def test_event(self):
# check that the workflow runs with event workspaces as input, junk data
event_data = CreateSampleWorkspace(
NumBanks=1,
BinWidth=20000,
PixelSpacing=0.1,
BankPixelWidth=100,
WorkspaceType="Event",
)
event_cal = CreateSampleWorkspace(
NumBanks=1,
BinWidth=20000,
PixelSpacing=0.1,
BankPixelWidth=100,
WorkspaceType="Event",
Function="Flat background",
)
event_bkg = CreateSampleWorkspace(
NumBanks=1,
BinWidth=20000,
PixelSpacing=0.1,
BankPixelWidth=100,
WorkspaceType="Event",
Function="Flat background",
)
pd_out = WANDPowderReduction(
InputWorkspace=event_data,
CalibrationWorkspace=event_cal,
BackgroundWorkspace=event_bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
)
x = pd_out.extractX()
y = pd_out.extractY()
self.assertAlmostEqual(x.min(), 0.03517355)
self.assertAlmostEqual(x.max(), 70.3119282)
self.assertAlmostEqual(y[0, 0], 0.0)
pd_out = WANDPowderReduction(
InputWorkspace=[event_data, event_data],
CalibrationWorkspace=event_cal,
BackgroundWorkspace="event_bkg,event_bkg",
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
)
x = pd_out.extractX()
y = pd_out.extractY()
self.assertAlmostEqual(x.min(), 0.03517355)
self.assertAlmostEqual(x.max(), 70.3119282)
self.assertAlmostEqual(y[0, 0], 0.0)
def test_edge_case(self):
# This is to capture the case where there are identical theta
# values for multiple spectra which cause problems when
# ResampleX is run with particular Xmin and Xmax causing the
# output to be all zeros.
# Create more real test workspace, using same properties as
# HB2C_558131
data = np.ones((256, 1920))
CreateWorkspace(
DataX=[0, 1],
DataY=data,
DataE=np.sqrt(data),
UnitX='Empty',
YUnitLabel='Counts',
NSpec=1966080 // 4,
OutputWorkspace='tmp_ws')
AddSampleLog('tmp_ws', LogName='HB2C:Mot:s2.RBV', LogText='29.9774', LogType='Number Series', NumberType='Double')
AddSampleLog('tmp_ws', LogName='HB2C:Mot:detz.RBV', LogText='0', LogType='Number Series', NumberType='Double')
tmp_ws = mtd['tmp_ws']
for n in range(tmp_ws.getNumberHistograms()):
s = tmp_ws.getSpectrum(n)
for i in range(2):
for j in range(2):
s.addDetectorID(int(n * 2 % 512 + n // (512 / 2) * 512 * 2 + j + i * 512))
LoadInstrument('tmp_ws', InstrumentName='WAND', RewriteSpectraMap=False)
out = WANDPowderReduction('tmp_ws', Target='Theta', XMin=29, XMax=31, NumberBins=10, NormaliseBy='None')
np.testing.assert_allclose(out.readY(0), [0, 0, 0, 0, 269.068237, 486.311606, 618.125152, 720.37274, 811.141863, 821.032586], rtol=5e-4)
tmp_ws.delete()
out.delete()
def test(self):
data, cal, bkg = self._create_workspaces()
# data normalised by monitor
pd_out = WANDPowderReduction(InputWorkspace=data,
Target='Theta',
NumberBins=1000)
x = pd_out.extractX()
y = pd_out.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0.00328244)
self.assertAlmostEqual(y.max(), 4.88908824)
self.assertAlmostEqual(x[0, y.argmax()], 45.094311535)
# data and calibration, limited range
pd_out2 = WANDPowderReduction(InputWorkspace=data,
CalibrationWorkspace=cal,
Target='Theta',
NumberBins=2000,
XMin=10,
XMax=40)
x = pd_out2.extractX()
y = pd_out2.extractY()
self.assertAlmostEqual(x.min(), 10.0075)
self.assertAlmostEqual(x.max(), 39.9925)
self.assertAlmostEqual(y.min(), 1.5)
self.assertAlmostEqual(y.max(), 12.6107234)
self.assertAlmostEqual(x[0, y.argmax()], 30.0025)
# data, cal and background, normalised by time
pd_out3 = WANDPowderReduction(InputWorkspace=data,
CalibrationWorkspace=cal,
BackgroundWorkspace=bkg,
Target='Theta',
NumberBins=1000,
NormaliseBy='Time')
x = pd_out3.extractX()
y = pd_out3.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0)
self.assertAlmostEqual(y.max(), 19.97968357)
self.assertAlmostEqual(x[0, y.argmax()], 45.008708196)
# data, cal and background. To d spacing
pd_out4 = WANDPowderReduction(InputWorkspace=data,
CalibrationWorkspace=cal,
BackgroundWorkspace=bkg,
Target='ElasticDSpacing',
EFixed=30,
NumberBins=1000)
x = pd_out4.extractX()
y = pd_out4.extractY()
self.assertAlmostEqual(x.min(), 1.92800159)
self.assertAlmostEqual(x.max(), 11.7586705)
self.assertAlmostEqual(y.min(), 0)
self.assertAlmostEqual(y.max(), 19.03642005)
self.assertAlmostEqual(x[0, y.argmax()], 2.1543333)
# data, cal and background with mask angle, to Q.
pd_out4 = WANDPowderReduction(InputWorkspace=data,
CalibrationWorkspace=cal,
BackgroundWorkspace=bkg,
Target='ElasticQ',
EFixed=30,
NumberBins=2000,
MaskAngle=60)
x = pd_out4.extractX()
y = pd_out4.extractY()
self.assertAlmostEqual(x.min(), 0.53479223)
self.assertAlmostEqual(x.max(), 3.21684994)
self.assertAlmostEqual(y.min(), 0)
self.assertAlmostEqual(y.max(), 19.9948756)
self.assertAlmostEqual(x[0, y.argmax()], 2.9122841)
# data, cal and background, scale background
pd_out4 = WANDPowderReduction(InputWorkspace=data,
CalibrationWorkspace=cal,
BackgroundWorkspace=bkg,
BackgroundScale=0.5,
Target='Theta',
NumberBins=1000,
NormaliseBy='Time')
x = pd_out4.extractX()
y = pd_out4.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0.75)
self.assertAlmostEqual(y.max(), 20.72968357)
self.assertAlmostEqual(x[0, y.argmax()], 45.008708196)
def test(self):
data, cal, bkg = self._create_workspaces()
# data normalised by monitor
pd_out = WANDPowderReduction(InputWorkspace=data,
Target="Theta",
NumberBins=1000)
x = pd_out.extractX()
y = pd_out.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0.00328244)
self.assertAlmostEqual(y.max(), 4.88908824)
self.assertAlmostEqual(x[0, y.argmax()], 45.094311535)
# data normalised by monitor <- duplicate input as two
# NOTE:
# still needs to check physics
pd_out_multi = WANDPowderReduction(InputWorkspace=[data, data],
Target="Theta",
NumberBins=1000)
x = pd_out_multi.extractX()
y = pd_out_multi.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0.00328244 * 2)
self.assertAlmostEqual(y.max(), 4.88908824 * 2)
self.assertAlmostEqual(x[0, y.argmax()], 45.094311535)
# data and calibration, limited range
pd_out2 = WANDPowderReduction(
InputWorkspace=data,
CalibrationWorkspace=cal,
Target="Theta",
NumberBins=2000,
XMin=10,
XMax=40,
)
x = pd_out2.extractX()
y = pd_out2.extractY()
self.assertAlmostEqual(x.min(), 10.0075)
self.assertAlmostEqual(x.max(), 39.9925)
self.assertAlmostEqual(y.min(), 1.5)
self.assertAlmostEqual(y.max(), 12.6107234)
self.assertAlmostEqual(x[0, y.argmax()], 30.0025)
# data and calibration, limited range
# NOTE:
# still needs to check physics
pd_out2_multi = WANDPowderReduction(
InputWorkspace=[data, data],
CalibrationWorkspace=cal,
Target="Theta",
NumberBins=2000,
XMin=10,
XMax=40,
)
x = pd_out2_multi.extractX()
y = pd_out2_multi.extractY()
self.assertAlmostEqual(x.min(), 10.0075)
self.assertAlmostEqual(x.max(), 39.9925)
self.assertAlmostEqual(y.min(), 1.5)
self.assertAlmostEqual(y.max(), 12.6107234)
self.assertAlmostEqual(x[0, y.argmax()], 30.0025)
# data, cal and background, normalised by time
pd_out3 = WANDPowderReduction(
InputWorkspace=data,
CalibrationWorkspace=cal,
BackgroundWorkspace=bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="Time",
)
x = pd_out3.extractX()
y = pd_out3.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0)
self.assertAlmostEqual(y.max(), 19.97968357)
self.assertAlmostEqual(x[0, y.argmax()], 45.008708196)
# data, cal and background, normalised by time
# NOTE:
# still needs to check physics
pd_out3_multi = WANDPowderReduction(
InputWorkspace=[data, data],
CalibrationWorkspace=cal,
BackgroundWorkspace="bkg,bkg",
Target="Theta",
NumberBins=1000,
NormaliseBy="Time",
)
x = pd_out3_multi.extractX()
y = pd_out3_multi.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0)
self.assertAlmostEqual(y.max(), 19.97968357)
self.assertAlmostEqual(x[0, y.argmax()], 45.008708196)
# data, cal and background. To d spacing
pd_out4 = WANDPowderReduction(
InputWorkspace=data,
CalibrationWorkspace=cal,
BackgroundWorkspace=bkg,
Target="ElasticDSpacing",
EFixed=30,
NumberBins=1000,
)
x = pd_out4.extractX()
y = pd_out4.extractY()
self.assertAlmostEqual(x.min(), 1.92800159)
self.assertAlmostEqual(x.max(), 11.7586705)
self.assertAlmostEqual(y.min(), 0)
self.assertAlmostEqual(y.max(), 19.03642005)
self.assertAlmostEqual(x[0, y.argmax()], 2.1543333)
pd_out4_multi = WANDPowderReduction(
InputWorkspace=[data, data],
CalibrationWorkspace=cal,
BackgroundWorkspace="bkg,bkg",
Target="ElasticDSpacing",
EFixed=30,
NumberBins=1000,
)
x = pd_out4_multi.extractX()
y = pd_out4_multi.extractY()
self.assertAlmostEqual(x.min(), 1.92800159)
self.assertAlmostEqual(x.max(), 11.7586705)
self.assertAlmostEqual(y.min(), 0)
self.assertAlmostEqual(y.max(), 19.03642005)
self.assertAlmostEqual(x[0, y.argmax()], 2.1543333)
# data, cal and background with mask angle, to Q.
pd_out4 = WANDPowderReduction(
InputWorkspace=data,
CalibrationWorkspace=cal,
BackgroundWorkspace=bkg,
Target="ElasticQ",
EFixed=30,
NumberBins=2000,
MaskAngle=60,
)
x = pd_out4.extractX()
y = pd_out4.extractY()
self.assertAlmostEqual(x.min(), 0.53479223, places=4)
self.assertAlmostEqual(x.max(), 3.21684994, places=4)
self.assertAlmostEqual(y.min(), 0, places=4)
self.assertAlmostEqual(y.max(), 19.9948756, places=4)
self.assertAlmostEqual(x[0, y.argmax()], 2.9122841, places=4)
# NOTE:
# Need to check the physics
pd_out4_multi = WANDPowderReduction(
InputWorkspace=[data, data],
CalibrationWorkspace=cal,
BackgroundWorkspace="bkg,bkg",
Target="ElasticQ",
EFixed=30,
NumberBins=2000,
MaskAngle=60,
)
x = pd_out4_multi.extractX()
y = pd_out4_multi.extractY()
self.assertAlmostEqual(x.min(), 0.53479223, places=4)
self.assertAlmostEqual(x.max(), 3.21684994, places=4)
self.assertAlmostEqual(y.min(), 0, places=4)
self.assertAlmostEqual(y.max(), 19.9948756, places=4)
self.assertAlmostEqual(x[0, y.argmax()], 2.9122841, places=4)
# data, cal and background, scale background
pd_out4 = WANDPowderReduction(
InputWorkspace=data,
CalibrationWorkspace=cal,
BackgroundWorkspace=bkg,
BackgroundScale=0.5,
Target="Theta",
NumberBins=1000,
NormaliseBy="Time",
)
x = pd_out4.extractX()
y = pd_out4.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0.75)
self.assertAlmostEqual(y.max(), 20.72968357)
self.assertAlmostEqual(x[0, y.argmax()], 45.008708196)
pd_out4_multi = WANDPowderReduction(
InputWorkspace=[data, data],
CalibrationWorkspace=cal,
BackgroundWorkspace="bkg,bkg",
BackgroundScale=0.5,
Target="Theta",
NumberBins=1000,
NormaliseBy="Time",
)
x = pd_out4_multi.extractX()
y = pd_out4_multi.extractY()
self.assertAlmostEqual(x.min(), 8.07086781)
self.assertAlmostEqual(x.max(), 50.82973519)
self.assertAlmostEqual(y.min(), 0.75)
self.assertAlmostEqual(y.max(), 20.72968357)
self.assertAlmostEqual(x[0, y.argmax()], 45.008708196)
ws = LoadNexus(nxs_van_file)
else:
ws = LoadWAND(upstream_van_file, Grouping='4x4')
SaveNexus(ws, nxs_van_file)
return ws
if powder:
from mantid.simpleapi import LoadWAND, WANDPowderReduction, SavePlot1D, SaveFocusedXYE, Scale
data = LoadWAND(filename, Grouping='4x4')
runNumber = data.getRunNumber()
cal = get_vanadium(runNumber)
WANDPowderReduction(InputWorkspace=data,
CalibrationWorkspace=cal,
Target='Theta',
NumberBins=1200,
OutputWorkspace='reduced')
Scale(InputWorkspace='reduced',OutputWorkspace='reduced',Factor=100)
SaveFocusedXYE('reduced', Filename=os.path.join(outdir, output_file+'.xye'), SplitFiles=False, IncludeHeader=False)
div = SavePlot1D('reduced', OutputType='plotly')
request = publish_plot('HB2C', runNumber, files={'file': div})
else: # Single Crystal
with h5py.File(filename, 'r') as f:
offset = decode(f['/entry/DASlogs/HB2C:Mot:s2.RBV/average_value'].value[0])
title = decode(f['/entry/title'].value[0])
mon = decode(f['/entry/monitor1/total_counts'].value[0])
duration = decode(f['/entry/duration'].value[0])
run_number = decode(f['/entry/run_number'].value[0])
def test_event(self):
# check that the workflow runs with event workspaces as input, junk data
event_data = CreateSampleWorkspace(
NumBanks=1,
BinWidth=20000,
PixelSpacing=0.1,
BankPixelWidth=100,
WorkspaceType="Event",
)
event_cal = CreateSampleWorkspace(
NumBanks=1,
BinWidth=20000,
PixelSpacing=0.1,
BankPixelWidth=100,
WorkspaceType="Event",
Function="Flat background",
)
event_bkg = CreateSampleWorkspace(
NumBanks=1,
BinWidth=20000,
PixelSpacing=0.1,
BankPixelWidth=100,
WorkspaceType="Event",
Function="Flat background",
)
# CASE 1
# input single workspace, output single workspace
pd_out = WANDPowderReduction(
InputWorkspace=event_data,
CalibrationWorkspace=event_cal,
BackgroundWorkspace=event_bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
Sum=False,
)
assert isinstance(pd_out, MatrixWorkspace)
x = pd_out.extractX()
y = pd_out.extractY()
self.assertAlmostEqual(x.min(), 0.03517355)
self.assertAlmostEqual(x.max(), 70.3119282)
self.assertAlmostEqual(y[0, 0], 0.0)
# CASE 2
# input multiple single ws, output (single) summed ws
pd_out = WANDPowderReduction(
InputWorkspace=[event_data, event_data],
CalibrationWorkspace=event_cal,
BackgroundWorkspace=event_bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
Sum=True,
)
x = pd_out.extractX()
y = pd_out.extractY()
self.assertAlmostEqual(x.min(), 0.03517355)
self.assertAlmostEqual(x.max(), 70.3119282)
self.assertAlmostEqual(y[0, 0], 0.0)
assert isinstance(pd_out, MatrixWorkspace)
# CASE 3
# input group ws containing several ws, output group ws containing several ws
pd_out = WANDPowderReduction(
InputWorkspace=[event_data, event_data],
CalibrationWorkspace=event_cal,
BackgroundWorkspace=event_bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
Sum=False,
)
for i in pd_out:
x = i.extractX()
y = i.extractY()
self.assertAlmostEqual(x.min(), 0.03517355)
self.assertAlmostEqual(x.max(), 70.3119282)
self.assertAlmostEqual(y[0, 0], 0.0)
assert isinstance(pd_out, WorkspaceGroup)
assert len(pd_out) == 2
event_data2 = CloneWorkspace(event_data)
event_data_group = WorkspaceGroup()
event_data_group.addWorkspace(event_data)
event_data_group.addWorkspace(event_data2)
# CASE 4 - input group ws, output group ws
pd_out = WANDPowderReduction(
InputWorkspace=event_data_group,
CalibrationWorkspace=event_cal,
BackgroundWorkspace=event_bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
Sum=False,
)
for i in pd_out:
x = i.extractX()
y = i.extractY()
self.assertAlmostEqual(x.min(), 0.03517355)
self.assertAlmostEqual(x.max(), 70.3119282)
self.assertAlmostEqual(y[0, 0], 0.0)
assert isinstance(pd_out, WorkspaceGroup)
assert len(pd_out) == 2
event_data2 = CloneWorkspace(event_data)
event_data_group = GroupWorkspaces([event_data, event_data2])
pd_out = WANDPowderReduction(
InputWorkspace=event_data_group,
CalibrationWorkspace=event_cal,
BackgroundWorkspace=event_bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
Sum=False,
)
for i in pd_out:
x = i.extractX()
y = i.extractY()
self.assertAlmostEqual(x.min(), 0.03517355)
self.assertAlmostEqual(x.max(), 70.3119282)
self.assertAlmostEqual(y[0, 0], 0.0)
assert isinstance(pd_out, WorkspaceGroup)
assert len(pd_out) == 2
Function="Flat background",
)
event_bkg = CreateSampleWorkspace(
NumBanks=1,
BinWidth=20000,
PixelSpacing=0.1,
BankPixelWidth=100,
WorkspaceType="Event",
Function="Flat background",
)
pd_out = WANDPowderReduction(
InputWorkspace=[event_data, event_data],
CalibrationWorkspace=event_cal,
BackgroundWorkspace=event_bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
Sum=True,
)
pd_out2 = WANDPowderReduction(
InputWorkspace=[event_data, event_data],
CalibrationWorkspace=event_cal,
BackgroundWorkspace=event_bkg,
Target="Theta",
NumberBins=1000,
NormaliseBy="None",
Sum=True,
)