Example for the correlated roughness

Examples/python/simulation/ex008_CorrelatedRoughness/CorrelatedRoughness.py

+import numpy
+import matplotlib
+import pylab
+from math import degrees, pi, sin, cos, radians
+from libBornAgainCore import *
+
+
+wavelength = 1.0 # angstrom
+ai = 0.2 # degrees
+
+Nalpha = 200
+Nphi = 200
+
+astart = 0 # degrees
+aend = 1 # degrees
+pstart = -0.5
+pend = 0.5
+
+
+def plot_with_pylab(data):
+    result = data.getArray() + 1  # for log scale
+    axis_phi = data.getAxis(0)
+    axis_alpha = data.getAxis(1)
+
+    pylab.imshow(numpy.rot90(result, 1), norm=matplotlib.colors.LogNorm(),
+                 extent=[degrees(axis_phi.getMin()), degrees(axis_phi.getMax()), degrees(axis_alpha.getMin()), degrees(axis_alpha.getMax())])
+    pylab.xlabel(r'$\phi_f$', fontsize=20)
+    pylab.ylabel(r'$\alpha_f$', fontsize=20)
+
+    pylab.show()
+
+
+def get_sample():
+    """
+    Build and return the sample representing the layers with correlated roughness.
+    """
+    m_ambience = MaterialManager.getHomogeneousMaterial("ambience", 0.0, 0.0)
+    m_part_a = MaterialManager.getHomogeneousMaterial("PartA", 5e-6, 0.0)
+    m_part_b = MaterialManager.getHomogeneousMaterial("PartB", 10e-6, 0.0)
+    m_substrate = MaterialManager.getHomogeneousMaterial("substrate", 15e-6, 0.0)
+
+    l_ambience = Layer(m_ambience, 0)
+    l_part_a = Layer(m_part_a, 2.5*nanometer)
+    l_part_b = Layer(m_part_b, 5.0*nanometer)
+    l_substrate = Layer(m_substrate, 0)
+
+    roughness = LayerRoughness()
+    roughness.setSigma(1.0*nanometer)
+    roughness.setHurstParameter(0.3)
+    roughness.setLatteralCorrLength(5*nanometer)
+
+    my_sample = MultiLayer()
+
+    # adding layers
+    my_sample.addLayer(l_ambience)
+
+    n_repetitions = 5
+    for i in range(n_repetitions):
+        my_sample.addLayerWithTopRoughness(l_part_a, roughness)
+        my_sample.addLayerWithTopRoughness(l_part_b, roughness)
+
+    my_sample.addLayerWithTopRoughness(l_substrate, roughness)
+    my_sample.setCrossCorrLength(1e-4)
+
+    return my_sample
+
+
+def set_parameters():
+    """
+    characterizing the input beam and output detector
+    """
+    simulation = Simulation()
+    simulation.setDetectorParameters(Nphi, pstart*degree, pend*degree, Nalpha, astart*degree, aend*degree, True)
+    simulation.setBeamParameters(wavelength*angstrom, ai*degree, 0.0*degree)
+
+    return simulation
+
+
+def run_simulation():
+    sample = get_sample()
+    simulation = set_parameters()
+    simulation.setSample(sample)
+    simulation.runSimulation()
+    data = simulation.getIntensityData()
+    return data
+
+
+if __name__ == '__main__':
+    data = run_simulation()
+    plot_with_pylab(data)
+

Tests/FunctionalTests/TestCore/IsGISAXS03BAsize/IsGISAXS03BAsize.cpp

@@ -6,6 +6,7 @@
 #include "Units.h"
 #include "MathFunctions.h"
 #include "SimulationRegistry.h"
+#include "OutputDataFunctions.h"
 #include <iostream>
 #include <cmath>
 
@@ -40,19 +41,11 @@ int FunctionalTests::IsGISAXS03BAsize::analyseResults()
     const double threshold(2e-10);
 
     // Calculating average relative difference.
-    *m_result -= *m_reference;
-    *m_result /= *m_reference;
-
-    double diff(0);
-    for(OutputData<double>::const_iterator it =
-            m_result->begin(); it!=m_result->end(); ++it) {
-        diff+= std::fabs(*it);
-    }
-    diff /= m_result->getAllocatedSize();
+    double diff = OutputDataFunctions::GetDifference(*m_result,*m_reference);
 
     // Assess result.
 	bool status_ok(true);
-    if( diff > threshold || std::isnan(diff)) status_ok=false;
+    if( diff > threshold ) status_ok=false;
 
     std::cout << " diff " << diff << std::endl;
     std::cout << m_name << " " << m_description << " " <<