def test_kswin_functionality(test_path):
kswin = KSWIN(alpha=0.0001, window_size=200, stat_size=100)
test_file = os.path.join(test_path, 'drift_stream.npy')
data_stream = np.load(test_file)
expected_indices = [1045, 1145]
detected_indices = []
for i in range(data_stream.size):
kswin.update(data_stream[i])
if kswin.change_detected:
detected_indices.append(i)
assert detected_indices == expected_indices
def test_kswin_initialization():
"""
KSWIN Test
Content:
alpha initialisation test.
alpha has range from (0,1)
pre obtained data initialisation test.
data must be list
KSWIN window size initialisation test.
0 < stat_size < window_size
KSWIN change detector size initialisation test.
At least 1 false positive must arisie due to the sensitive alpha, when testing the standard
Sea generator
"""
with pytest.raises(ValueError):
KSWIN(alpha=-0.1)
with pytest.raises(ValueError):
KSWIN(alpha=1.1)
kswin = KSWIN(alpha=0.5)
assert kswin.alpha == 0.5
kswin = KSWIN(data="st")
assert isinstance(kswin.window, np.ndarray)
kswin = KSWIN(data=np.array([0.75, 0.80, 1, -1]))
assert isinstance(kswin.window, np.ndarray)
try:
KSWIN(window_size=-10)
except ValueError:
assert True
else:
assert False
try:
KSWIN(window_size=10, stat_size=30)
except ValueError:
assert True
else:
assert False
def test_kswin_coverage():
with pytest.raises(ValueError):
KSWIN(alpha=-0.1)
with pytest.raises(ValueError):
KSWIN(alpha=1.1)
kswin = KSWIN(alpha=0.5)
assert kswin.alpha == 0.5
kswin = KSWIN(window="st")
assert isinstance(kswin.window, np.ndarray)
kswin = KSWIN(window=np.array([0.75, 0.80, 1, -1]))
assert isinstance(kswin.window, np.ndarray)
try:
KSWIN(window_size=-10)
except ValueError:
assert True
else:
assert False
try:
KSWIN(window_size=10, stat_size=30)
except ValueError:
assert True
else:
assert False
kswin = KSWIN()
kswin.reset()
assert kswin.p_value == 0
assert kswin.window.shape[0] == 0
assert kswin.change_detected is False
def test_kswin():
kswin = KSWIN(alpha=0.0001, window_size=200, stat_size=100)
expected_indices = [1043, 1143]
detected_indices = perform_test(kswin, data_stream_1)
assert detected_indices == expected_indices
def test_kswin_reset():
kswin = KSWIN()
kswin.reset()
assert kswin.p_value == 0
assert kswin.window.shape[0] == 0
assert kswin.change_detected == False