LoadWANDSCD v1#

Summary#

Load WAND single crystal data into a detector space vs rotation MDHisto

Properties#

Name	Direction	Type	Default	Description
Filename	Input	list of str lists		Files to load. Allowed extensions: [‘.nxs.h5’]
IPTS	Input	number	Optional	IPTS number to load from
RunNumbers	Input	int list		Run numbers to load
VanadiumIPTS	Input	number	Optional	IPTS number to load Vanadium normalization
VanadiumRunNumber	Input	number	Optional	Run number to load Vanadium normalization
VanadiumFile	Input	string		File with Vanadium normalization scan data. Allowed extensions: [‘.nxs’]
VanadiumWorkspace	Input	IMDHistoWorkspace		MDHisto workspace containing vanadium normalization data
NormalizedBy	Input	string	None	Normalization flux. Valid options are Counts, Monitor, and Time. Allowed values: [‘None’, ‘Counts’, ‘Monitor’, ‘Time’]
NormalizeData	Input	boolean	True	When False, skip normalization even if vanadium data is provided.
Grouping	Input	string	None	Group pixels (shared by input and normalization). Allowed values: [‘None’, ‘2x2’, ‘4x4’]
ApplyGoniometerTilt	Input	boolean	False	Apply the goniometer sgl and sgu tilts to the UB-matrix. This is for backwards compatability only.
OutputWorkspace	Output	`Workspace`	Mandatory	signal for each pixel and run index. May be normalized by flux and vanadium signal.
OutputNormalizationWorkspace	Output	`Workspace`		Optional: output the normalization workspace containing the product of flux and vanadium signal for each pixel and run index.
OutputGroupingWorkspace	Output	`Workspace`		Optional: output GroupingWorkspace mapping every detector’s workspace index to its group ID. Only produced when Grouping is ‘2x2’ or ‘4x4’.

Description#

This algorithm will load a series of runs into a MDHistoWorkspace that has dimensions x and y detector pixels vs scanIndex. The scanIndex is the omega rotation of the sample. The instrument attached to the OutputWorkspace is directly copied from the FIRST run, therefore it is crucial to have the correct instrument attached to the first run. In addition the s1 (omega rotation), duration, run_number and monitor count is read from every file and included in the logs of the OutputWorkspace.

Normalization can be optionally performed in the same process, provided that the necessary Vanadium data is specified. There’s also the posibility of not normalizing the data but saving the normalization workspace alongside the main data workspace for later use in MDNorm.

By default, the algorithm will try to locate the Vanadium data using IPTS and run number. If failed, it will check the Vanadium filename entry to see if the data can be loaded directly from file. If neither is provided, the algorithm will try to check if the Vanadium data is provided as a workspace in memory.

Option NormalizedBy selects the data-to-Vanadium flux ratio as:

Monitor: ratio of data to Vanadium monitor counts.
Time: ratio of data to Vanadium duration.
Counts: 1.0 / mean Vanadium signal.
None: 1.0

If option NormalizeData is True, a normalization workspace is constructed and then divided from the data:

normalization_workspace = flux_ratio * vanadium_signal
normalized_data = sample_signal / normalization_workspace

If a workspace name is provided in OutputNormalizationWorkspace, then normalization_workspace will be saved, irrespective of the value of NormalizeData.

If the “HB2C:CS:CrystalAlign:UBMatrix” property exists and apply goniometer tilt is true, it will be converted into the OrientedLattice on the OutputWorkspace. The goniometer tilts (sgu and sgl) are combined into the UB Matrix so that only omega (s1) needs to be taken into account during rotation.

This algorithm doesn’t use Mantid loaders but instead h5py and numpy to load and integrate the events.

There is a grouping option to group pixels by either 2x2 or 4x4 which will help in reducing memory usage and speed up the later reduction steps. In most cases you will not see a difference in reduced data with 4x4 pixel grouping. Also, both input data and the Vanadium data will share the same grouping scheme.

The loaded workspace is designed to be the input to ConvertWANDSCDtoQ v1.

Usage#

Load one file, Vanadium for normalisation

norm = LoadWANDSCD(IPTS=7776, RunNumbers=26509)
print(repr(norm))

Output:

MDHistoWorkspace
Title:
Dim 0: (y) 0.5 to 512.5 in 512 bins
Dim 1: (x) 0.5 to 3840.5 in 3840 bins
Dim 2: (scanIndex) 0.5 to 1.5 in 1 bins

Inelastic: ki-kf
Instrument: ...

Run start: 2018-Mar-12 17:10:59
Run end:  not available
Sample: a 1.0, b 1.0, c 1.0; alpha 90, beta 90, gamma 90

Load multiple data files

data = LoadWANDSCD(IPTS=7776, RunNumbers='26640-27944')
print("Memory used: {}GiB".format(data.getMemorySize()/2**30))
print(repr(data))
print('s1 = {}'.format(data.getExperimentInfo(0).run().getProperty('s1').value[0:10]))
print('monitor_counts = {}'.format(data.getExperimentInfo(0).run().getProperty('monitor_counts').value[0:10]))
print('duration = {}'.format(data.getExperimentInfo(0).run().getProperty('duration').value[0:10]))
print('run_number = {}'.format(data.getExperimentInfo(0).run().getProperty('run_number').value[0:10]))

Output:

Memory used: 59GB

MDHistoWorkspace
Title:
Dim 0: (y) 0.5 to 512.5 in 512 bins
Dim 1: (x) 0.5 to 3840.5 in 3840 bins
Dim 2: (scanIndex) 0.5 to 1305.5 in 1305 bins
Inelastic: ki-kf
Instrument: ...
Run start: 2018-May-02 13:34:10
Run end:  not available
Sample: a 5.7, b 5.7, c 5.6; alpha 93, beta 90, gamma 98

s2 = [-180,-179.9,-179.8,-179.7,-179.6,-179.5,-179.4,-179.3,-179.2,-179.1]
monitor_count = [44571,44598,44567,44869,44453,44238,44611,44120,44762,44658]
duration = [2.05,2.05,2.03333,2.05,2.03333,2.03333,2.05,2.01667,2.05,2.05]
run_number = [26640,26641,26642,26643,26644,26645,26646,26647,26648,26649]

Load data and Vanadium, normalize data

data = LoadWANDSCD(
    IPTS=7776, RunNumbers='26640-26700',
    VanadiumIPTS=7776, VanadiumRunNumber=26509,
    NormalizedBy='Monitor', NormalizeData=True
    )

Load data and Vanadium. Don’t normalize the output data but output the normalization workspace

data, norm = LoadWANDSCD(
    IPTS=7776, RunNumbers='26640-26700',
    VanadiumIPTS=7776, VanadiumRunNumber=26509,
    NormalizedBy='Monitor', NormalizeData=False,
    Grouping="4x4", OutputNormalizationWorkspace='norm'
    )
print(repr(norm.id()))
print([norm.getDimension(i).getNBins() for i in range(norm.getNumDims())])

Output:

'MDHistoWorkspace'
[128, 960, 61]

Load data and Vanadium. Normalize the output data and output the grouping workspace

data = LoadWANDSCD(
    IPTS=7776, RunNumbers='26640-26700',
    VanadiumIPTS=7776, VanadiumRunNumber=26509,
    Grouping='2x2',
    NormalizedBy='Monitor', NormalizeData=True,
    OutputGroupingWorkspace='grouping'
    )
grouping = mtd['grouping']
print(int(grouping.readY(255)[0]))

Output:

Load data and Vanadium. Don’t normalize the output data and output the normalization and grouping workspaces

data, norm, grouping = LoadWANDSCD(
    IPTS=7776, RunNumbers='26640-26700',
    VanadiumIPTS=7776, VanadiumRunNumber=26509,
    Grouping='2x2',
    NormalizedBy='Monitor', NormalizeData=False,
    OutputNormalizationWorkspace='norm',
    OutputGroupingWorkspace='grouping'
    )
print([norm.getDimension(i).getNBins() for i in range(norm.getNumDims())])
print(int(grouping.readY(255)[0]))

Output:

[256, 1920, 61]
128

Load with different grouping comparing memory usage

data = LoadWANDSCD(IPTS=7776, RunNumbers='26640-27944')
data_2x2 = LoadWANDSCD(IPTS=7776, RunNumbers='26640-27944', Grouping='2x2')
data_4x4 = LoadWANDSCD(IPTS=7776, RunNumbers='26640-27944', Grouping='4x4')
print("Memory used by {}: {}GiB".format(data,data.getMemorySize()/2**30))
print("Memory used by {}: {}GiB".format(data_2x2,data_2x2.getMemorySize()/2**30))
print("Memory used by {}: {}GiB".format(data_4x4,data_4x4.getMemorySize()/2**30))
print(repr(data_4x4))

# Integrate y and plot
data_integrated = IntegrateMDHistoWorkspace('data_4x4', P1Bin='0,129')
import matplotlib.pyplot as plt
from mantid import plots
fig, ax = plt.subplots(subplot_kw={'projection':'mantid'})
c = ax.pcolormesh(data_integrated, vmax=100)
cbar=fig.colorbar(c)
cbar.set_label('Intensity (arb. units)')
#fig.savefig('LoadWANDSCD.png')

Output:

Memory used by data: 59GiB
Memory used by data_2x2: 14GiB
Memory used by data_4x4: 3GiB

MDHistoWorkspace
Title:
Dim 0: (y) 0.5 to 128.5 in 128 bins
Dim 1: (x) 0.5 to 960.5 in 960 bins
Dim 2: (scanIndex) 0.5 to 1305.5 in 1305 bins
Inelastic: ki-kf
Instrument: ...
Run start: 2018-May-02 13:34:10
Run end:  not available
Sample: a 5.7, b 5.7, c 5.6; alpha 93, beta 90, gamma 98