Method I | new measurement method

[chraibi]

In Gitlab by @gjaeger on Nov 12, 2018, 08:21 [origin]

Data for all individuals (all ID/PersID in trajectory file) independent of the measuring area, which is based on the Voronoi density and velocity v

[chraibi]

In Gitlab by @gjaeger on Nov 12, 2018, 13:36

The method should deliver the following output in a txt-file:

column names:

• PersID (person/agent ID)
• Frame (frame)
• x/m (x-position)
• y/m (y-position)
• z/m (z-position)
• rho (individual density)
• vel (individual velocity)
• Voronoi_Polygon (polygon string)

[chraibi]

In Gitlab by @gjaeger on Nov 12, 2018, 13:36

changed the description

[chraibi]

In Gitlab by @chraibi on Nov 12, 2018, 14:25

Don't we get this output from method D already?

[chraibi]

In Gitlab by @PaulGeoerg on Nov 12, 2018, 14:30

I think @gjaeger means to collect the information into one single .txt-file. Actually method D gives the columns #frame, speed and density. If you want information e.g. the voronoi-polygon, you will need a workaround because the information is stored in different output-files. Is that correct, Gregor?

[chraibi]

In Gitlab by @gjaeger on Nov 12, 2018, 14:30

@chraibi:

Don't we get this output from method D already?

Only depending on the measurement area.

I would like to have the data independent of a measurement area.

[chraibi]

In Gitlab by @gjaeger on Nov 12, 2018, 14:31

@PaulGeoerg: no

[chraibi]

In Gitlab by @PaulGeoerg on Nov 12, 2018, 14:32

@gjaeger ah okay, I got it =)

[chraibi]

In Gitlab by @gjaeger on Feb 7, 2019, 09:47

new branch 96-method_i with initial commit 938e43221c442e3b3f840404e3bfdbc91dc3021a

:warning: Note: Renaming of function polygon_to_string (L 533) to polygon_to_string2 required.

[chraibi]

In Gitlab by @gjaeger on Feb 7, 2019, 09:57

Idea from issue #99: consideration of the corners of the considered area when calculating the voronoi cells.

[chraibi]

In Gitlab by @chraibi on Feb 7, 2019, 12:52

Note: Renaming of function polygon_to_string (L 533) to polygon_to_string2 required.

why?

[chraibi]

In Gitlab by @gjaeger on Feb 7, 2019, 14:01

that's why:

[ 93%] Building CXX object CMakeFiles/jpsreport.dir/methods/Method_D.cpp.o
[ 96%] Building CXX object CMakeFiles/jpsreport.dir/methods/Method_I.cpp.o
[100%] Linking CXX executable ../bin/jpsreport
duplicate symbol polygon_to_string[abi:cxx11](boost::geometry::model::polygon<boost::geometry::model::d2::point_xy<double, boost::geometry::cs::cartesian>, true, true, std::vector, std::vector, std::allocator, std::allocator> const&) in:
    CMakeFiles/jpsreport.dir/methods/Method_D.cpp.o
    CMakeFiles/jpsreport.dir/methods/Method_I.cpp.o
ld: 1 duplicate symbol for architecture x86_64
collect2: error: ld returned 1 exit status
make[2]: *** [../bin/jpsreport] Error 1
make[1]: *** [CMakeFiles/jpsreport.dir/all] Error 2
make: *** [all] Error 2

[chraibi]

In Gitlab by @chraibi on Feb 8, 2019, 08:52

Funny that boost has a function with a similar name. :sweat_smile:

[chraibi]

In Gitlab by @gjaeger on Feb 8, 2019, 09:14

The function polygon_to_string was duplicate in Method_D.cpp and Method_I.cpp.

[chraibi]

In Gitlab by @chraibi on Feb 8, 2019, 09:15

I suggest to move it in Analysis, so that we have it only once. In case we later find a bug in this function and need to fix it, we don't have to do it in two different files.

[chraibi]

In Gitlab by @chraibi on Feb 11, 2019, 10:58

assigned to @chraibi

[chraibi]

In Gitlab by @gjaeger on Feb 18, 2019, 12:57

@chraibi: commit 938e432 - @ArneGraf and my work of today.

[chraibi]

In Gitlab by @chraibi on Feb 20, 2019, 08:55

mentioned in merge request !14

[chraibi]

In Gitlab by @gjaeger on Feb 22, 2019, 07:39

version
Commit hash: v0.8.3-56-gbfc4ed4

test case
ini_EO_300_free-velocity_1_test.xml

Result
The output folder is empty. No data was saved. log_UO_300-300-120_test.txt

[chraibi]

In Gitlab by @gjaeger on Feb 25, 2019, 21:49

version
Commit hash: v0.8.3-59-g3f685a3

*test case
ini_EO_300_free-velocity_1_test.xml

Result
I can compile the program and I get the following warning:

[ 89%] Building CXX object CMakeFiles/geometrycore.dir/methods/Method_I.cpp.o
/Users/gjaeger/Documents/hubs/JuPedSim_dev/jpsreport/methods/Method_I.cpp: In member function 'bool Method_I::Process(const PedData&, const std::filesystem::__cxx11::path&, const double&)':
/Users/gjaeger/Documents/hubs/JuPedSim_dev/jpsreport/methods/Method_I.cpp:219:1: warning: control reaches end of non-void function [-Wreturn-type]
 }
 ^
[ 93%] Linking CXX shared library ../lib/libgeometrycore.dylib

Where does the missing return belong? I find no difference to Method D.

I can start the calculation (see logfile). Only one frame is analyzed (see traj_EO_300_free-volocity_1_exp.txt_id_2.dat) I need every PersID with every frame.

I only need the file in IndividualFD for the data output. The other files are redundant.

[chraibi]

In Gitlab by @chraibi on Feb 26, 2019, 06:26

Where does the missing return belong?

I don't understand your meaning.

I only need the file in IndividualFD for the data output. The other files are redundant.

:+1:

[chraibi]

In Gitlab by @gjaeger on Feb 27, 2019, 22:50

version
Commit hash: v0.8.3-63-gf15df4f

test case
ini_EO_300_free-velocity_1_test.xml

result
I can compile the program and I get no warning.

jpsreport calculates something with some warnings (log file: log_UO_300-300-120_test.txt).

There seems to be a bug in the calculation of the voronoi density. The individual density is very high ($\approx 500.000\ P/m^{2}$), see traj_EO_300_free-velocity_1_exp.txt_id_2.dat.

[chraibi]

In Gitlab by @chraibi on Feb 28, 2019, 18:46

There seems to be a bug in the calculation of the voronoi density. The individual density is very high ($\approx 500.000\ P/m^{2}$), see traj_EO_300_free-velocity_1_exp.txt_id_2.dat.

That cannot be. Nothing changed in the calculation method. The bug is somewhere else.

[chraibi]

In Gitlab by @chraibi on Mar 3, 2019, 19:37

closed via merge request !14