def test_connected_regions(self):
labels, regions = lulu.connected_regions(self.img)
assert_array_equal(labels, self.img)
assert_equal(len(regions), 3)
crh.set_value(regions[0], 5)
assert_array_equal(crh.todense(regions[0]),
[[5, 5, 5, 5, 0],
[5, 0, 0, 0, 0],
[5, 0, 0, 0, 0],
[5, 5, 5, 5, 0],
[5, 5, 5, 5, 0]])
assert_array_equal(crh.todense(regions[1]),
[[0, 0, 0, 0, 1],
[0, 0, 0, 0, 1],
[0, 0, 0, 0, 1],
[0, 0, 0, 0, 1],
[0, 0, 0, 0, 1]])
assert_array_equal(crh.todense(regions[2]),
[[0, 0, 0, 0, 0],
[0, 2, 2, 2, 0],
[0, 2, 2, 2, 0],
[0, 0, 0, 0, 0],
[0, 0, 0, 0, 0]])
def test_merge_diagonal(self):
a = ConnectedRegion(shape=(2, 4),
value=1,
rowptr=[0, 2, 4],
colptr=[0, 1, 1, 2])
b = ConnectedRegion(shape=(2, 4),
value=1,
rowptr=[0, 2, 4],
colptr=[2, 3, 3, 4])
crh.merge(a, b)
assert_array_equal(crh.todense(a), [[1, 0, 1, 0], [0, 1, 0, 1]])
assert_equal(crh.nnz(a), 4)
a = ConnectedRegion(shape=(2, 2),
value=1,
rowptr=[0, 2, 4],
colptr=[0, 1, 1, 2])
b = ConnectedRegion(shape=(2, 2),
value=1,
rowptr=[0, 2, 4],
colptr=[1, 2, 0, 1])
crh.merge(a, b)
assert_array_equal(crh.todense(a), [[1, 1], [1, 1]])
assert_equal(crh.nnz(a), 4)
def test_pickle(self):
import pickle
from StringIO import StringIO
s = StringIO()
pickle.dump(self.c, s)
s.seek(0)
new_c = pickle.load(s)
assert np.all(crh.todense(new_c) == crh.todense(self.c))
def test_pickle(self):
import pickle
from StringIO import StringIO
s = StringIO()
pickle.dump(self.c, s)
s.seek(0)
new_c = pickle.load(s)
assert np.all(crh.todense(new_c) == crh.todense(self.c))
def test_pickle(self):
import pickle
try:
from StringIO import StringIO
except:
from io import BytesIO as StringIO
s = StringIO()
pickle.dump(self.c, s)
s.seek(0)
new_c = pickle.load(s)
assert np.all(crh.todense(new_c) == crh.todense(self.c))
def test_maximal(self):
c = ConnectedRegion(shape=(3, 3), value=1, rowptr=[0, 2], colptr=[0, 1])
img = crh.todense(c)
assert_equal(crh.boundary_maximum(c, img), 0)
crh.set_value(c, -1)
assert_equal(crh.boundary_maximum(c, img), 0)
def test_maximal(self):
c = ConnectedRegion(shape=(3, 3), value=1, rowptr=[0, 2], colptr=[0, 1])
img = crh.todense(c)
assert_equal(crh.boundary_maximum(c, img), 0)
crh.set_value(c, -1)
assert_equal(crh.boundary_maximum(c, img), 0)
def test_reshape(self):
d = crh.copy(self.c)
crh.reshape(d, (4, 5))
assert_array_equal(crh.todense(d), self.dense[:4, :])
crh.reshape(d, (5, 5))
crh.reshape(d)
assert_array_equal(crh.get_shape(d), (4, 5))
def test_reshape(self):
d = crh.copy(self.c)
crh.reshape(d, (4, 5))
assert_array_equal(crh.todense(d), self.dense[:4, :])
crh.reshape(d, (5, 5))
crh.reshape(d)
assert_array_equal(crh.get_shape(d), (4, 5))
def test_start_row(self):
c = ConnectedRegion(shape=(2,2),
value=1, start_row=0,
rowptr=[0,2],
colptr=[0,1])
assert_array_equal(crh.todense(c), [[1, 0],
[0, 0]])
crh.set_start_row(c, 0)
crh.set_start_row(c, 1)
assert_raises(ValueError, crh.set_start_row, c, 2)
def test_merge_diagonal(self):
a = ConnectedRegion(shape=(2, 4), value=1,
rowptr=[0, 2, 4], colptr=[0, 1, 1, 2])
b = ConnectedRegion(shape=(2, 4), value=1,
rowptr=[0, 2, 4], colptr=[2, 3, 3, 4])
crh.merge(a, b)
assert_array_equal(crh.todense(a), [[1, 0, 1, 0],
[0, 1, 0, 1]])
assert_equal(crh.nnz(a), 4)
a = ConnectedRegion(shape=(2, 2), value=1,
rowptr=[0, 2, 4], colptr=[0, 1, 1, 2])
b = ConnectedRegion(shape=(2, 2), value=1,
rowptr=[0, 2, 4], colptr=[1, 2, 0, 1])
crh.merge(a, b)
assert_array_equal(crh.todense(a), [[1, 1],
[1, 1]])
assert_equal(crh.nnz(a), 4)
def test_connected_regions(self):
labels, regions = lulu.connected_regions(self.img)
assert_array_equal(labels, self.img)
assert_equal(len(regions), 3)
crh.set_value(regions[0], 5)
assert_array_equal(crh.todense(regions[0]),
[[5, 5, 5, 5, 0], [5, 0, 0, 0, 0], [5, 0, 0, 0, 0],
[5, 5, 5, 5, 0], [5, 5, 5, 5, 0]])
assert_array_equal(crh.todense(regions[1]),
[[0, 0, 0, 0, 1], [0, 0, 0, 0, 1], [0, 0, 0, 0, 1],
[0, 0, 0, 0, 1], [0, 0, 0, 0, 1]])
assert_array_equal(crh.todense(regions[2]),
[[0, 0, 0, 0, 0], [0, 2, 2, 2, 0], [0, 2, 2, 2, 0],
[0, 0, 0, 0, 0], [0, 0, 0, 0, 0]])
def test_start_row(self):
c = ConnectedRegion(shape=(2, 2),
value=1,
start_row=0,
rowptr=[0, 2],
colptr=[0, 1])
assert_array_equal(crh.todense(c), [[1, 0], [0, 0]])
crh.set_start_row(c, 0)
crh.set_start_row(c, 1)
assert_raises(ValueError, crh.set_start_row, c, 2)
def test_merge(self):
a = ConnectedRegion(shape=(3, 3), value=1,
rowptr=[0, 2, 4, 6],
colptr=[0, 3, 0, 1, 0, 3])
b = ConnectedRegion(shape=(3, 3), value=1,
start_row=1,
rowptr=[0,2],
colptr=[1,3])
crh.merge(a, b)
assert_array_equal(crh.todense(a), np.ones((3, 3)))
a = ConnectedRegion(shape=(1, 2), value=2,
rowptr=[0, 2],
colptr=[1, 2])
b = ConnectedRegion(shape=(1, 3), value=1,
rowptr=[0, 2],
colptr=[2, 3])
crh.merge(a, b)
assert_array_equal(crh.todense(a), [[0, 2, 2]])
def test_merge(self):
a = ConnectedRegion(shape=(3, 3), value=1,
rowptr=[0, 2, 4, 6],
colptr=[0, 3, 0, 1, 0, 3])
b = ConnectedRegion(shape=(3, 3), value=1,
start_row=1,
rowptr=[0,2],
colptr=[1,3])
crh.merge(a, b)
assert_array_equal(crh.todense(a), np.ones((3, 3)))
a = ConnectedRegion(shape=(1, 2), value=2,
rowptr=[0, 2],
colptr=[1, 2])
b = ConnectedRegion(shape=(1, 3), value=1,
rowptr=[0, 2],
colptr=[2, 3])
crh.merge(a, b)
assert_array_equal(crh.todense(a), [[0, 2, 2]])
def test_outside_boundary_beyond_border(self):
c = ConnectedRegion(shape=(2, 2),
value=1,
rowptr=[0, 2, 4],
colptr=[0, 1, 1, 2])
assert_array_equal(crh.todense(c), np.eye(2))
y, x = crh.outside_boundary(c)
assert_array_equal(iarr.to_list(y),
[-1, -1, -1, 0, 0, 0, 1, 1, 1, 2, 2, 2])
assert_array_equal(iarr.to_list(x),
[-1, 0, 1, -1, 1, 2, -1, 0, 2, 0, 1, 2])
def test_outside_boundary_beyond_border(self):
c = ConnectedRegion(shape=(2, 2),
value=1,
rowptr=[0, 2, 4],
colptr=[0, 1, 1, 2])
assert_array_equal(crh.todense(c), np.eye(2))
y, x = crh.outside_boundary(c)
assert_array_equal(iarr.to_list(y),
[-1, -1, -1, 0, 0, 0, 1, 1, 1, 2, 2, 2])
assert_array_equal(iarr.to_list(x),
[-1, 0, 1, -1, 1, 2, -1, 0, 2, 0, 1, 2])
def test_merge_different_shapes(self):
a = ConnectedRegion(shape=(4, 3), value=1,
start_row=1,
rowptr=[0, 4, 8, 10],
colptr=[0, 1, 2, 3, 0, 1, 2, 3, 0, 3])
b = ConnectedRegion(shape=(3, 4), value=1,
rowptr=[0, 2, 6, 10],
colptr=[1, 4, 1, 2, 3, 4, 1, 2, 3, 4])
crh.merge(a, b)
assert_array_equal(crh.todense(a),
[[0, 1, 1, 1],
[1, 1, 1, 1],
[1, 1, 1, 1],
[1, 1, 1, 0]])
def test_internal_build_ops(self):
c = ConnectedRegion(shape=(2, 2), rowptr=[0])
assert_equal(crh._current_row(c), 0)
crh._append_colptr(c, 0, 1)
crh._new_row(c)
assert_equal(crh._current_row(c), 1)
crh._append_colptr(c, 1, 2)
crh._new_row(c)
crh.set_value(c, 5)
assert_array_equal(crh.todense(c), [[5, 0], [0, 5]])
def test_internal_build_ops(self):
c = ConnectedRegion(shape=(2, 2), rowptr=[0])
assert_equal(crh._current_row(c), 0)
crh._append_colptr(c, 0, 1)
crh._new_row(c)
assert_equal(crh._current_row(c), 1)
crh._append_colptr(c, 1, 2)
crh._new_row(c)
crh.set_value(c, 5)
assert_array_equal(crh.todense(c), [[5, 0], [0, 5]])
import sys
sys.path.append('..')
import numpy as np
import matplotlib.pyplot as plt
from lulu import ConnectedRegion
from lulu import connected_region_handler as crh
from lulu import int_array
c = ConnectedRegion(shape=(5,5),
value=1, start_row=1,
rowptr=[0,4,6,10,14],
colptr=[2,3,4,5,0,5,0,1,2,5,0,2,3,5])
print(crh.todense(c))
dense = np.zeros((7,7,3))
dense[1:6, 1:6, 0] = crh.todense(c)
plt.subplot(1, 2, 1)
plt.imshow(dense, interpolation='nearest')
plt.title('Connected region')
plt.xticks([])
plt.yticks([])
ii, jj = crh.outside_boundary(c)
dense_outside = dense.copy()
for i, j in zip(int_array.to_list(ii), int_array.to_list(jj)):
dense_outside[i + 1, j + 1] = [0, 1, 0]
def test_contains(self):
d = crh.todense(self.c)
for y, x in np.ndindex(crh.get_shape(self.c)):
crh.contains(self.c, y, x) == d[y, x]
sys.path.append('..')
import numpy as np
import matplotlib.pyplot as plt
from lulu import ConnectedRegion
from lulu import connected_region_handler as crh
from lulu import int_array
c = ConnectedRegion(shape=(5, 5),
value=1,
start_row=1,
rowptr=[0, 4, 6, 10, 14],
colptr=[2, 3, 4, 5, 0, 5, 0, 1, 2, 5, 0, 2, 3, 5])
print(crh.todense(c))
dense = np.zeros((7, 7, 3))
dense[1:6, 1:6, 0] = crh.todense(c)
plt.subplot(1, 2, 1)
plt.imshow(dense, interpolation='nearest')
plt.title('Connected region')
plt.xticks([])
plt.yticks([])
ii, jj = crh.outside_boundary(c)
dense_outside = dense.copy()
for i, j in zip(int_array.to_list(ii), int_array.to_list(jj)):
dense_outside[i + 1, j + 1] = [0, 1, 0]
sys.path.append('..')
import numpy as np
import matplotlib.pyplot as plt
from lulu import ConnectedRegion
from lulu import connected_region_handler as crh
from lulu import int_array
c = ConnectedRegion(shape=(5, 5),
value=1,
start_row=1,
rowptr=[0, 4, 6, 10, 14],
colptr=[2, 3, 4, 5, 0, 5, 0, 1, 2, 5, 0, 2, 3, 5])
print crh.todense(c)
dense = np.zeros((7, 7, 3))
dense[1:6, 1:6, 0] = crh.todense(c)
plt.subplot(1, 2, 1)
plt.imshow(dense, interpolation='nearest')
plt.title('Connected region')
plt.xticks([])
plt.yticks([])
ii, jj = crh.outside_boundary(c)
dense_outside = dense.copy()
for i, j in zip(int_array.to_list(ii), int_array.to_list(jj)):
dense_outside[i + 1, j + 1] = [0, 1, 0]
def test_copy(self):
x = [0, 1]
c = ConnectedRegion(shape=(1, 1), value=1, rowptr=[0, 2], colptr=x)
d = crh.copy(c)
assert crh.todense(d) == crh.todense(c)
def test_basic(self):
assert_array_equal(crh.todense(self.c), self.dense)
assert_array_equal(crh.todense(crh.copy(self.c)), self.dense)
def test_contains(self):
d = crh.todense(self.c)
for y, x in np.ndindex(crh.get_shape(self.c)):
crh.contains(self.c, y, x) == d[y, x]
def test_basic(self):
assert_array_equal(crh.todense(self.c), self.dense)
assert_array_equal(crh.todense(crh.copy(self.c)), self.dense)
import sys
sys.path.append('..')
import numpy as np
import matplotlib.pyplot as plt
from lulu import ConnectedRegion
from lulu import connected_region_handler as crh
from lulu import int_array
c = ConnectedRegion(shape=(5,5),
value=1, start_row=1,
rowptr=[0,4,6,10,14],
colptr=[2,3,4,5,0,5,0,1,2,5,0,2,3,5])
print crh.todense(c)
dense = np.zeros((7,7,3))
dense[1:6, 1:6, 0] = crh.todense(c)
plt.subplot(1, 2, 1)
plt.imshow(dense, interpolation='nearest')
plt.title('Connected region')
plt.xticks([])
plt.yticks([])
ii, jj = crh.outside_boundary(c)
dense_outside = dense.copy()
for i, j in zip(int_array.to_list(ii), int_array.to_list(jj)):
dense_outside[i + 1, j + 1] = [0, 1, 0]
def test_copy(self):
x = [0, 1]
c = ConnectedRegion(shape=(1,1), value=1, rowptr=[0, 2], colptr=x)
d = crh.copy(c)
assert crh.todense(d) == crh.todense(c)
