[Camino-users] questions about satasynt

[Alexander, Daniel]

Hi,

If you run the following (from the tutorial page):


datasynth -walkers ${WALKERS} -tmax ${TIMESTEPS} -geometry inflammation -numcylinders 100 -p 0.0 -initial uniform -voxels 1 \
-increments 1 -separateruns -latticesize ${LATSIZE} -schemefile ${SCHEMEFILE} -gamma ${GAMA} ${GAMB} -diffusivity ${DIFF} \
> ${OUTPUTDIR}/${FNAME}

it will always produce the same set of cylinders with the same configuration, because the random seed is fixed.  If you change "-voxels 1" to "-voxels 100" it will produce 100 copies each with the same set of cylinders.

To get a different substrate, ie a different set of cylinders and configuration drawn from the same distribution, you need to run datasynth 10 times with a different random seed, which you can specify with "-seed <int>", where <int> is some integer.  This tutorial, http://cmic.cs.ucl.ac.uk/camino//index.php?n=Tutorials.MCSimulator, and some of the man pages discuss best practice in using -seed.

The other problem you had with this command:


ParseSimOutput.pl LamD1a.txt

is because of a typo on the tutorial page, which I've now fixed.  Specifically, this


datasynth -walkers 5 -tmax 5 -geometry inflammation -numcylinders 100 -p 0.0 -initial uniform -voxels 1 \
-increments 1 -separateruns -latticesize ${LATSIZE} -schemefile ActiveAxG140_PM.scheme1 -gamma ${GAMA} ${GAMB} \
-diffusivity ${DIFF} -substrateinfo > /tmp/test.Bfloat 2> LamD1a.info

should have "2> LamD1a.txt" at the end instead of "2> LamD1a.info".

Danny



On 30 May 2014, at 09:26, stefania oliviero <stefania.oliviero at hotmail.it<mailto:stefania.oliviero at hotmail.it>> wrote:

Hi All,
I am trying to generate a substrate of 100 voxels with cylinders distributed second a gamma function distribution , following the instructions of the tutorial Activeax.
Due to the huge time required to generate 100 voxels I would like to generate 10 substrates each with 10 voxels.
 The questions are:

- the spatial distribution of the cylinders in each voxel is fixed or not?

- if not, is there a seed used to generate the different spatial distribution varying the voxel? How can I control it to avoid to use the same seed for each of the 10 substrates? In the worst case I will have only one substrate with 10 voxel and not 10 different substrates with 10 voxels.

I am trying to solve the problem but I don't find a solution:

- at first I used the option -drawncrosssection of the command datasynth to view the spatial distribution of the voxels. For each different run (varying the number of voxel) I obtain alway the same image file debug_crossSec_113.gray (a fixed spatial distribution of just one voxel)

- at second I used the option -cylfile of the command datasynth to read the positions of all the cylinders: I obstained always the same file testcyls.csv with the same distribution

- finally I generated the file *.info generated using the command suggested in the tutorial Activeax

datasynth -walkers 5 -tmax 5 -geometry inflammation -numcylinders 100 -p 0.0 -initial uniform -voxels 1 \ -increments 1 -separateruns -latticesize ${LATSIZE} -schemefile ActiveAxG140_PM.scheme1 -gamma ${GAMA} ${GAMB} \ -diffusivity ${DIFF} -substrateinfo > /tmp/test.Bfloat 2> LamD1a.info

The file LamD1a.info contains the position and size of each cylinder:

cylinder 0 pos = (6.65278756664775E-6, 1.8736510967768196E-6, 0.0)      radius = 2.4836106445030362E-6
cylinder 1 pos = (7.075179421691923E-6, 6.310782215503249E-6, 0.0)      radius = 1.2288218199625427E-6
cylinder 2 pos = (9.040679260286337E-7, 1.358410391168818E-5, 0.0)      radius = 1.2057082606855263E-6
cylinder 3 pos = (4.391993962910324E-6, 7.3503496846908985E-6, 0.0)     radius = 1.0606081830751333E-6
cylinder 4 pos = (9.776132252482087E-6, 1.1165129374727735E-5, 0.0)     radius = 1.0479898302635493E-6
cylinder 5 pos = (2.2709875877317475E-7, 2.8099470916851546E-6, 0.0)    radius = 1.0133293909377957E-6
cylinder 6 pos = (9.976382544843197E-6, 1.3158389965085646E-5, 0.0)     radius = 9.45079565714962E-7
cylinder 7 pos = (1.01430033661826E-5, 2.5476110220568397E-6, 0.0)      radius = 9.190732393643317E-7
cylinder 8 pos = (1.2759519960773037E-5, 1.3126264521291948E-5, 0.0)    radius = 9.166534852189462E-7
:
:


Here is a simple perl script to filter out just the list of cylinder radii: ParseSimOutput.pl<http://cmic.cs.ucl.ac.uk/camino//uploads/Tutorials/ParseSimOutput.pl>. Thus,

ParseSimOutput.pl LamD1a.txt


Produces just the list:

2.4836106445030362E-6
1.2288218199625427E-6
1.2057082606855263E-6
1.0606081830751333E-6
1.0479898302635493E-6
1.0133293909377957E-6
9.45079565714962E-7
9.190732393643317E-7
9.166534852189462E-7
:
:


perl ParseSimOutput.pl LamD1a_test.txt
output: Can't open LamD1a_test.txt: No such file or directory at ParseSimOutput.pl line 6.

perl ParseSimOutput.pl LamD1a_test.info
no output


Thanks a lot and sorry for the too long mail
Cheers

Stefania Oliviero
Università degli Studi di Napoli Federico II
ITAB, Università degli Studi G. d'Annunzio di Chieti



_______________________________________________
Camino-users mailing list
Camino-users at www.nitrc.org<mailto:Camino-users at www.nitrc.org>
http://www.nitrc.org/mailman/listinfo/camino-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.nitrc.org/pipermail/camino-users/attachments/20140602/c84653f4/attachment.html>