Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Topics - Martin

Pages: [1] 2
1
FAQ / Aggregate operator example
« on: January 15, 2013, 16:09:28  »
The aggregate operator takes as parameter a numbered (frequency) and interval chance node (severity) and must itself be an interval discrete function node.

It is a requirement that the state values of the numbered node form an increasing sequence (0,1,2,3,4,5....) and that the interval state range in the discrete function node is sufficient to cover zero as well as min and max product of frequency and severity state values.
(If you get insufficient state range messages when compiling a network using discrete function node and aggregate operator, it is most likely the intervals of the discrete function node that needs to be revisited to make sure that this last requirement is met).
 
Also there is a small description in the C API manual on page 74 (look for h_operator_aggregate).

Here are two example network files:
aggregate1.oobn
http://download.hugin.com/webdocs/forum_example/aggregate/aggregate1.oobn

aggregate2.oobn
http://download.hugin.com/webdocs/forum_example/aggregate/aggregate2.oobn

In aggregate1 the interval range is finite and in aggregate2 it goes to infinity.

In case you are familiar with the old convolution dialog, the aggregate operator can produce the same results.
The aggregate2 example meets the requirements of the old convolution dialog (interval range from 0 to infinity). One can then use convolution (select frequency+severity, network->analysis->convolution) and compute the total impact distribution - which should produce same result as the discrete function node distribution produced by the aggregate operator.
But where the convolution dialog only lets us inspect numbers, the aggregate operator allows us to use the numbers for further computaions in the network through the child nodes of the discrete function node.
 
Finally, as with normal function nodes the belief updating can only go forward through discrete function nodes. This means that evidence entered on the descendants of a discrete function node has no impact on beliefs of the parents.

3
FAQ / Deploy Hugin with Active Directory service
« on: November 20, 2007, 16:31:48  »

5
FAQ / Silent Installation
« on: November 07, 2007, 13:31:06  »
Question: I am an administrator and have to install Hugin on a large number of computers. Is it possible to perform a silent installation that requires no user interaction?

Answer: Yes this is possible with installation packages that ends with '.msi'.

This is the command line:
msiexec /i <hugin-installer>.msi /quiet PACKAGE_NAME="<insert packagename>" USER_NAME="<insert username>" ORGANIZATION="<insert organization>" SERIAL_NUMBER="<insert serial>" HUGINROOT=<target install dir>

If the parameter HUGINROOT is omitted, Hugin will be installed in <programfiles folder>\Hugin Expert\<hugin dist name>

A few notes:
- The license info is not checked for validity during a silent installation (verify this manually by doing a regular point-and-click installation before rolling out).
- In case the installation fails, it will fail silently and no dialog will appear to inform the user. (but the target computer will remain unchanged)
- The PACKAGE_NAME property is needed from version 7.7

6
FAQ / Difference between 32-bit and 64-bit Hugin Packages
« on: July 18, 2007, 17:05:55  »
The Hugin packages come in a 32-bit version and a 64-bit version. This is a description of the difference between them.

32-bit Hugin64-bit Hugin
Features:
32-bit GUI
32-bit C API*
32-bit C++ API*
32-bit C# API*
32-bit Java API*
32-bit ActiveX API*

* API's are only provided with developer, researcher and OEM packages.

32-bit memory constraints:
Maximum memory utilization: between ~2 and ~4 Gb, depending on operating system.

Supported platforms:
PC Windows 2000/XP/Vista/7 32-bit
Solaris 8, 9 or 10 32-bit
Linux Red Hat Enterprise 32-bit
Mac OS X 10.4, 10.5, 10.6, 10.7, 10.8 32-bit

PC Windows XP/Vista/7/8 64-bit
Solaris 8, 9 or 10 64-bit
Linux Red Hat Enterprise 64-bit
Features:
64-bit GUI
64-bit C API*
64-bit C++ API*
64-bit C# API*
64-bit Java API*


64-bit memory constraints:
Maximum memory utilization: All available system memory (> 4 Gb)

Supported platforms:
PC Windows XP/Vista/7/8 64-bit
Solaris 8, 9 or 10 64-bit
Linux Red Hat Enterprise 64-bit
Mac OS X 10.6, 10.7, 10.8 64-bit


Should I get a 32- or 64-bit version?
Keep in mind that 64-bit Hugin does not run on a 32-bit operating system. This is also the case for 64-bit computers running a 32-bit operating system. So remember to check that your computer is running one of the supported 64-bit operating systems before getting a 64-bit Hugin.

Reasons to choose 32-bit Hugin:
- Your computer are running one of the supported 32-bit operating systems.
- You do application development for a 32-bit environment.
- You need a site license and have computers running both 32-bit and 64-bit versions of the same operating system.

Reasons to choose 64-bit Hugin:
- Your computer are running one of the supported 64-bit operating systems.
- You do application development for a 64-bit environment.
- You are working with large models and run into trouble with the memory limitation of a 32-bit operating system.

7
FAQ / Why is the CDVT function always disabled?
« on: July 03, 2007, 13:37:54  »
The CDVT (Continuous-Discrete Node Value Transfer) link is only supported for "flat" models, i.e., CDVT links are not supported for OOBNs. Since new networks are OOBNs by default, the network has to be saved as ".net" before the functionality becomes available.

The button becomes "non-grey" when a pair of "legal" nodes are selected, i.e., one continuous node and one interval node including intervals with both -oo and oo.

The functionality is also available under "Network->analysis->CDVT function".

8
FAQ / Linux: java.lang.UnsatisfiedLinkError
« on: July 02, 2007, 14:19:54  »
Symptom:
When I run the Hugin GUI on linux, the application quits with the following message:
Exception in thread "main" java.lang.UnsatisfiedLinkError:
[...]



Cause:
(1 - Most likely)
Hugin will only run with Sun's java distributions. The error is most likely because you are running hugin with a java distribution that does not come from Sun.
Check your java distribution by writing 'java -version'


(2)
If the exact text of the error message says
Exception in thread "main" java.lang.UnsatisfiedLinkError: no hgapiXX in java.library.path
Then, java.library.path in the hugin startup script is not configured correctly. java.library.path must point to the directory containing libhgapiXX.so

9
Network Repository / Water
« on: June 29, 2007, 10:49:07  »
Preliminary model of the biological processes of a water purification plant. The network consists of four structurally identical subnetworks, each representing a time slice of 15 minutes. The network was developed by Finn V. Jensen, Uffe Kjærulff, Kristian G. Olesen, and Jan Pedersen.

(network size using optimal elimination order:  3,028,305)

10
Network Repository / Munin4
« on: June 29, 2007, 10:48:14  »
Subset of the Munin network (1041 nodes).   

(network size using associated elimination order: 8,860,074)

11
Network Repository / Munin3
« on: June 29, 2007, 10:47:12  »
Subset of the Munin network (1045 nodes).

(network size using associated elimination order: 3,077,688)

12
Network Repository / Munin2
« on: June 29, 2007, 10:46:01  »
Subset of the Munin network (1003 nodes).

(network size using associated elimination order: 2,049,942)

13
Network Repository / Munin1
« on: June 29, 2007, 10:44:45  »
Subset of the Munin network (189 nodes). For a description of the Munin network see Andreasen et al. (1989), MUNIN - An Expert EMG Assistant, In John E. Desmedt (ed.), Computer-Aided Electromyography and Expert Systems, ch. 21, Elsevier Science Publishers, Amsterdam.

(Network size using associated elimination order: 83,735,758)

14
Network Repository / Mildew
« on: June 29, 2007, 10:43:17  »
A preliminary model for deciding on the amount of fungicides to be used against attack of mildew in wheat. The model has been developed by Finn V. Jensen, Jørgen Olesen, and Uffe Kjærulff.

(network size using the associated optimal elimination order: 3,400,464)

15
Network Repository / Link
« on: June 29, 2007, 10:40:41  »
Pedigree for linkage analysis created by Claus Skaanning Jensen and Augustine Kong. See Jensen and Kong (1996), Blocking Gibbs Sampling for Linkage Analysis in Large Pedigrees with Many Loops, Research Report R-96-2048, Department of Computer Science, Aalborg University, Denmark

(Network size using associated elimination order: 23,983,962)

Pages: [1] 2