def test_ifas_masked_std():
""" This tests the standard deviation computation with masked
arrays."""
# Creating the testing array of integers.
test_array = test.base.create_prime_test_array(shape=(5,5))
# Creating the mask for this array.
column_indexes = [0,1,3,4]
row_indexes = [1,4,2,3]
mask_array = mask.mask_single_pixels(data_array=test_array,
column_indexes=column_indexes,
row_indexes=row_indexes)
# Creating a masked array.
masked_array = np_ma.array(test_array, mask=mask_array)
# The std.
std = core.math.ifas_masked_std(array=masked_array)
# Test the population std against the expected value.
CHECK_STRING = '29.08662486490562'
CHECK_NUMBER = sy.Float(CHECK_STRING)
# Checking the mean itself.
assert_message = ("The check std value is: {check} "
"The std value is: {std} "
"The array is: \n {array}"
.format(check=CHECK_NUMBER, std=std,
array=masked_array))
assert math.isclose(std, CHECK_NUMBER), assert_message
# All done.
return None
def test_ifas_masked_median():
""" This tests the median computation with masked arrays."""
# Creating the testing array of integers.
test_array = test.base.create_prime_test_array(shape=(5,5))
# Creating the mask for this array.
column_indexes = [0,1,3,4]
row_indexes = [1,4,2,3]
mask_array = mask.mask_single_pixels(data_array=test_array,
column_indexes=column_indexes,
row_indexes=row_indexes)
# Creating a masked array.
masked_array = np_ma.array(test_array, mask=mask_array)
# The median.
median = core.math.ifas_masked_median(array=masked_array)
# Test the median against the expected value.
CHECK_STRING = '37'
CHECK_NUMBER = sy.Integer(CHECK_STRING)
# Checking the mean itself.
assert_message = ("The check median value is: {check} "
"The median value is: {median} "
"The array is: \n {array}"
.format(check=CHECK_NUMBER, median=median,
array=masked_array))
assert median == CHECK_NUMBER, assert_message
# All done.
return None
def test_mask_single_pixels():
""" This tests the masking of single pixels."""
# Creating the testing array.
test_array = test.base.create_prime_test_array(shape=(10, 10))
# Prescribed masking parameters
# Every other column.
column_indexes = [1, 2, 3, 4, 8, 7, 6, 5, 1, 1, 8, 8]
row_indexes = [1, 2, 3, 4, 5, 6, 7, 8, 8, 6, 1, 3]
# Create the mask.
test_mask = mask.mask_single_pixels(data_array=test_array,
column_indexes=column_indexes,
row_indexes=row_indexes)
# Create a masked array for both convince and testing.
test_masked_array = np_ma.array(test_array, mask=test_mask, dtype=int)
# A properly completed mask should have the same product value
# as this number. This is how the mask is checked.
CHECK_STRING = '192.684278256839293972761174821265114117452620'
CHECK_LOGARITHM = sy.Float(CHECK_STRING)
__, __, product_log10 = core.math.ifas_large_integer_array_product(
integer_array=test_masked_array.compressed())
# Finally, check. As we are dealing with large single power
# prime composite numbers and long decimals, and the smallest
# factor change of removing the 2 product still changes the
# logarithm enough, checking if the logs are close is good
# enough.
assert_message = ("The check logarithm is: {check} "
"The product logarithm is: {log} "
"The masked array is: \n {array}".format(
check=CHECK_LOGARITHM,
log=product_log10,
array=test_masked_array))
assert math.isclose(product_log10, CHECK_LOGARITHM), assert_message
# All done.
return None