Exporting data to excel (csv) or matlab (mat)

[rediponi]

I have generated a set of simulations and i would be interested to export some information like:
1 - SAR slice : How to determine the slice number which is closer to a coordinate: lets say at 0cm but due to gridding it corresponds to 0.013 cm.
2- In the extracted SAR slice I am interested in the SAR values in a line along for example Y-axis. Again i know the absolute value of the line but i dont know exactly to which coordinate would correspond due to gridding.

I generated a python script of a manually extracted line, then i run the generated script, but it gives errors.

Could you show an example script of generating and exporting the SAR slice and a line in that slice?

[Sylvain]

1. You need teh function FindCell (see example below)
2. You also need FindCell, together with raw access to the underlying numpy data array and a way to "reshape" it properly (see same example below)

	# Creating the analysis pipeline
	# Adding a new SimulationExtractor
	simulation = document.AllSimulations["SCC34 Benchmark Simulation"]
	simulation_extractor = simulation.Results()

	# Adding a new EmSensorExtractor
	em_sensor_extractor = simulation_extractor["Overall Field"]
	em_sensor_extractor.FrequencySettings.ExtractedFrequency = u"All"
	em_sensor_extractor.SurfaceCurrent.SurfaceResolution = 0.001, units.Meters
	document.AllAlgorithms.Add(em_sensor_extractor)


	
	# Get data object
	em_sensor_extractor.Outputs["SAR(x,y,z,f0)"].Update()  # output needs to be updated before extracting data
	sar = em_sensor_extractor.Outputs["SAR(x,y,z,f0)"].Data
	# a location in space:
	x = 0
	y = -0.00910
	z = -0.00025
	# get grid cell indices closest to that location:
	id_x, id_y, id_z = sar.Grid.FindCell(x,y,z)
	# Create Slice Viewer at that location
	

	inputs = [em_sensor_extractor.Outputs["SAR(x,y,z,f0)"]]
	slice_field_viewer = analysis.viewers.SliceFieldViewer(inputs=inputs)
	slice_field_viewer.Data.Mode = slice_field_viewer.Data.Mode.enum.QuantityRealModulus
	slice_field_viewer.Slice.Index = id_z
	slice_field_viewer.UpdateAttributes()
	document.AllAlgorithms.Add(slice_field_viewer)
	
	# Get raw data array (in the form of a numpy array)
	sar_array = em_sensor_extractor.Outputs["SAR(x,y,z,f0)"].Data.Field(snapshot_index=0)
	nx, ny, nz = numpy.array(sar.Grid.Dimensions) - numpy.array((1,1,1))  # grid is defined on nodes, SAR data is in cell centers
	sar_array = numpy.reshape(sar_array, (nx, ny, nz), order='F')
	# now data can be accessed easily:
	print sar_array[:, id_y, id_z]

[rediponi]

This post is deleted!

[rediponi]

Thank you Sylvain.
Could you possibly add one example of code writing array (for the line) or matrix ( for the slice ) using the Matlab export ?

[Sylvain]

Currently the Matlab exporter only allows exporting a whole quantity (i.e. the whole 3D array if it is a 3D quantity, like the SAR(x,y,z,f0)).
Once exported, though, it is relatively straightforward to slice through the array in Matlab. See for example this post: https://forum.zurichmedtech.com/topic/47/exporting-data-to-matlab-how-to-access-the-data-as-a-volume-matrix

[AntoninoMC]

Dear Redi, you can export arbitrary data to Matlab (e.g. field array and grid coordinates) using the Scipy.io library. A generic use of it is the following:

import scipy.io as sio
dict={'field':fieldarray,'xgrid':xgrid,'ygrid':ygrid,'zgrid':zgrid}
sio.savemat(filename,dict)

where xgrid,ygrid,zgrid are the vectors containing the cartesian coordinates of the grid.