def test_from_file(self):
df = pd.DataFrame(dict(time=self.time, flux=self.flux, pulse_number=0))
h5_filename = "testing.h5"
df.to_hdf(h5_filename, 'df')
csv_filename = "testing.csv"
df.to_csv(csv_filename)
h5data = TimeDomainData.from_file(h5_filename)
csvdata = TimeDomainData.from_file(csv_filename)
os.remove(h5_filename)
os.remove(csv_filename)
self.assertTrue(all(np.abs(h5data.flux - csvdata.flux) < 1e-15))
self.assertTrue(all(np.abs(h5data.time - csvdata.time) < 1e-15))
def test_read_h5(self):
df = pd.DataFrame(dict(time=self.time, flux=self.flux, pulse_number=0))
filename = "testing.h5"
df.to_hdf(filename, 'df')
data = TimeDomainData.from_h5(filename)
os.remove(filename)
self.assertTrue(np.all(np.abs(data.time - df.time.values) < 1e-15))
def test_read_csv(self):
df = pd.DataFrame(dict(time=self.time, flux=self.flux, pulse_number=0))
filename = "testing.csv"
df.to_csv(filename, float_format='%.16f')
data = TimeDomainData.from_csv(filename)
os.remove(filename)
self.assertTrue(np.all(np.abs(data.time - df.time.values) < 1e-15))
def test_estimate_pulse_time(self):
passes = 0
total = 10
time = np.linspace(-1, 1, 1000)
for _ in range(total):
pulse_time = np.random.uniform(-0.5, 0.5)
pulse_width = 0.01
noise = np.random.normal(0, 1, len(time))
flux = 4 * np.exp(-(time - pulse_time)**2 / pulse_width) + noise
data = TimeDomainData.from_array(time=time, flux=flux)
if np.abs(data.estimate_pulse_time() - pulse_time) < 0.1:
passes += 1
self.assertTrue(passes == total)
def test_pulse_number(self):
time = np.linspace(0, 3, 1000)
flux = np.random.normal(0, 1, len(time))
pulse_number = np.zeros(len(time))
pulse_number[time > 1] = 1
pulse_number[time > 2] = 2
df = pd.DataFrame(dict(time=time, flux=flux,
pulse_number=pulse_number))
filename = "testing.csv"
df.to_csv(filename, float_format='%.16f')
data = TimeDomainData.from_csv(filename, pulse_number=0)
os.remove(filename)
idxs = df.pulse_number == 0
self.assertTrue(
np.all(np.abs(data.time - df[idxs].time.values) < 1e-15))
def test_range_flux(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.range_flux,
np.max(self.flux) - np.min(self.flux))
def test_max_flux(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertAlmostEqual(data.max_flux, np.exp(0), 3)
def test_from_list(self):
time = [0, 1, 2]
flux = [0, 1, 2]
TimeDomainData.from_array(time, flux)
def test_from_array_unequal_shape(self):
with self.assertRaises(ValueError):
TimeDomainData.from_array(self.time[:10], self.flux)
def test_delta_time(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
data.reference_time = 10
delta_time = data.time - 10
self.assertTrue(all(data.delta_time - delta_time == 0))
def test_reference_time_set(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
data.reference_time = 0.3
self.assertEqual(data.reference_time, 0.3)
def test_time_unit_set(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
data.time_unit = "s"
self.assertEqual(data.time_unit, "s")
def test_time_unit_default(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.time_unit, "days")
def test_from_unsorted(self):
time = [0, 1, 2, 1.5]
flux = [0, 1, 2, 1.5]
with self.assertRaises(ValueError):
TimeDomainData.from_array(time, flux)
def test_midtime(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.midtime, 0)
def test_truncate_data(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
duration = data.duration
data.truncate_data(0.5)
self.assertLess(data.duration, 0.5 * duration)
def test_reference_time_default(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.reference_time, 0)
def test_duration(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.duration, self.time[-1] - self.time[0])
def test_max_time(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertAlmostEqual(data.max_time, 0, 3)
def test_start(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.start, self.time[0])
def test_RMS_flux(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.RMS_flux, np.sqrt(np.mean(data.flux**2)))
def test_end(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.end, self.time[-1])
def test_from_array(self):
data = TimeDomainData.from_array(self.time, self.flux)
self.assertTrue(all(self.time == data.time))
self.assertTrue(all(self.flux == data.flux))
def test_N(self):
data = TimeDomainData.from_array(time=self.time, flux=self.flux)
self.assertEqual(data.N, len(self.time))
def test_from_array_ndim_greater_one(self):
with self.assertRaises(ValueError):
TimeDomainData.from_array(np.random.uniform(0, 1, (2, 3)),
np.random.uniform(0, 1, (2, 3)))
def setUp(self):
self.time = np.linspace(-1, 1, 10000)
self.flux = np.exp(-self.time**2 / 2)
self.data = TimeDomainData.from_array(time=self.time, flux=self.flux)