You should upgrade or use an alternative browser.
-
Forums
-
Mathematics
-
MATLAB, Maple, Mathematica, LaTeX
Installing Modnum with Scicoslab
-
Thread starter
skippy1729 -
Start date
Apr 18, 2011
- Apr 18, 2011
- #1
I am using Ubuntu 10.4 LTS 64 bit for AMD processor.
First I tried Scicoslab 4.4.1 (scicoslab-gtk_4.4.1-1_amd64.maverick.deb) and Modnum 4.2.2. When I run Scicoslab, it complains that Modnum was compiled with Scicoslab 4.4 beta 7.
So, I remove 4.4.1 and install 4.4 beta 7. Now I get this message when I launch scicoslab:
Startup execution:
loading initial environment
… Search Scilab version …
Version of ScicosLab is : 4.4b7
Version of Scicos is : scicos4.4
/home/skippy/modnum_422//interf/scicoslab/../../src/libmodnum_lib.so: wrong ELF class: ELFCLASS32
«/../../src/libmodnum_lib.so»)
!—error 236
link: the shared archive was not loaded
at line 55 of exec file called by :
exec(MODNUM+»/interf/scicoslab/loader.sce»);
at line 16 of exec file called by :
exec(startup,-1);mclose(startup)
at line 272 of exec file called by :
exec(‘SCI/scilab.star’,-1);;quit
So I try «./configure —with-scisrc=/usr/lib64/scicoslab-gtk-4.4b7» followed by «make all» gives pages of error messages.
Can you suggest which version of Scicoslab and which version of Modnum will be compatible?
Any help or suggestions would be greatly appreciated.
Skippy
PS There is a installation support email address on the Modnum download page but it is dead.
Answers and Replies
- Apr 21, 2011
- #2
I just installed Modnum to Scicoslab 4.4.1 under Debian ‘squeeze’ this afternoon. here’s what I did:
— Tried to run the binary of Modnum 4.2.2 and got the same complaint as you did.
— Downloaded the source package instead and ran «exec (‘builder.sce’);)
— I had to install a few packages (and recursive dependencies) over and above what I already had (your mileage may vary): libgtk2.0-dev libvte-dev libwebkit-dev
It doesn’t install to the scicoslab directories, but everything compiles and works beautifully. Give it a go.
Cheers,
The Plumber
- Apr 21, 2011
-
- #3
Installing modnum into the scicoslab directories so that it loads automatically is done thus:
— Copy the modnum directory tree, as built using the instructions above, into (in my case): /usr/lib/scicoslab-gtk-4.4.1/contrib/
(you end up with the directory /usr/lib/scicoslab-gtk-4.4.1/contrib/modnum, for example)
— Edit /usr/lib/scicoslab-gtk-4.4.1/contrib/loader.sce to add:
exec(SCI+’/contrib/modnum/loader.sce’);
Now scicoslab/scicos loads your shiny new modnum blocks, palettes, help, etc.
Cheers,
Pb
- Apr 21, 2011
- #4
Suggested for: Installing Modnum with Scicoslab
- Last Post
- Aug 4, 2022
- Last Post
- Jul 28, 2018
- Last Post
- Jun 3, 2022
- Last Post
- Jul 25, 2022
- Last Post
- Oct 1, 2022
- Last Post
- Nov 1, 2021
- Last Post
- May 13, 2022
- Last Post
- Nov 5, 2022
- Last Post
- Feb 10, 2022
- Last Post
- Mar 21, 2015
-
Forums
-
Mathematics
-
MATLAB, Maple, Mathematica, LaTeX
feanoris 0 / 0 / 3 Регистрация: 11.10.2012 Сообщений: 56 |
||||
1 |
||||
25.10.2012, 17:06. Показов 1723. Ответов 6 Метки нет (Все метки)
Работаю в среде RadAsm. Хочу запустить пример с учебника. Создаю новый проект—выбираю DOS app. Но при сборке выдает ошибку создания Erorr during creation C:radasmmasm32BinDOSLNK.EXE /TINY 111112.obj;
__________________
0 |
639 / 482 / 12 Регистрация: 10.01.2011 Сообщений: 1,045 |
|
25.10.2012, 17:16 |
2 |
feanoris, в папке C:radasmmasm32Bin есть файл DOSLNK.EXE?
0 |
0 / 0 / 3 Регистрация: 11.10.2012 Сообщений: 56 |
|
25.10.2012, 17:43 [ТС] |
3 |
feanoris, в папке C:radasmmasm32Bin есть файл DOSLNK.EXE? нет есть только link, link16.exe
0 |
639 / 482 / 12 Регистрация: 10.01.2011 Сообщений: 1,045 |
|
25.10.2012, 18:08 |
4 |
Значит настраивайте этот Ваш RadAsm на link16.exe для линковки файлов под Dos.
0 |
0 / 0 / 3 Регистрация: 11.10.2012 Сообщений: 56 |
|
25.10.2012, 18:20 [ТС] |
5 |
Значит настраивайте этот Ваш RadAsm на link16.exe для линковки файлов под Dos. Подскажите в каком файле это необходимо прописать?
0 |
639 / 482 / 12 Регистрация: 10.01.2011 Сообщений: 1,045 |
|
25.10.2012, 18:44 |
6 |
feanoris, понятия не имею. Справки нет в RadAsm`е?
1 |
0 / 0 / 3 Регистрация: 11.10.2012 Сообщений: 56 |
|
25.10.2012, 19:36 [ТС] |
7 |
feanoris, понятия не имею. Справки нет в RadAsm`е? Все, разобрался, link16 переименовываем на doslnk.exe
0 |
- Forum
- The Ubuntu Forum Community
- Other Discussion and Support
- Education & Science
- Installing Modnum with Scicoslab
-
Installing Modnum with Scicoslab
I am having problems installing Modnum for use with Scicoslab.
I am using Ubuntu 10.4 LTS 64 bit for AMD processor.
First I tried Scicoslab 4.4.1 (scicoslab-gtk_4.4.1-1_amd64.maverick.deb) and Modnum 4.2.2. When I run Scicoslab, it complains that Modnum was compiled with Scicoslab 4.4 beta 7.
So, I remove 4.4.1 and install 4.4 beta 7. Now I get this message when I launch scicoslab:
Startup execution:
loading initial environment
… Search Scilab version …
Version of ScicosLab is : 4.4b7
Version of Scicos is : scicos4.4
/home/skippy/modnum_422//interf/scicoslab/../../src/libmodnum_lib.so: wrong ELF class: ELFCLASS32
«/../../src/libmodnum_lib.so»)
!—error 236
link: the shared archive was not loaded
at line 55 of exec file called by :
exec(MODNUM+»/interf/scicoslab/loader.sce»);
at line 16 of exec file called by :
exec(startup,-1);mclose(startup)
at line 272 of exec file called by :
exec(‘SCI/scilab.star’,-1);;quitSo I try «./configure —with-scisrc=/usr/lib64/scicoslab-gtk-4.4b7» followed by «make all» gives pages of error messages.
Can you suggest which version of Scicoslab and which version of Modnum will be compatible?
Any help or suggestions would be greatly appreciated.
Skippy
Bookmarks
Bookmarks
Posting Permissions
From Wikipedia, the free encyclopedia
Screenshot of ScicosLab 4.4 beta 7 running under Windows 7 |
|
Developer(s) | Metalau Project-team and ENPC |
---|---|
Stable release |
4.4.2 |
Operating system | Linux, Windows, Mac OS X |
Type | Technical computing |
License | Scilab License |
Website | www.scicoslab.org |
ScicosLab is a software package providing a multi-platform environment for scientific computation. It is based on the official Scilab 4.x (BUILD4) distribution, and includes the modeling and simulation tool Scicos and a number of other toolboxes.
The latest stable version of ScicosLab is ScicosLab 4.4.2.[1]
It is possible that Scilab/Scicos is currently the most complete alternative to commercial packages for dynamic systems modeling and simulation packages such as MATLAB/Simulink and MATRIXx/SystemBuild.»[2]
Features[edit]
ScicosLab runs, and is available in binary format, for the main available platforms like Unix/Linux workstations, Microsoft Windows, and MacOSX. Scicoslab was based on Scilab and Scicos, but it was forked from them. Currently it is separated from the new versions evolution in order to maintain compatibility among them.
See also[edit]
- Scilab
- Scicos
External links[edit]
- ScicosLab Homepage
- Scilab License
- Scicos Homepage
- Maxplus Homepage
- Scicos-FLEX Homepage — Scicos-FLEX is a toolbox for code generation for embedded microcontrollers
- E4Coder: The toolset based on ScicosLab for simulation and code generation for embedded devices
References[edit]
- ^ «ScicosLab Download». ScicosLab. Retrieved 2016-05-27.
- ^ Campbell, Stephen L.; Chancelier, Jean-Philippe; Nikoukhah, Ramine (2010). Modeling and Simulation in Scilab/Scicos with ScicosLab 4.4 (Second ed.). Springer Science+Business Media. p. V. doi:10.1007/978-1-4419-5527-2. ISBN 978-1-4419-5527-2.
-
Summary
-
Files
-
Reviews
-
Support
-
Mailing Lists
-
Bugs
-
News
-
Discussion
-
Mercurial ▾
- kernel-includes
- kernel-src
- matlab
- scicoslab
- scilabthe
-
SVN
Menu
▾
▴
Error in the generated code with RTAI
Created:
2011-05-23
Updated:
2013-05-01
-
Hello,
I think I found a bug when using RTSS and RTAI with Scicos. When I try to compile a Scicos diagram with RTSS blocks, ScicosLab is unable to compile the generated code and returns the following error:
«I recovered from the following error:
unix_w: rtmain44.c: In function ‘rtRegisterScope’:
in RTAICodeGen_ action.»
When I run the makefile manually the compilation error is always the same and is due to an error in the generated code in the file «.c». At the lines where the dynamics parameters are reported there are carriage returns with wrong «__CONST__» assignements.The code just below is taken from the part of the code which cause ScicosLab to fail. (I removed the parts of the code which seems to be right)
/* Object parameters declaration */
/* Routine name of block: rtss_scs_robot_cf4
* Gui name of block: rt_robot_if
* Compiled structure index: 1
*/
__CONST__ double opar_1={-3.1415927,0,0,0,0,0.758,0.1164,-0.0013,0.0149,
__CONST__ 0.0008893,0.0097063,0.0091825,-0.0001221,-0.000011,0.0046667,0.000014,
__CONST__ 78.25,0.03,0,0};/* Routine name of block: rtss_scs_gravload_cf4
* Gui name of block: rt_gravload_if
* Compiled structure index:__CONST__ double opar_8={-3.1415927,0,0,0,0,0.758,0.1164,-0.0013,0.0149,
__CONST__ 0.0008893,0.0097063,0.0091825,-0.0001221,-0.000011,0.0046667,0.000014,
__CONST__ 78.25,0.03,0,0};
-
Hi,
I’m sorry for the late reply.
I’m really under heavy load right now, and I cannot investigate the problem until the end of the next week. Anyway, can you provide some detail of your configuration? ScicosLab and RTAILab versions in particular (I assume you’re running RTSS-1.0.0b1, not the development (SVN) version…)
Thanks.
Regards,
Matteo
-
Hi Matteo,
Thank you for your answer.
Indeed I use RTSS-1.0.0b1 and not the SVN version. The version of ScicosLab I use is 4.4.1 and the version of RTAI is 3.8.1.
I have to deal with the same issue with blocks which are not include in the RTSS library when I compile with the rtss_rtai toolchain. It seems it occurs when the parameters of a block is over a given number of characters (which I cannot identify).I hope my message is clear enough.
Thanks for your help.
Regards,
Olivier
-
Hi Olivier,
I use RTSS-1.0.0b1 and not the SVN version. The version of ScicosLab I use is 4.4.1
The 1.0.0b1 distrib works with the following configuration: SL-4.3 + RTAI-3.8.1. I have a patch for SL-4.4b7 in the svn version, but I don’t know if it works for SL-4.4.1 as well. Try it by checking it out as:
svn co -r284 https://rtss.svn.sourceforge.net/svnroot/rtss/software/scicoslab/rtss/branches/rel-1.0.x robot
and please let me know if this fixes your issue. If not, I’ll further investigate the problem and I’ll try to release a fix for the next week.
See you soon,
- Matteo
-
Hi Matteo,
Unfortunately the patch doesn’t seem to work with the version I use.
Thank you for your help.
Olivier
-
Hi Olivier,
I’m able to reproduce your problem. I’ll further investigate it and I’ll provide a fix in the next few days. Anyway, I have 2 questions for you:
I have to deal with the same issue with blocks which are not include in the RTSS library when I compile with the rtss_rtai toolchain
1. Which are these blocks? Please give me their names.
2. Are you sure that the code generation works if you use the standard rtai toolchain?Regards,
- Matteo
-
Hi Matteo,
I reproduced the error with common blocks, it seems that it comes from the scope block. The problem appears even with the rtai toolchain, so I guess RTSS is not involved when using common blocks.
I also notice that it does not occur everytime and I am able to compile within ScicosLab most generally when using common blocks (not from RTSS).Below is the generated code with RTAI toolchain:
/* Routine name of block: rtscope
* Gui name of block: rtai4_scope
* Compiled structure index: 155
*/
__CONST__ unsigned char opar_81={101,114,114,111,114,83,97,109,112,108,101,83,
__CONST__ 116,97,116,101,0};
__CONST__ unsigned char opar_82={101,115,115,49,0};
__CONST__ unsigned char opar_83={101,115,115,50,0};
__CONST__ unsigned char opar_84={101,115,115,51,0};
__CONST__ unsigned char opar_85={101,115,115,52,0};And the error message from ScicosLab which is similar to the message we can get when compiling with rtss_rtai toolchain :
Executing «ok=compile_standalone();»…
I recovered from the following error:
unix_w: rtmain44.c: In function ‘rtRegisterScope’:
in RTAICodeGen_ action.When using RTSS blocks the errors appears even if we are not using scope block.
Thank you for your help.
See you soon,
Olivier
-
Hi Olivier,
In fact, there is a problem in RTAICodeGen_.sci for blocks using the object parameters list (opar). RTSS blocks make an extensive use of the opar attribute, and you can see that rtai4_scope block does use it as well:
/* Routine name of block: rtscope * Gui name of block: rtai4_scope * Compiled structure index: 155 */ +-----> opar! | __CONST__ unsigned char opar_81[]={101,114,114,111,114,83,97,109,112,108,101,83, __CONST__ 116,97,116,101,0};
I’ve reported the issue to Roberto Bucher (RTAILab lead developer) and he said he’ll provide a fix in a few days.
In the mean time, you can patch your RTAICodeGen_.sci by replacing lines 2472-2487 with the following code:
for j=1:nopar if mat2scs_c_nb(opar(opptr(i)+j-1)) <> 11 then Co_body = cformatline(mat2c_typ(opar(opptr(i)+j-1)) +... // <MM120611> ' opar_'+string(opptr(i)+j-1) + '[]={'+... strcat(string(opar(opptr(i)+j-1)),',')+'};',70); Co_body(1) = ' __CONST__ '+Co_body(1); Code_opar =[Code_opar; Co_body] // </MM120611> else //** cmplx test Co_body = cformatline(mat2c_typ(opar(opptr(i)+j-1)) +... // <MM120611> ' opar_'+string(opptr(i)+j-1) + '[]={'+... strcat(string([real(opar(opptr(i)+j-1)(:)); imag(opar(opptr(i)+j-1)(:))]),',')+'};',70); Co_body(1) = ' __CONST__ '+Co_body(1); Code_opar =[Code_opar; Co_body] // </MM120611> end end
It does work for me, let me know if it works for you too.
Regards,
- Matteo
-
Hi Matteo,
Sorry for the late reply.
Unfortunately the patch you suggested does not seem to work for me…
I should probably wait that Roberto Butcher releases the fix and try again. If it still does not work I should consider reinstalling scicoslab.
What do you think ?Thanks for your help it is very appreciated.
Regards,
Olivier
-
I should probably wait that Roberto Butcher releases the fix and try again. If it still does not work I should consider reinstalling scicoslab
A better option would be to send me two of your diagrams (with and without RTSS blocks) which you are unable to generate code from.
You can send them at mmorelli@users.sourceforge.net.
See you soon,
- Matteo
Log in to post a comment.
edit: first comment is solution
I was running webui-user. All was going well, but the installer seemed to have froze while installing torch. I looked it up and read that you should close and reopen. I did that, then it was stuck on the last ‘loading bar’ download at 1.58/1.59 gigabytes. after waiting like half an hour, I closed and reopened, then got this error:
Installing requirements for Web UI
Launching Web UI with arguments:
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
making attention of type ‘vanilla’ with 512 in_channels
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
making attention of type ‘vanilla’ with 512 in_channels
Error verifying pickled file from C:Usersomall/.cachehuggingfacetransformersc506559a5367a918bab46c39c79af91ab88846b49c8abd9d09e699ae067505c6.6365d436cc844f2f2b4885629b559d8ff0938ac484c01a6796538b2665de96c7:
Traceback (most recent call last):
File «C:AI Shenanigansstable-diffusion-webuimodulessafe.py», line 61, in check_pt
with zipfile.ZipFile(filename) as z:
File «C:Pythonlibzipfile.py», line 1267, in __init__
self._RealGetContents()
File «C:Pythonlibzipfile.py», line 1334, in _RealGetContents
raise BadZipFile(«File is not a zip file»)
zipfile.BadZipFile: File is not a zip file
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File «C:AI Shenanigansstable-diffusion-webuimodulessafe.py», line 80, in load
check_pt(filename)
File «C:AI Shenanigansstable-diffusion-webuimodulessafe.py», line 72, in check_pt
unpickler.load()
_pickle.UnpicklingError: persistent IDs in protocol 0 must be ASCII strings
The file may be malicious, so the program is not going to read it.
You can skip this check with —disable-safe-unpickle commandline argument.
Traceback (most recent call last):
File «C:AI Shenanigansstable-diffusion-webuilaunch.py», line 165, in <module>
start_webui()
File «C:AI Shenanigansstable-diffusion-webuilaunch.py», line 159, in start_webui
import webui
File «C:AI Shenanigansstable-diffusion-webuiwebui.py», line 82, in <module>
shared.sd_model = modules.sd_models.load_model()
File «C:AI Shenanigansstable-diffusion-webuimodulessd_models.py», line 181, in load_model
sd_model = instantiate_from_config(sd_config.model)
File «C:AI Shenanigansstable-diffusion-webuirepositoriesstable-diffusionldmutil.py», line 85, in instantiate_from_config
return get_obj_from_str(config[«target»])(**config.get(«params», dict()))
File «C:AI Shenanigansstable-diffusion-webuirepositoriesstable-diffusionldmmodelsdiffusionddpm.py», line 461, in __init__
self.instantiate_cond_stage(cond_stage_config)
File «C:AI Shenanigansstable-diffusion-webuirepositoriesstable-diffusionldmmodelsdiffusionddpm.py», line 519, in instantiate_cond_stage
model = instantiate_from_config(config)
File «C:AI Shenanigansstable-diffusion-webuirepositoriesstable-diffusionldmutil.py», line 85, in instantiate_from_config
return get_obj_from_str(config[«target»])(**config.get(«params», dict()))
File «C:AI Shenanigansstable-diffusion-webuirepositoriesstable-diffusionldmmodulesencodersmodules.py», line 142, in __init__
self.transformer = CLIPTextModel.from_pretrained(version)
File «C:AI Shenanigansstable-diffusion-webuivenvlibsite-packagestransformersmodeling_utils.py», line 2006, in from_pretrained
loaded_state_dict_keys = [k for k in state_dict.keys()]
AttributeError: ‘NoneType’ object has no attribute ‘keys’