Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

5 Pages<1234>»
Options
Go to last post Go to first unread
Offline uni  
#21 Posted : 24 August 2013 16:37:51(UTC)
uni


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 10/11/2010(UTC)
Posts: 1,331
Man
Russian Federation

Was thanked: 1163 time(s) in 672 post(s)
Originally Posted by: mkraska Go to Quoted Post
Originally Posted by: uni Go to Quoted Post
Updated. Now the installer is located within the plugin.

There is an user interaction issue. After updating the plugin and re-starting smath,
the installer dialog pops up and offers installation. You are also informed that you do not need installation if you already have the runtime environment installed. In this case it is natural that the user chooses [Cancel]. Then, however, the dialog pops up every time you launch smath.

If you choose [Install] even if you know that nothing needs to be installed, then suddenly the installer gets aware of that and just finishes and no more messages appear on start-up.


Updated. I think I found a way to do the installation without the use of environment variable path. You can remove the path from the environment variable. Also, you can delete the installation files from c:\MATLAB6p5.
Russia ☭ forever
Viacheslav N. Mezentsev
thanks 1 user thanked uni for this useful post.
on 24/08/2013(UTC)
Offline mkraska  
#22 Posted : 24 August 2013 17:24:53(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,743
Germany

Was thanked: 955 time(s) in 606 post(s)
Works here.

Is there a chance to return scalars if the result matrix has 1x1 size by default (without touching each function separately)? Then we could use all the functions as direct replacement for the standard ones. The drawback would be that these functions cannot handle lists (except if you can recognize that in the plugin and do sys2mat and vice versa.

One could even think of working with any nested structures by flattening them out internally and afterwards restructuring the results according to input.

I guess that unit handling would not be trivial. However, for most special functions this is irrelevant anyways.
Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
Offline uni  
#23 Posted : 24 August 2013 18:36:32(UTC)
uni


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 10/11/2010(UTC)
Posts: 1,331
Man
Russian Federation

Was thanked: 1163 time(s) in 672 post(s)
Originally Posted by: mkraska Go to Quoted Post
Works here.

Is there a chance to return scalars if the result matrix has 1x1 size by default (without touching each function separately)? Then we could use all the functions as direct replacement for the standard ones.


Updated. Done. As for the rest, I'll think about it.
Russia ☭ forever
Viacheslav N. Mezentsev
thanks 2 users thanked uni for this useful post.
on 24/08/2013(UTC),  on 25/08/2013(UTC)
Offline mkraska  
#24 Posted : 24 August 2013 19:17:01(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,743
Germany

Was thanked: 955 time(s) in 606 post(s)
Great! Thanks a lot. Now all functions can be called with scalar argument and return scalars in that case. Is it sensible to include the mw... names in the doc (some of them are already in it) or can they be expected to change to generic ones in the near future?
Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
Offline uni  
#25 Posted : 24 August 2013 19:27:08(UTC)
uni


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 10/11/2010(UTC)
Posts: 1,331
Man
Russian Federation

Was thanked: 1163 time(s) in 672 post(s)
I'm not sure, but I think that next year we will not see support for namespaces and function overloading. Andrey is a very busy. Therefore, I recommend to use prefixes in the documentation.

For now we have 49 functions.

Russia ☭ forever
Viacheslav N. Mezentsev
thanks 3 users thanked uni for this useful post.
on 25/08/2013(UTC),  on 25/08/2013(UTC),  on 25/08/2013(UTC)
Offline mkraska  
#26 Posted : 25 August 2013 00:16:12(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,743
Germany

Was thanked: 955 time(s) in 606 post(s)
Thanks for the information. If the loss of list processing capability would be the only drawback of just re-defining the internal functions, I would vote for that.

Something went wrong with the diagonal scaling. I can't see such a function.
Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
Offline omorr  
#27 Posted : 25 August 2013 11:32:19(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,740
Man
Serbia

Was thanked: 318 time(s) in 268 post(s)
Originally Posted by: uni Go to Quoted Post
I'm not sure, but I think that next year we will not see support for namespaces and function overloading. Andrey is a very busy. Therefore, I recommend to use prefixes in the documentation.

For now we have 49 functions.


Thank you uni Good

As I could see in the first post of yours, when you manage to get those functions to work you are updating the plugin - blue starting with mw... (I counted more than 49 at the moment )Good and the rest in some green - not working yet. By the way, an insignificant observation about - on ode function in the list, mw... prefix is missing.

If you get most of those functions in the list to work in spite of the mentioned obstacles in SMath (different number of arguments etc.), it would be really something big Victory .

Regards,
Radovan
When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
thanks 1 user thanked omorr for this useful post.
on 25/08/2013(UTC)
Offline mkraska  
#28 Posted : 25 August 2013 12:38:43(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,743
Germany

Was thanked: 955 time(s) in 606 post(s)
The functions mwall() and mwany() do not work correctly.

Also, I recommend to modify the calling syntax.

mwany(M) test if any of the elements in M is nonzero. Corresponds to matlab any(M( : )), where the colon operator flattens the matrix into a vector.

mwany(M,1) or mwany(M,2) do the test along the columns and rows respectively (no change except bug fixing required)

Same for mwall.

Edited by user 25 August 2013 12:40:35(UTC)  | Reason: Not specified

File Attachment(s):
mwall.sm (11kb) downloaded 57 time(s).
mwany.sm (11kb) downloaded 51 time(s).
mkraska attached the following image(s):
mwall.PNG
Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
thanks 2 users thanked mkraska for this useful post.
on 25/08/2013(UTC),  on 25/08/2013(UTC)
Offline uni  
#29 Posted : 25 August 2013 13:04:51(UTC)
uni


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 10/11/2010(UTC)
Posts: 1,331
Man
Russian Federation

Was thanked: 1163 time(s) in 672 post(s)
Updated. 83 functions (see log.txt in the plugin folder).
Russia ☭ forever
Viacheslav N. Mezentsev
thanks 2 users thanked uni for this useful post.
on 25/08/2013(UTC),  on 25/08/2013(UTC)
Offline omorr  
#30 Posted : 25 August 2013 14:16:02(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,740
Man
Serbia

Was thanked: 318 time(s) in 268 post(s)
Thank you uni Good
Originally Posted by: uni Go to Quoted Post
Updated. 83 functions (see log.txt in the plugin folder).

I tried it first and could see that mwbesselx() functions do not work. Then I checked the log.txt file.
This is the end of it

....
8/25/2013 12:35:58 PM [INFO ] [Plugin.Initialize()] Successfully. 83 functions loaded.
8/25/2013 12:39:34 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:39:34 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:39:51 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:39:51 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:39:56 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:39:56 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:41:32 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:41:32 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:41:50 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
8/25/2013 12:41:50 PM [ERROR] [Plugin.NumericEvaluation()] mwbesseli(2) Exception has been thrown by the target of an invocation.
....

No success with mwbetainc() yet as well. I suppose you know all this, just wanted to confirm.

Regards,
Radovan
When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
Offline uni  
#31 Posted : 25 August 2013 14:35:46(UTC)
uni


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 10/11/2010(UTC)
Posts: 1,331
Man
Russian Federation

Was thanked: 1163 time(s) in 672 post(s)
Show me log.txt from the \bin\win32\ directory.

Russia ☭ forever
Viacheslav N. Mezentsev
Offline mkraska  
#32 Posted : 25 August 2013 15:29:05(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,743
Germany

Was thanked: 955 time(s) in 606 post(s)
I confirm the problem. The log in bin/win32 says

25.08.2013 14:18:54 [ERROR] [mwbesseli] Failed to load MEX-File : besselmx.
25.08.2013 14:19:00 [ERROR] [mwbesseli] Failed to load MEX-File : besselmx.
25.08.2013 14:19:09 [ERROR] [mwbesselj] Failed to load MEX-File : besselmx.
25.08.2013 14:19:16 [ERROR] [mwbesselj] Failed to load MEX-File : besselmx.
25.08.2013 14:19:20 [ERROR] [mwbesselj] Failed to load MEX-File : besselmx.
25.08.2013 14:20:55 [ERROR] [mwbesseli] Failed to load MEX-File : besselmx.

and I get an error message from SMath (translated)

The program cannot be launched because DFORMD.dll is missing on your computer. Re-install the program to solve the problem.

such a dll is present on my system as part of the ANSYS finite element package.
mkraska attached the following image(s):
bessel.PNG
Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
thanks 1 user thanked mkraska for this useful post.
on 25/08/2013(UTC)
Offline omorr  
#33 Posted : 25 August 2013 16:00:05(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,740
Man
Serbia

Was thanked: 318 time(s) in 268 post(s)
I checked the log.txt file. There is only one message repeating when I call any of the bessel functions

8/25/2013 2:22:55 PM [ERROR] [mwbesseli] Failed to load MEX-File : besselmx.

when I try to use mwbetainc() there is only one message about the first argument

8/25/2013 2:45:00 PM [ERROR] [mwbetainc] X must be in the interval [0,1].

although I used something like this

z:range(0,10)

mwbetainc(0.5,z,3)=#@#

with the usual error message ...-function is not defined

Regards,
Radovan

[EDIT] I do not have a message about missing dll like Martin

Edited by user 25 August 2013 16:03:10(UTC)  | Reason: Not specified

When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
thanks 1 user thanked omorr for this useful post.
on 25/08/2013(UTC)
Offline uni  
#34 Posted : 25 August 2013 16:13:29(UTC)
uni


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 10/11/2010(UTC)
Posts: 1,331
Man
Russian Federation

Was thanked: 1163 time(s) in 672 post(s)
Updated. Try now. As for mwbetainc(), I don't know yet what's going on.
Russia ☭ forever
Viacheslav N. Mezentsev
thanks 1 user thanked uni for this useful post.
on 25/08/2013(UTC)
Offline omorr  
#35 Posted : 25 August 2013 16:21:39(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,740
Man
Serbia

Was thanked: 318 time(s) in 268 post(s)
Originally Posted by: uni Go to Quoted Post
Updated. Try now. As for mwbetainc(), I don't know yet what's going on.

I can confirm that mwbesselx() functions are working for me Good . Thank you.

Regards,
Radovan

Edited by user 25 August 2013 16:24:13(UTC)  | Reason: Not specified

When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
thanks 1 user thanked omorr for this useful post.
on 26/08/2013(UTC)
Offline uni  
#36 Posted : 26 August 2013 12:37:51(UTC)
uni


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 10/11/2010(UTC)
Posts: 1,331
Man
Russian Federation

Was thanked: 1163 time(s) in 672 post(s)
Updated. 112 functions.

Added: mwrand(), mwrandn(), mwsize(), mwconv(), mwconv2(), mwnnz(), mwnonzeos(), mwtrace(), mwconj(), mwxor(), mwtril(), mwtriu(), mwcorrcoef(), mweye(), mwones(), mwzeros(), mwroots(), mwrot90(), mwsum().
Russia ☭ forever
Viacheslav N. Mezentsev
thanks 2 users thanked uni for this useful post.
on 26/08/2013(UTC),  on 26/08/2013(UTC)
Offline mkraska  
#37 Posted : 26 August 2013 16:45:32(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,743
Germany

Was thanked: 955 time(s) in 606 post(s)
I am trying to keep up... ;-)

When you assign a matrix generating function to a variable, then this is done symbolically by default in your plugin. Thus, a random matrix looks differently every time it is used. This behaviour is different from Davide's Statistical tools, where numeric evaluation seems to be default.

I guess that symbols with changing contents at every use are somewhat unexpected for the user. Therefore, I would therefore recommend to set evaluation to numeric by default, unless there is something to gain from symbolic evaluation (which I do not see so far, bit I may be wrong).

Edit: There are multiple implications of that SMath symbolics is unaware of scalar or vector properties of operands. Therefore I urgently recommend to suppress symbolic evaluation for (possibly) matrix valued functions. In the second example, replacing mwones() by Ones() solves the problem (mismatch between symbolic and numeric evaluation)

Edit2: To internally apply eval() in every function should help.

Edited by user 26 August 2013 18:06:15(UTC)  | Reason: Not specified

mkraska attached the following image(s):
rand.PNG
ones.PNG
Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
thanks 1 user thanked mkraska for this useful post.
on 27/08/2013(UTC)
Offline mkraska  
#38 Posted : 27 August 2013 00:57:17(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,743
Germany

Was thanked: 955 time(s) in 606 post(s)
some functions do not work...

Error message in log file: Argument to xxx must not be complex.
Error message in SMath: function is not defined.

I can't see any complex argument.
File Attachment(s):
Section math boolean matlab.sm (26kb) downloaded 52 time(s).
mkraska attached the following image(s):
boolean.PNG
Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
thanks 1 user thanked mkraska for this useful post.
on 27/08/2013(UTC)
Offline omorr  
#39 Posted : 27 August 2013 08:00:14(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,740
Man
Serbia

Was thanked: 318 time(s) in 268 post(s)
Confirmed

log.txt from the plugin folder

the last part

....
8/27/2013 6:50:53 AM [INFO ] [Plugin.Initialize()] Successfully. 112 functions loaded.
8/27/2013 6:52:42 AM [ERROR] [Plugin.NumericEvaluation()] mwxor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:52:42 AM [ERROR] [Plugin.NumericEvaluation()] mwxor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:52:55 AM [ERROR] [Plugin.NumericEvaluation()] mwxor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:52:55 AM [ERROR] [Plugin.NumericEvaluation()] mwxor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:08 AM [ERROR] [Plugin.NumericEvaluation()] mwand(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:08 AM [ERROR] [Plugin.NumericEvaluation()] mwand(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:13 AM [ERROR] [Plugin.NumericEvaluation()] mwxor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:13 AM [ERROR] [Plugin.NumericEvaluation()] mwxor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:13 AM [ERROR] [Plugin.NumericEvaluation()] mwand(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:13 AM [ERROR] [Plugin.NumericEvaluation()] mwand(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:19 AM [ERROR] [Plugin.NumericEvaluation()] mwxor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:19 AM [ERROR] [Plugin.NumericEvaluation()] mwxor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:48 AM [ERROR] [Plugin.NumericEvaluation()] mwor(2) Exception has been thrown by the target of an invocation.
8/27/2013 6:53:48 AM [ERROR] [Plugin.NumericEvaluation()] mwor(2) Exception has been thrown by the target of an invocation.


log.txt from win32 folder

8/27/2013 6:52:42 AM [ERROR] [mwxor] Operands to AND must not be complex.
8/27/2013 6:52:42 AM [ERROR] [mwxor] Operands to AND must not be complex.
8/27/2013 6:52:55 AM [ERROR] [mwxor] Operands to AND must not be complex.
8/27/2013 6:52:55 AM [ERROR] [mwxor] Operands to AND must not be complex.
8/27/2013 6:53:08 AM [ERROR] [mwand] Operands to AND must not be complex.
8/27/2013 6:53:08 AM [ERROR] [mwand] Operands to AND must not be complex.
8/27/2013 6:53:13 AM [ERROR] [mwxor] Operands to AND must not be complex.
8/27/2013 6:53:13 AM [ERROR] [mwxor] Operands to AND must not be complex.
8/27/2013 6:53:13 AM [ERROR] [mwand] Operands to AND must not be complex.
8/27/2013 6:53:13 AM [ERROR] [mwand] Operands to AND must not be complex.
8/27/2013 6:53:19 AM [ERROR] [mwxor] Operands to AND must not be complex.
8/27/2013 6:53:19 AM [ERROR] [mwxor] Operands to AND must not be complex.
8/27/2013 6:53:48 AM [ERROR] [mwor] Operands to OR must not be complex.
8/27/2013 6:53:48 AM [ERROR] [mwor] Operands to OR must not be complex.



Regards,
Radovan



When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
thanks 1 user thanked omorr for this useful post.
on 27/08/2013(UTC)
Offline uni  
#40 Posted : 27 August 2013 12:09:24(UTC)
uni


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 10/11/2010(UTC)
Posts: 1,331
Man
Russian Federation

Was thanked: 1163 time(s) in 672 post(s)
Updated. 136 functions.

mwrand(), mvrandn(), mwand(), mwor(), mwnot(), mwxor() - bugs fixed;



As for symbolic simplifications when dealing with matrices, I think that in this case it's a bug. The multiplication operator believes that the second operand returns a scalar, but this is not true.
uni attached the following image(s):
symbolic simplifications.png
Russia ☭ forever
Viacheslav N. Mezentsev
thanks 1 user thanked uni for this useful post.
on 27/08/2013(UTC)
Users browsing this topic
5 Pages<1234>»
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.