class ColumnOps_test(unittest.TestCase):
def setUp(self):
"""Create a test data set."""
x = np.linspace(1, 10, 10)
y = 2 * x - 3
dy = np.abs(y / 100)
z = x + 4
dz = np.abs(z / 100)
self.data = Data(
np.column_stack((x, y, dy, z, dz)),
column_headers=["Tine", "Signal 1", "d1", "Signal 2", "d2"])
def test_add(self):
self.data.add(1, 3, header="Add")
self.assertTrue(
np.all(self.data // "Add" == (self.data // "Signal 1" +
self.data // "Signal 2")),
"Failed to add column correctly")
self.data.add((1, 2), (3, 4), header="Add", index="Add", replace=True)
d_man = np.sqrt((self.data // "d1")**2 + (self.data // "d2")**2)
self.assertTrue(np.allclose(self.data // -1, d_man),
"Failed to calculate error in add")
self.data.add(1, 3.0, index="Add", replace=True)
self.assertTrue(np.all(self.data // 5 == self.data // 1 + 3),
"Failed to add with a constant")
self.data.add(1, np.ones(10), index=5, replace=True)
self.assertTrue(np.all(self.data // 5 == self.data // 1 + 1),
"Failed to add with a array")
def test_subtract(self):
self.data.subtract(1, 3, header="Subtract")
self.assertTrue(
np.all(self.data // "Subtract" == (self.data // "Signal 1" -
self.data // "Signal 2")),
"Failed to add column correctly")
self.data.subtract((1, 2), (3, 4),
header="Subtract",
index="Subtract",
replace=True)
d_man = np.sqrt((self.data // "d1")**2 + (self.data // "d2")**2)
self.assertTrue(np.allclose(self.data // -1, d_man),
"Failed to calculate error in add")
self.data.subtract(1, 3.0, index="Subtract", replace=True)
self.assertTrue(np.all(self.data // 5 == self.data // 1 - 3),
"Failed to subtract with a constant")
self.data.subtract(1, np.ones(10), index=5, replace=True)
self.assertTrue(np.all(self.data // 5 == self.data // 1 - 1),
"Failed to subtract with a array")
def test_multiply(self):
self.data.multiply(1, 3, header="Multiply")
self.assertTrue(
np.all((self.data // "Multiply") == ((self.data // "Signal 1") *
(self.data // "Signal 2"))),
"Failed to add column correctly")
self.data.multiply((1, 2), (3, 4),
header="Multiply",
index="Multiply",
replace=True)
d_man = np.sqrt(2) * 0.01 * np.abs(self.data // -2)
self.assertTrue(np.allclose(self.data // -1, d_man),
"Failed to calculate error in add")
self.data.multiply(1, 3.0, index="Multiply", replace=True)
self.assertTrue(np.all(self.data // 5 == (self.data // 1) * 3),
"Failed to multiply with a constant")
self.data.multiply(1, 2 * np.ones(10), index=5, replace=True)
self.assertTrue(np.all(self.data // 5 == (self.data // 1) * 2),
"Failed to multiply with a array")
def test_divide(self):
self.data.divide(1, 3, header="Divide")
self.assertTrue(
np.all((self.data // "Divide") == ((self.data // "Signal 1") /
(self.data // "Signal 2"))),
"Failed to add column correctly")
self.data.divide((1, 2), (3, 4),
header="Divide",
index="Divide",
replace=True)
d_man = np.sqrt(2) * 0.01 * np.abs(self.data // -2)
self.assertTrue(np.allclose(self.data // -1, d_man),
"Failed to calculate error in diffsum")
self.data.divide(1, 3.0, index="Divide", replace=True)
self.assertTrue(np.all(self.data // 5 == (self.data // 1) / 3),
"Failed to add with a constant")
self.data.divide(1, 2 * np.ones(10), index=5, replace=True)
self.assertTrue(np.all(self.data // 5 == (self.data // 1) / 2),
"Failed to add with a array")
def test_diffsum(self):
self.data.diffsum(1, 3, header="Diffsum")
a = self.data // 1
b = self.data // 3
man = (a - b) / (a + b)
self.assertTrue(np.all((self.data // "Diffsum") == man),
"Failed to diffsum column correctly")
self.data.diffsum((1, 2), (3, 4),
header="Diffsum",
index="Diffsum",
replace=True)
def test_limits(self):
self.data.setas = "x..ye"
self.assertEqual(self.data.min(1), (-1, 0), "Minimum method failed")
self.assertEqual(self.data.min(), (5.0, 0), "Minimum method failed")
self.assertEqual(self.data.min(1, bounds=lambda r: r[0] > 2),
(3.0, 2.0), "Max with bounds failed")
self.assertEqual(self.data.max(), (14, 9), "Max method failed")
self.assertEqual(self.data.max(1), (17, 9), "Max method failed")
self.assertEqual(self.data.max(1, bounds=lambda r: r[0] < 5),
(5.0, 3.0), "Max with bounds failed")
self.assertEqual(self.data.span(), (5.0, 14.0), "span method failed")
self.assertEqual(self.data.span(1), (-1.0, 17.0), "span method failed")
self.assertEqual(self.data.span(1, bounds=lambda r: 2 < r.i < 8),
(5.0, 13.0), "span with bounds failed")
def test_stats(self):
self.assertEqual(self.data.mean(1), 8.0, "Simple channel mean failed")
self.assertTrue(
np.allclose(self.data.mean(1, sigma=2),
(0.048990998729652346, 0.031144823004794875)),
"Channel mean with sigma failed")
self.assertAlmostEqual(self.data.std(1),
6.0553007081949835,
msg="Simple Standard Deviation failed")
self.assertAlmostEqual(self.data.std(1, 2),
2.7067331877422456,
msg="Simple Standard Deviation failed")
t_col = d.find_col(pat, force_list=True)
if len(t_col) == 1:
t_col = t_col[0]
break
else:
t_col = select_col(d, "Select column for temperature data :")
for pat in r_pat:
r_col = d.find_col(pat, force_list=True)
if len(r_col) == 1:
r_col = r_col[0]
break
else:
r_col = select_col(d, "Select column for resistance data :")
rho0 = d.min(r_col)[0]
A = rho0 * 40
thetaD = 300.0
p0 = [thetaD, rho0, A]
print("Initial guesses: {}".format(p0))
d.del_rows(0, lambda x, r: np.any(np.isnan(r)))
popt, pcov = d.curve_fit(bg_wrapper,
xcol=t_col,
ycol=r_col,
p0=p0,
absolute_sigma=False)
perr = np.sqrt(np.diag(pcov))
labels = [r"\theta_D", r"\rho_0", r"A"]
