| Geant4 Cross Reference |
1 1
2 Examples for event biasing: B01 << 2 Examples for event biasing
3 ------------------------------- << 3 --------------------------
>> 4
>> 5 This directory includes example applications to demonstrate the usage of
>> 6 different variance reduction techniques supported in Geant4, or possible
>> 7 from the user applications.
4 8
5 B01, B02 and B03 applications demonstrate the <<
6 reduction techniques supported in Geant4, or p <<
7 applications. <<
8 9
9 General remark to variance reduction 10 General remark to variance reduction
10 ------------------------------------ 11 ------------------------------------
11 The tools provided for importance sampling (or 12 The tools provided for importance sampling (or geometrical splitting and
12 Russian roulette) and for the weight window te 13 Russian roulette) and for the weight window technique require the user to
13 have a good understanding of the physics in th 14 have a good understanding of the physics in the problem. This is because
14 the user has to decide which particle types ha 15 the user has to decide which particle types have to be biased, define the
15 cells (physical volumes, replicas) and assign 16 cells (physical volumes, replicas) and assign importances or weight
16 windows to that cells. If this is not done pro 17 windows to that cells. If this is not done properly it can not be
17 expected that the results describe a real expe 18 expected that the results describe a real experiment. The examples given
18 here only demonstrate how to use the tools tec 19 here only demonstrate how to use the tools technically. They don't intend
19 to produce physical correct results. 20 to produce physical correct results.
20 21
21 General remark to scoring 22 General remark to scoring
22 ------------------------- 23 -------------------------
23 Scoring is carried out using the built-in Mult << 24 A interface G4VScorer is provided for the user. The user may create his
24 parallel geometries this requires a special sc << 25 own class to perform the desired scoring. The user defined class
25 See examples/extended/runAndEvent (especailly << 26 therefore should inherit from the interface G4VScorer.
>> 27 An example of an implementation of a scorer is G4Scorer
>> 28 which may be found in source/event.
>> 29 The scoring in these examples is done with a G4Scorer.
>> 30 The variance reduction techniques and scoring do not support all options
>> 31 of the Geant4 geometry. It only supports physical volumes and simple
>> 32 replicas.
>> 33 To identify a physical volume (or replica) objects of the class
>> 34 G4GeometryCell are used. Scoring is done according to these
>> 35 cells and importance values or the weight windows may be assigned to
>> 36 them.
>> 37 When scoring is done in a parallel geometry special action has to be taken
>> 38 to prevent counting of "collisions" with boundaries of the mass geometry
>> 39 as interactions. This is differently handled when scoring is done in the
>> 40 mass geometry.
>> 41
>> 42 --> G4GeometryCell of the parallel geometry must not share boundaries with
>> 43 the world volume! <--
26 44
27 Known problems - should not happen << 45 Known problems
28 ---------------------------------- << 46 --------------
29 In the following scenario it can happen that a 47 In the following scenario it can happen that a particle is not
30 biased and it's weight is therefore not change 48 biased and it's weight is therefore not changed even if it crosses
31 a boundary where biasing should happen. 49 a boundary where biasing should happen.
32 Importance and weight window sampling create p 50 Importance and weight window sampling create particles on boundaries
33 between volumes. If the GPIL method of a physi 51 between volumes. If the GPIL method of a physical process returns
34 0 as step length for a particle on a boundary 52 0 as step length for a particle on a boundary and if the PostStepDoIt of
35 that process changes the direction of the part 53 that process changes the direction of the particle to go back in the
36 former volume the biasing won't be invoked. 54 former volume the biasing won't be invoked.
37 This will produce particles with weights that 55 This will produce particles with weights that do not correspondent to the
38 importance of the current volumes. 56 importance of the current volumes.
39 57
40 Further information: 58 Further information:
41 -------------------- 59 --------------------
42 Short description of importance sampling and s 60 Short description of importance sampling and scoring:
43 https://geant4.web.cern.ch/collaboration/worki << 61 http://cern.ch/geant4/working_groups/geometry/biasing/Sampling.html
44 62
45 Example B01 63 Example B01
46 =========== 64 ===========
47 65
48 The example uses importance sampling or the we 66 The example uses importance sampling or the weight window technique
49 according to an input parameter. It uses scori 67 according to an input parameter. It uses scoring in both cases.
50 Importance values or weight windows are define 68 Importance values or weight windows are defined according to the mass
51 geometry. In this example the weight window te 69 geometry. In this example the weight window technique is configured such
52 that it behaves equivalent to importance sampl 70 that it behaves equivalent to importance sampling: The window is actually
53 not a window but simply the inverse of the imp 71 not a window but simply the inverse of the importance value and only
54 one energy region is used that covers all ener 72 one energy region is used that covers all energies in the problem.
55 The user may change the weight window configur 73 The user may change the weight window configuration by changing the
56 initialization of the weight window algorithm 74 initialization of the weight window algorithm in example,cc.
57 Different energy bounds for the weight window 75 Different energy bounds for the weight window technique may be specified
58 in B01DetectorConstruction. 76 in B01DetectorConstruction.
59 77
60 The executable takes one optional argument: 0 78 The executable takes one optional argument: 0 or 1. Without argument or
61 with argument: 0, the importance sampling is a 79 with argument: 0, the importance sampling is applied with argument: 1,
62 the weight window technique is applied. 80 the weight window technique is applied.
63 81
64 A modular approach is applied to the physicsli <<
65 82
66 Example B02 83 Example B02
67 =========== 84 ===========
68 85
69 This example uses a parallel geometry to defin 86 This example uses a parallel geometry to define G4GeometryCell objects
70 for scoring and importance sampling. The outpu << 87 for scoring and importance sampling. In addition it customizes
>> 88 the scoring. In this example one scorer creates a histogram.
>> 89
>> 90 Compiling and running
>> 91 ---------------------
>> 92 Can be compiled and executed on a RedHat-7.3 system with gcc-3.2.3
>> 93 compiler and the tcsh shell.
>> 94 To compile this example you need AIDA 3.2.1 installed. To link
>> 95 and run it you need a AIDA compliant analysis package. The
>> 96 GNUmakefile of this example shows how to use AIDA through PI as
>> 97 analysis interface.
>> 98 Histograms are saved in HBOOK format. It can be displayed with PAW or
>> 99 compatible packages.
>> 100 You need to set the following variables in your environment:
>> 101 "G4ANALYSIS_USE"
>> 102 "PI_BASE_DIR" (where PI has been installed)
>> 103 Finally, source the script setupPI.csh.
>> 104
>> 105 Now you should be able to run gmake and to run exampleB02.
>> 106
>> 107 The example stores the plot in the file b02.hbook.
>> 108 To look at the histogram using lizard you also may use PI 1.2.1
>> 109 http://cern.ch/PI.
71 110
72 A modular approach is applied to the physicsli <<
73 The parallel geometry is included in this exte <<
74 111
75 Example B03 112 Example B03
76 =========== 113 ===========
77 114
78 This example uses a parallel geometry to defin << 115 This example uses Geant4 and in particular importance sampling and
79 for scoring and importance sampling. The outpu << 116 scoring through python. It creates a simple histogram. It's meant
80 equivalent to B02 (and B01). << 117 to demonstrate how to use a customized scorer and importance sampling
81 << 118 in combination with a scripting language, python.
82 This demonstrates a customised "flat" physics << 119 Geant4 code is executed from a python session. Therefore, swig is used
83 of biasing. Complementary approach to the modu << 120 to create python shadow classes and to generate the code necessary to
84 << 121 use the Geant4 libraries from a python session.
85 << 122 It can be built and run using PI: http://cern.ch/PI.
86 _____________________________________________ << 123 At the end a histogram called "trackentering.hbook" is created and can be
87 << 124 displayed using standard packages (such as PAW).
88 << 125
89 Generic biasing examples GB0 << 126 Building, compiling and running
90 ---------------------------- << 127 -------------------------------
91 << 128
92 These examples illustrate the usage of a biasi << 129 You need to set the following variables in your environment:
93 version Geant4 10.0. << 130 "G4ANALYSIS_USE"
94 The scheme is meant to be extensible, not limi << 131 "PI_BASE_DIR" (where PI has been installed)
95 << 132 "SWIG_BASE_DIR" (where SWIG 1.3.15 has been installed)
96 Example GB01: << 133 Finally source the script setupPI.csh.
97 ============= << 134
98 << 135 You may run gmake now.
99 This example illustrates how to bias process c << 136 You should be able to execute the file B03RunApplication.py from your
100 << 137 shell or from a lizard session now.
101 << 138
102 Example GB02: << 139 At the moment the plotting is not available using a python script, but
103 ============= << 140 it is planned in future releases. A histogram is created and later displayed
104 << 141 using standard analysis packages.
105 Illustrates a force collision scheme similar t << 142
106 << 143 To clean all the added files, just type gmake clean_all.
107 << 144
108 Example GB03: << 145 Files in B03;
109 ============= << 146 B03Application.py: Is a example class utilizing importance sampling
110 << 147 and scoring using python.
111 Illustrates geometry based biasing. << 148 B03RunApplication.py: Is a python script running the example.
112 << 149 It may be executed from the shell or in a python session.
113 << 150 B03App.py: Is created by swig using swig.
114 Example GB04: <<
115 ============= <<
116 <<
117 Illustrates a bremsstrahlung splitting. <<
118 <<
119 <<
120 Example GB05: <<
121 ============= <<
122 <<
123 Illustrates a "splitting by cross-section" tec <<
124 technique using absorption cross-section to co <<
125 <<
126 <<
127 Example GB06: <<
128 ============= <<
129 <<
130 Illustrates the usage of parallel geometries w <<
131 <<
132 Example GB07: <<
133 ============= <<
134 <<
135 Illustrates the usage of leading particle bias <<
136 <<
137 <<
138 _____________________________________________ <<
139 <<
140 <<
141 Reverse MonteCarlo Technique exam <<
142 --------------------------------- <<
143 <<
144 Example ReverseMC01 <<
145 =================== <<
146 <<
147 Example illustrating the use of the Reverse Mo <<
148 application. See details in ReverseMC01/README <<
149 <<