ctypes: pass the dimension parameters in the correct order

[paskino]

The dimension parameter in v24.0.0 is passed reversed to what it was in version v22.0.0, see details below for FGP_TV.

In this PR, we invert the dims list at creation so that the parameters match what was in version 22.

• closes Issue with 3D GPU regularisers in v. 24.0.0 #204

Details

v22.0.0

The dimensions dims are created as:

CCPi-Regularisation-Toolkit/src/Python/src/cpu_regularisers.pyx

Lines 139 to 141 in 71f8d30

	dims[0] = inputData.shape[0]
	dims[1] = inputData.shape[1]
	dims[2] = inputData.shape[2]

and passed inverted:

CCPi-Regularisation-Toolkit/src/Python/src/cpu_regularisers.pyx

Lines 149 to 155 in 71f8d30

	TV_FGP_CPU_main(&inputData[0,0,0], &outputData[0,0,0], &infovec[0], regularisation_parameter,
	iterationsNumb,
	tolerance_param,
	methodTV,
	nonneg,
	dims[2], dims[1], dims[0])
	return (outputData,infovec)

24.0.0

The dimensions are created in the same order as in version 22, but used in the opposite order:

CCPi-Regularisation-Toolkit/src/Python/ccpi/filters/TV.py

Lines 82 to 88 in b0e1888

	dims = list(inputData.shape)
	if inputData.ndim == 2:
	dims.append(1)
	# float TV_FGP_CPU_main(float *Input, float *Output, float *infovector, float lambdaPar,
	# int iterationsNumb, float epsil, int methodTV, int nonneg, int dimX, int dimY, int dimZ);
	cilreg.TV_FGP_CPU_main(in_p, out_p, infovector_p, lambdaPar, iterationsNumb, epsil, methodTV, nonneg, dims[0], dims[1], dims[2])

[dkazanc]

Thanks @paskino, this does look right to me. Shall I add some tests (3D) here as well, so we're more certain that all of the methods do what they should?

[paskino]

Tests are very useful, please add them.

[paskino]

This may not be sufficient to run the GPU tests.

In CIL we also verify that a NVIDIA GPU is installed by calling nvidia-smi along these lines

#nvidia
try:
    subprocess.check_output('nvidia-smi')
    has_nvidia = True
except:
    has_nvidia = False

[dkazanc]

sure, will add that. thanks.

[paskino]

@dkazanc can we merge this and open an issue to add unit tests?

[dkazanc]

@paskino Yes, we can. I'm struggling to find time, but I'll come back to it. thanks.

[casperdcl]

released as v24.0.1