def test_even_height_odd_width():
"""
WHEN: There is an image with odd height, an even width and three channels
AND: The make_square function is called
WHAT SHOULD HAPPEN: A square image is returned.
Number of channels is preserverd.
Returns:
Nothing
"""
in_height = 4
in_width = 5
for in_channels in [1, 3]:
shape = in_height, in_width, in_channels
input_image = np.zeros(shape)
output_image = make_square(input_image)
out_height, out_width, out_channels = output_image.shape
assert out_height == out_width
assert out_channels == in_channels
def test_even_height_odd_width():
"""
WHEN: There is an image with odd height, an even width and three channels
AND: The make_square function is called
WHAT SHOULD HAPPEN: A square image is returned.
Number of channels is preserverd.
Returns:
Nothing
"""
in_height = 4
in_width = 5
for in_channels in [1, 3]:
shape = in_height, in_width, in_channels
input_image = np.zeros(shape)
output_image = make_square(input_image)
out_height, out_width, out_channels = output_image.shape
assert out_height == out_width
assert out_channels == in_channels
def test_even_height_odd_width_square():
"""
WHEN: There is a square image with odd height, an odd width and three channels
AND: The make_square function is called
WHAT SHOULD HAPPEN: The original image is returned.
Number of channels is preserverd.
Returns:
Nothing
"""
in_height = 4
in_width = 4
for in_channels in [1, 3]:
shape = in_height, in_width, in_channels
input_image = np.zeros(shape)
output_image = make_square(input_image)
assert np.array_equal(output_image, input_image)
def test_even_height_odd_width_square():
"""
WHEN: There is a square image with odd height, an odd width and three channels
AND: The make_square function is called
WHAT SHOULD HAPPEN: The original image is returned.
Number of channels is preserverd.
Returns:
Nothing
"""
in_height = 4
in_width = 4
for in_channels in [1, 3]:
shape = in_height, in_width, in_channels
input_image = np.zeros(shape)
output_image = make_square(input_image)
assert np.array_equal(output_image, input_image)