def test_read_write_streamarray():
with TempDirectory() as directory:
io = npy.NumpyDirectory(directory.path)
original = utilities.fake_stream_array()
name = 'stream_array'
io.write(original, name)
assert original == io.read(name)
def test_read_write_streamarray():
with TempDirectory() as directory:
filename = 'test.nc'
io = nc.NCFile(os.path.join(directory.path, filename))
original = utilities.fake_stream_array()
name = 'stream_array'
io.write(original, name)
assert original == io.read(name)
def test_cached_stream_array():
with TempDirectory() as directory:
filename = 'test.nc'
io = nc.NCFile(os.path.join(directory.path, filename), cache_s21_raw=True)
original = utilities.fake_stream_array()
name = 'measurement'
io.write(original, name)
assert np.all(original.s21_raw == io.read(name).s21_raw)
def test_read_write_streamarray():
with TempDirectory() as directory:
filename = 'test.nc'
io = nc.NCFile(os.path.join(directory.path, filename))
original = utilities.fake_stream_array()
name = 'stream_array'
io.write(original, name)
assert original == io.read(name)
def test_cached_stream_array():
with TempDirectory() as directory:
filename = 'test.nc'
io = nc.NCFile(os.path.join(directory.path, filename),
cache_s21_raw=True)
original = utilities.fake_stream_array()
name = 'measurement'
io.write(original, name)
assert np.all(original.s21_raw == io.read(name).s21_raw)
def test_s21_raw_mean():
num_tones = 4
num_samples = 128
s21 = np.empty((num_tones, num_samples), dtype=np.complex)
s21[0, :] = np.nan * (1 + 1j)
s21[1, :num_samples/2] = 1 + 2j
s21[1, num_samples/2:] = np.nan * (1 + 1j)
s21[2, :] = np.linspace(-1, 1, num_samples) + 1j * np.linspace(-2, 0, num_samples)
s21[3, :] = np.linspace(-100, 0, num_samples) + 1j * np.linspace(-100, 100, num_samples)
sa = utilities.fake_stream_array(num_tones=num_tones)
sa.s21_raw = s21
with warnings.catch_warnings(record=True) as w:
warnings.simplefilter('always')
s21_raw_mean = sa.s21_raw_mean
assert len(w) == 1
assert issubclass(w[0].category, RuntimeWarning)
assert np.isnan(s21_raw_mean[0])
np.testing.assert_allclose(s21_raw_mean[1], 1 + 2j)
np.testing.assert_allclose(s21_raw_mean[2], 0 - 1j)
np.testing.assert_allclose(s21_raw_mean[3], -50 + 0j)
def test_s21_raw_mean_error():
num_tones = 4
num_samples = 128
s21 = np.empty((num_tones, num_samples), dtype=np.complex)
s21[0, ::2] = 1 + 2j
s21[0, 1::2] = np.nan * (1 + 1j)
correct_s21_error_0 = 0
s21[1, ::2] = 1 + 0j
s21[1, 1::2] = 0 + 1j
correct_s21_error_1 = (0.5 + 0.5j) / np.sqrt(num_samples)
s21[2, ::4] = -1 + 0j
s21[2, 2::4] = 0 - 1j
s21[2, 1::2] = np.nan * (1 + 1j)
correct_s21_error_2 = (0.5 + 0.5j) / np.sqrt(
num_samples / 2) # Half the samples are NaN.
# Using np.nan * (1 + 1j) is crucial here because the error calculation takes real and imag parts of s21_raw.
# The correct error for index 3 is NaN
s21[3, :] = np.nan * (1 + 1j)
# Test StreamArray code path
sa = utilities.fake_stream_array(num_tones=num_tones)
sa.s21_raw = s21
with warnings.catch_warnings(record=True) as w:
warnings.simplefilter('always')
s21_raw_mean_error = sa.s21_raw_mean_error
assert len(w) == 2
assert issubclass(w[0].category, RuntimeWarning) # From real
assert issubclass(w[1].category, RuntimeWarning) # From imag
np.testing.assert_allclose(s21_raw_mean_error[0], correct_s21_error_0)
np.testing.assert_allclose(s21_raw_mean_error[1], correct_s21_error_1)
np.testing.assert_allclose(s21_raw_mean_error[2], correct_s21_error_2)
assert np.isnan(s21_raw_mean_error[3])
# Test SingleStream code path.
assert sa[0].s21_raw_mean_error == correct_s21_error_0
assert sa[1].s21_raw_mean_error == correct_s21_error_1
assert sa[2].s21_raw_mean_error == correct_s21_error_2
with warnings.catch_warnings(record=True) as w:
warnings.simplefilter('always')
assert np.isnan(sa[3].s21_raw_mean_error)
assert len(w) == 2
assert issubclass(w[0].category, RuntimeWarning) # From real
assert issubclass(w[1].category, RuntimeWarning) # From imag
def test_s21_raw_mean_error():
num_tones = 4
num_samples = 128
s21 = np.empty((num_tones, num_samples), dtype=np.complex)
s21[0, ::2] = 1 + 2j
s21[0, 1::2] = np.nan * (1 + 1j)
correct_s21_error_0 = 0
s21[1, ::2] = 1 + 0j
s21[1, 1::2] = 0 + 1j
correct_s21_error_1 = (0.5 + 0.5j) / np.sqrt(num_samples)
s21[2, ::4] = -1 + 0j
s21[2, 2::4] = 0 - 1j
s21[2, 1::2] = np.nan * (1 + 1j)
correct_s21_error_2 = (0.5 + 0.5j) / np.sqrt(num_samples/ 2) # Half the samples are NaN.
# Using np.nan * (1 + 1j) is crucial here because the error calculation takes real and imag parts of s21_raw.
# The correct error for index 3 is NaN
s21[3, :] = np.nan * (1 + 1j)
# Test StreamArray code path
sa = utilities.fake_stream_array(num_tones=num_tones)
sa.s21_raw = s21
with warnings.catch_warnings(record=True) as w:
warnings.simplefilter('always')
s21_raw_mean_error = sa.s21_raw_mean_error
assert len(w) == 2
assert issubclass(w[0].category, RuntimeWarning) # From real
assert issubclass(w[1].category, RuntimeWarning) # From imag
np.testing.assert_allclose(s21_raw_mean_error[0], correct_s21_error_0)
np.testing.assert_allclose(s21_raw_mean_error[1], correct_s21_error_1)
np.testing.assert_allclose(s21_raw_mean_error[2], correct_s21_error_2)
assert np.isnan(s21_raw_mean_error[3])
# Test SingleStream code path.
assert sa[0].s21_raw_mean_error == correct_s21_error_0
assert sa[1].s21_raw_mean_error == correct_s21_error_1
assert sa[2].s21_raw_mean_error == correct_s21_error_2
with warnings.catch_warnings(record=True) as w:
warnings.simplefilter('always')
assert np.isnan(sa[3].s21_raw_mean_error)
assert len(w) == 2
assert issubclass(w[0].category, RuntimeWarning) # From real
assert issubclass(w[1].category, RuntimeWarning) # From imag
def test_s21_raw_mean():
num_tones = 4
num_samples = 128
s21 = np.empty((num_tones, num_samples), dtype=np.complex)
s21[0, :] = np.nan * (1 + 1j)
s21[1, :num_samples / 2] = 1 + 2j
s21[1, num_samples / 2:] = np.nan * (1 + 1j)
s21[2, :] = np.linspace(
-1, 1, num_samples) + 1j * np.linspace(-2, 0, num_samples)
s21[3, :] = np.linspace(
-100, 0, num_samples) + 1j * np.linspace(-100, 100, num_samples)
sa = utilities.fake_stream_array(num_tones=num_tones)
sa.s21_raw = s21
with warnings.catch_warnings(record=True) as w:
warnings.simplefilter('always')
s21_raw_mean = sa.s21_raw_mean
assert len(w) == 1
assert issubclass(w[0].category, RuntimeWarning)
assert np.isnan(s21_raw_mean[0])
np.testing.assert_allclose(s21_raw_mean[1], 1 + 2j)
np.testing.assert_allclose(s21_raw_mean[2], 0 - 1j)
np.testing.assert_allclose(s21_raw_mean[3], -50 + 0j)
