def test_units(observation):
ps = PowerSpectrum(observation=observation)
assert ps.horizon_buffer.to("littleh/Mpc").unit == units.littleh / units.Mpc
assert ps.k_21.to("littleh/Mpc").unit == units.littleh / units.Mpc
assert ps.delta_21.to("mK^2").unit == units.mK**2
assert callable(ps.p21)
assert ps.k_min.to("littleh/Mpc").unit == units.littleh / units.Mpc
assert ps.k_max.to("littleh/Mpc").unit == units.littleh / units.Mpc
assert ps.k1d.to("littleh/Mpc").unit == units.littleh / units.Mpc
assert isinstance(ps.power_normalisation(0.1 * units.littleh / units.Mpc), float)
assert ps.horizon_limit(10).to("littleh/Mpc").unit == units.littleh / units.Mpc
ps = PowerSpectrum(
observation=observation,
k_21=np.array([1, 2, 3]) * units.littleh / units.Mpc,
delta_21=np.array([1, 2, 3]) * units.mK**2,
)
ps2 = PowerSpectrum(
observation=observation,
k_21=np.array([1, 2, 3]) / units.Mpc,
delta_21=np.array([1, 2, 3]) * units.mK**2,
)
assert np.all(ps.k_21 < ps2.k_21)
def test_limited_k_range(observation, caplog):
ps = PowerSpectrum(
observation=observation,
k_21=np.array([1, 2, 3]) * units.littleh / units.Mpc,
delta_21=np.array([1, 2, 3]) * units.mK**2,
)
ps.k1d
assert any(
"The minimum k value is being restricted" in rec.msg for rec in caplog.records
)
def test_sensitivity_2d(observation):
ps = PowerSpectrum(observation=observation)
sense_thermal = ps.calculate_sensitivity_2d(thermal=True, sample=False)
sense_full = ps.calculate_sensitivity_2d()
assert all(np.all(sense_thermal[key] <= sense_full[key]) for key in sense_thermal)
with pytest.raises(ValueError):
ps.calculate_sensitivity_2d(thermal=False, sample=False)
def test_sensitivity_2d_grid(observation, caplog):
ps = PowerSpectrum(observation=observation)
sense_ungridded = ps.calculate_sensitivity_2d(thermal=True, sample=True)
kperp = (
np.array([x.value for x in sense_ungridded.keys()])
* list(sense_ungridded.keys())[0].unit
)
sense = ps.calculate_sensitivity_2d_grid(
kperp_edges=np.linspace(kperp.min().value, kperp.max().value, 10) * kperp.unit,
kpar_edges=ps.k1d,
)
assert sense.shape == (9, len(ps.k1d) - 1)
ps.calculate_sensitivity_2d_grid(
kperp_edges=np.linspace(ps.k_21.min().value / 2, ps.k_21.max().value * 2, 10)
* ps.k_21.unit,
kpar_edges=ps.k_21 / 2,
)
assert "minimum kbin is being restricted" in caplog.text
assert "maximum kbin is being restricted" in caplog.text
def test_plots(observation):
# this is a dumb test, just checking that it doesn't error.
ps = PowerSpectrum(observation=observation)
sense2d = ps.calculate_sensitivity_2d()
ps.plot_sense_2d(sense2d)
def test_sensitivity_1d_binned(observation):
ps = PowerSpectrum(observation=observation)
assert np.all(
ps.calculate_sensitivity_1d() == ps.calculate_sensitivity_1d_binned(ps.k1d)
)
def test_write_to_custom_filename(observation, tmp_path):
out = tmp_path / "outfile.h5"
ps = PowerSpectrum(observation=observation)
out2 = ps.write(filename=out)
assert out2 == out
def test_infs_in_trms(observation):
# default dumb layout should have lots of infs..
assert np.any(np.isinf(observation.Trms))
ps = PowerSpectrum(observation=observation)
ps.calculate_sensitivity_2d()
def test_sensitivity_optimistic(observation):
ps = PowerSpectrum(observation=observation, foreground_model="optimistic")
assert ps.horizon_limit(10.0) > ps.horizon_limit(5.0)