def test_interpolation_dropped(self):
"""Dropped channels are correctly interpolated"""
dt = self.dt_data / 3.0
tstart = (self.data.t[0] + 5 * self.dt_data).values
tend = (self.data.t[-1] - 10 * self.dt_data).values
chan_to_drop = 1
data = deepcopy(self.data)
data.attrs["dropped"] = self.dropped
data.dropped.loc[dict(chan=chan_to_drop)] = data.sel(chan=chan_to_drop)
data.loc[dict(chan=chan_to_drop)] = np.full_like(
data.sel(chan=chan_to_drop), np.nan)
try:
_data = convert_in_time_dt(tstart, tend, dt, data)
except Exception as e:
raise e
_dt = (_data.t[1] - _data.t[0]).values
assert np.all(_data.t <= data.t.max())
assert np.all(_data.t >= data.t.min())
assert _dt == approx(dt)
_dt = (_data.error.t[1] - _data.error.t[0]).values
assert np.all(_data.error.t <= data.error.t.max())
assert np.all(_data.error.t >= data.error.t.min())
assert _dt == approx(dt)
_dt = (_data.dropped.t[1] - _data.dropped.t[0]).values
assert np.all(_data.dropped.t <= data.dropped.t.max())
assert np.all(_data.dropped.t >= data.dropped.t.min())
assert _dt == approx(dt)
def test_wrong_end_time(self):
"""Checks end time wrongly set"""
dt = self.dt_data
tstart = (self.data.t[0] + 5 * self.dt_data).values
tend = (self.data.t[-1] + 10 * self.dt_data).values
try:
_ = convert_in_time_dt(tstart, tend, dt, self.data)
except ValueError as e:
assert e
def test_identity(self):
"""Checks identity"""
dt = self.dt_data * 1.0
tstart = self.data.t[0].values
tend = self.data.t[-1].values
try:
_data = convert_in_time_dt(tstart, tend, dt, self.data)
except Exception as e:
raise e
assert np.all(_data == self.data)
def test_identity_dt(self):
"""Checks identity for dt = dt_data"""
dt = self.dt_data * 1.0
tstart = (self.data.t[0] + 5 * self.dt_data).values
tend = (self.data.t[-1] - 10 * self.dt_data).values
try:
_data = convert_in_time_dt(tstart, tend, dt, self.data)
except Exception as e:
raise e
for t in _data.t:
delta = np.min(np.abs(t - self.data.t))
try:
assert approx(delta) == 0
except AssertionError as e:
print("Original and new time axis aren't identical")
raise e
delta = np.min(
np.abs(_data.sel(t=t) - self.data.sel(t=t, method="nearest")))
try:
assert approx(delta) == 0
except AssertionError as e:
print("Original and new data aren't identical")
raise e
for t in _data.error.t:
delta = np.min(np.abs(t - self.data.error.t))
try:
assert approx(delta) == 0
except AssertionError as e:
print("Original and new time axis of error aren't identical")
raise e
delta = np.min(
np.abs(
_data.error.sel(t=t) -
self.data.error.sel(t=t, method="nearest")))
try:
assert approx(delta) == 0
except AssertionError as e:
print("Original and new error aren't identical")
raise e
def test_binning(self):
"""Checks binning works as expected and returned data is withing limits"""
dt = self.dt_data * 3.0
tstart = (self.data.t[0] + 5 * self.dt_data).values
tend = (self.data.t[-1] - 10 * self.dt_data).values
try:
_data = convert_in_time_dt(tstart, tend, dt, self.data)
except Exception as e:
raise e
_dt = (_data.t[1] - _data.t[0]).values
assert np.all(_data.t <= self.data.t.max())
assert np.all(_data.t >= self.data.t.min())
assert _dt == approx(dt)
_dt = (_data.error.t[1] - _data.error.t[0]).values
assert np.all(_data.error.t <= self.data.error.t.max())
assert np.all(_data.error.t >= self.data.error.t.min())
assert _dt == approx(dt)
