
h_blobs = algorithm.get_blobs(algorithm.find_regions(figures[7]))
top_g = g_blobs[algorithm.get_top(g_blobs)]
bottom_h = h_blobs[algorithm.get_bottom(h_blobs)]
possible_answers = []
for i, score in enumerate(scores):
if score != 0.0:
possible_answers.append( (i, solutions[i]))
comparisons = []
for answer in possible_answers:
solution_blobs = algorithm.get_blobs(algorithm.find_regions(answer[1]))
top_solution = solution_blobs[algorithm.get_top(solution_blobs)]
bottom_solution = solution_blobs[algorithm.get_bottom(solution_blobs)]
x = ( answer[0], algorithm.calc_rms(top_g, top_solution) + algorithm.calc_rms(bottom_h, bottom_solution))
comparisons.append(x)
m = min(comparisons, key=lambda t: t[1])
return m
def inner_shape(source, compare):
source_blobs = algorithm.get_blobs(algorithm.find_regions(source))
compare_blobs = algorithm.get_blobs(algorithm.find_regions(compare))
source_inner = source_blobs[algorithm.get_center(source_blobs)]
compare_inner = compare_blobs[algorithm.get_center(compare_blobs)]
return equality(source_inner, compare_inner)
def shape_delta(source, compare):
source_count = len(algorithm.find_regions(source))
compare_count = len(algorithm.find_regions(compare))
if source_count < compare_count:
return 'added' # + str(compare_count - source_count)
elif source_count > compare_count:
return 'removed' # + str(source_count - compare_count)
else:
return 'unchanged'
def compare_top_corners(figures):
a_blobs = algorithm.get_blobs(algorithm.find_regions(figures[0]))
c_blobs = algorithm.get_blobs(algorithm.find_regions(figures[2]))
d_blobs = algorithm.get_blobs(algorithm.find_regions(figures[3]))
f_blobs = algorithm.get_blobs(algorithm.find_regions(figures[5]))
if len(a_blobs) < 2 or len(c_blobs) < 2 or len(d_blobs) < 2 or len(f_blobs) < 2:
return False
if transformation.strict_equality(a_blobs[algorithm.get_top(a_blobs)], c_blobs[algorithm.get_top(c_blobs)]) \
and transformation.strict_equality(d_blobs[algorithm.get_top(d_blobs)], f_blobs[algorithm.get_top(f_blobs)]):
return True
else:
return False
def compare_bottom_bc_ef(figures):
b_blobs = algorithm.get_blobs(algorithm.find_regions(figures[1]))
c_blobs = algorithm.get_blobs(algorithm.find_regions(figures[2]))
e_blobs = algorithm.get_blobs(algorithm.find_regions(figures[4]))
f_blobs = algorithm.get_blobs(algorithm.find_regions(figures[5]))
if len(b_blobs) < 2 or len(c_blobs) < 2 or len(e_blobs) < 2 or len(f_blobs) < 2:
return False
if transformation.strict_equality(b_blobs[algorithm.get_bottom(b_blobs)], c_blobs[algorithm.get_bottom(c_blobs)]) \
and transformation.strict_equality(e_blobs[algorithm.get_bottom(e_blobs)], f_blobs[algorithm.get_bottom(f_blobs)]):
return True
else:
return False
def compare_top_corners(figures):
a_blobs = algorithm.get_blobs(algorithm.find_regions(figures[0]))
c_blobs = algorithm.get_blobs(algorithm.find_regions(figures[2]))
d_blobs = algorithm.get_blobs(algorithm.find_regions(figures[3]))
f_blobs = algorithm.get_blobs(algorithm.find_regions(figures[5]))
if len(a_blobs) < 2 or len(c_blobs) < 2 or len(d_blobs) < 2 or len(
f_blobs) < 2:
return False
if transformation.strict_equality(a_blobs[algorithm.get_top(a_blobs)], c_blobs[algorithm.get_top(c_blobs)]) \
and transformation.strict_equality(d_blobs[algorithm.get_top(d_blobs)], f_blobs[algorithm.get_top(f_blobs)]):
return True
else:
return False
def compare_bottom_bc_ef(figures):
b_blobs = algorithm.get_blobs(algorithm.find_regions(figures[1]))
c_blobs = algorithm.get_blobs(algorithm.find_regions(figures[2]))
e_blobs = algorithm.get_blobs(algorithm.find_regions(figures[4]))
f_blobs = algorithm.get_blobs(algorithm.find_regions(figures[5]))
if len(b_blobs) < 2 or len(c_blobs) < 2 or len(e_blobs) < 2 or len(
f_blobs) < 2:
return False
if transformation.strict_equality(b_blobs[algorithm.get_bottom(b_blobs)], c_blobs[algorithm.get_bottom(c_blobs)]) \
and transformation.strict_equality(e_blobs[algorithm.get_bottom(e_blobs)], f_blobs[algorithm.get_bottom(f_blobs)]):
return True
else:
return False
def outer_shape(source, compare):
source_blobs = algorithm.get_blobs(algorithm.find_regions(source))
compare_blobs = algorithm.get_blobs(algorithm.find_regions(compare))
if len(source_blobs) < 2 or len(compare_blobs) < 2:
return None
else:
source_inner = algorithm.get_center(source_blobs)
compare_inner = algorithm.get_center(compare_blobs)
source_blobs.pop(source_inner)
compare_blobs.pop(compare_inner)
source_outer = algorithm.write_blobs(source_blobs)
compare_outer = algorithm.write_blobs(compare_blobs)
return equality(source_outer, compare_outer)
def compare_rows_or_cols(scores, init_set, solution_set, solutions):
abc_pixel_count = algorithm.fill_ratio(init_set[0]) + algorithm.fill_ratio(init_set[1]) \
+ algorithm.fill_ratio(init_set[2])
abc_shape_count = algorithm.find_regions(init_set[0]) + algorithm.find_regions(init_set[1]) \
+ algorithm.find_regions(init_set[2])
gh_pixel_count = algorithm.fill_ratio(
solution_set[0]) + algorithm.fill_ratio(solution_set[1])
gh_shape_count = algorithm.find_regions(
solution_set[0]) + algorithm.find_regions(solution_set[1])
possible_answers = []
for i, score in enumerate(scores):
if score != 0.0:
possible_answers.append((i, solutions[i]))
comparisons = []
for answer in possible_answers:
x = (gh_pixel_count + algorithm.fill_ratio(answer[1]),
len(gh_shape_count) + len(algorithm.find_regions(answer[1])))
comparisons.append(x)
x = (abc_pixel_count, len(abc_shape_count))
closest = utility.closest_node(x, comparisons)
return (possible_answers[closest][0], possible_answers[closest][1])