Some general questions

0 views
Skip to first unread message

Tungsing email: u0700082@nus.edu.sg

unread,
May 24, 2011, 6:32:00 AM5/24/11
to dexen_users
Hi patrick and Cihat,

i got some questions to ask:

1) the default settings for houdini is that the Y axis is pointing
upwards, whereas for patrick, u seem to mention that Y axis pointing
upwards will cause the radiance to think that the model is sitting
sideways.
Thus, should we all set the Z axis upwards as a standard for both
radiance and energyplus??

2) the dexen seem to always generate 40 short indivduals after two
tests that i have conducted for max birth of 10000. The pareto graph
that i generated gives 9960 Individuals after i add job score. I
checked the internet explorer with the asa address with port 8000. It
seems that dexen have generated 10000 individuals inside the internet
explorer, but somehow didnt evaluate the last 40 for performance eg
lux.

3) how do we actually save the information on the archive server?
this is referring to the files on the server, that we can browse on
internet explorer, thus we can see the INDividual files and the
details. The information on the pareto graph is limiting, and we may
need to refer back to the files on the server to compare the data on
the pareto graph. Letting the cmd windows remain unclose is the
current method i use to keep the data on the archive server, which
means that i can never off the computer.
But is there a way to save this data away for reference?

Thank you and regards
tungsing

Patrick Janssen

unread,
May 24, 2011, 6:55:16 AM5/24/11
to dexen...@googlegroups.com
1) Yes, you need to make z point up. You go to "Edit > Preferences > 3D View Ports", the for "Orientation" select "Z Up"

2) There are actually 10,000 individuals, but 40 of them have not yet been evaluated. So they can't be plotted. This is normal. Whenever you stop the evolution, there will always be some individuals that have not yet been evaluated.

3) Currently you need to do this manually. For example, lets say you have been running a job, and it has produced some interesting results. So you decide that you want to save these results for later reference. What you need to do is to:

a) generate all the graphs that you need using the CLI
b) go to the archive server folder and copy the folder to another location on you hard disk
c) save all your graphs to the same location

Now you have a backup of that job - you can view the graphs later, and access the individuals (including the bgeo files for the individuals).

Note that the backup cannot be opened in Dexen once the server has been closed. However, it is probably not necessary to do so in any case. Just make sure you generate all the pareto graphs etc that you need before you close the server.
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::




--
-----------------------------------------------------------------
http://groups.google.com/group/dexen_users

Tungsing9

unread,
May 24, 2011, 8:03:23 PM5/24/11
to dexen_users
Hi patrick, thanks for the info

However, how do we actually go to the archive server folder and copy
the folder to another location
on our hard disk?

where is this archive server folder stored?

Another thing is that how do we open this stored files again, in order
to access the individuals?

Thanks
Tungsing





On May 24, 6:55 pm, Patrick Janssen <patr...@janssen.name> wrote:
> 1) Yes, you need to make z point up. You go to "Edit > Preferences > 3D View
> Ports", the for "Orientation" select "Z Up"
>
> 2) There are actually 10,000 individuals, but 40 of them have not yet been
> evaluated. So they can't be plotted. This is normal. Whenever you stop the
> evolution, there will always be some individuals that have not yet been
> evaluated.
>
> 3) Currently you need to do this manually. For example, lets say you have
> been running a job, and it has produced some interesting results. So you
> decide that you want to save these results for later reference. What you
> need to do is to:
>
> a) generate all the graphs that you need using the CLI
> b) go to the archive server folder and copy the folder to another location
> on you hard disk
> c) save all your graphs to the same location
>
> Now you have a backup of that job - you can view the graphs later, and
> access the individuals (including the bgeo files for the individuals).
>
> Note that the backup cannot be opened in Dexen once the server has been
> closed. However, it is probably not necessary to do so in any case. Just
> make sure you generate all the pareto graphs etc that you need before you
> close the server.
> ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
>
> On 24 May 2011 18:32, Tungsing email: u0700...@nus.edu.sg <

Patrick Janssen

unread,
May 24, 2011, 10:39:11 PM5/24/11
to dexen...@googlegroups.com
In the CLI command, type "asa" (archive_server_address).

This will tell you the folder.
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::



--
-----------------------------------------------------------------
http://groups.google.com/group/dexen_users

Reply all
Reply to author
Forward
0 new messages