gplab-users Mailing List for GPLAB (Page 3)
Status: Beta
Brought to you by:
gplab
You can subscribe to this list here.
2004 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
(5) |
Sep
(1) |
Oct
|
Nov
(2) |
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2005 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
(7) |
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
|
2006 |
Jan
|
Feb
(5) |
Mar
|
Apr
|
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
(2) |
Nov
|
Dec
|
2007 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
(3) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2008 |
Jan
|
Feb
(1) |
Mar
(3) |
Apr
(3) |
May
|
Jun
(3) |
Jul
(3) |
Aug
|
Sep
(3) |
Oct
|
Nov
|
Dec
(1) |
2009 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(3) |
Jul
(2) |
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
|
2010 |
Jan
(3) |
Feb
(2) |
Mar
|
Apr
|
May
(2) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2011 |
Jan
|
Feb
|
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
|
Dec
|
2012 |
Jan
(1) |
Feb
|
Mar
(2) |
Apr
(1) |
May
(1) |
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(1) |
2013 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(2) |
2014 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(2) |
Nov
|
Dec
|
2015 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2016 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
(1) |
Dec
|
2017 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
(2) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2018 |
Jan
(3) |
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(3) |
Sep
(2) |
Oct
|
Nov
|
Dec
|
2021 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Rosanny S. <sa...@is...> - 2008-07-01 05:21:28
|
hi all, i'm having problem with adjusting the crossover probability. if i wanna set my crossover probability to 90%, how can i do this? i have read the manual, but i still couldn't understand how -- regards, Rosanny Sihombing |
From: Murat K. <mka...@in...> - 2008-06-26 14:08:45
|
Dear Sara, Thank you for your tips. It is quite easy and quick. I used to convert eqs. from the tree which was very confusing. Cheers, Murat Karakus Sara Silva wrote: > Dear Rosanny, > > Among those parameters you see at the end there is the equation of the > best individual found during the run. Namely, you can find it in: > > state.bestsofar.str > > Alternatively, you can call gplab with two output parameters, for > example [v,b]=gplab(100,50) and the best individual will be stored in b. > To access its equation you just have to do: > > b.str > > I hope this helps! > Cheers, > Sara > > > > Rosanny Sihombing wrote: > >> hi all, >> >> i don't understand how to show the equation result after running gplab >> function. >> for example, if i'm running gplab(100,50) with exp_x.txt as the input and >> exp_y.txt as the output, MATLAB will show me some parameters in the end of >> the running. but it didn't show the equation result. >> >> how can i get the equation result? i'm sorry that i'm a newbie. >> >> >> ---- >> really need a help on this, >> Rosanny Sihombing >> >> >> ------------------------------------------------------------------------- >> Check out the new SourceForge.net Marketplace. >> It's the best place to buy or sell services for >> just about anything Open Source. >> http://sourceforge.net/services/buy/index.php >> _______________________________________________ >> gplab-users mailing list >> gpl...@li... >> https://lists.sourceforge.net/lists/listinfo/gplab-users >> >> >> > > ------------------------------------------------------------------------- > Check out the new SourceForge.net Marketplace. > It's the best place to buy or sell services for > just about anything Open Source. > http://sourceforge.net/services/buy/index.php > _______________________________________________ > gplab-users mailing list > gpl...@li... > https://lists.sourceforge.net/lists/listinfo/gplab-users > > |
From: Sara S. <sa...@de...> - 2008-06-26 13:49:30
|
Dear Rosanny, Among those parameters you see at the end there is the equation of the best individual found during the run. Namely, you can find it in: state.bestsofar.str Alternatively, you can call gplab with two output parameters, for example [v,b]=gplab(100,50) and the best individual will be stored in b. To access its equation you just have to do: b.str I hope this helps! Cheers, Sara Rosanny Sihombing wrote: > hi all, > > i don't understand how to show the equation result after running gplab > function. > for example, if i'm running gplab(100,50) with exp_x.txt as the input and > exp_y.txt as the output, MATLAB will show me some parameters in the end of > the running. but it didn't show the equation result. > > how can i get the equation result? i'm sorry that i'm a newbie. > > > ---- > really need a help on this, > Rosanny Sihombing > > > ------------------------------------------------------------------------- > Check out the new SourceForge.net Marketplace. > It's the best place to buy or sell services for > just about anything Open Source. > http://sourceforge.net/services/buy/index.php > _______________________________________________ > gplab-users mailing list > gpl...@li... > https://lists.sourceforge.net/lists/listinfo/gplab-users > > |
From: Rosanny S. <sa...@is...> - 2008-06-26 04:52:11
|
hi all, i don't understand how to show the equation result after running gplab function. for example, if i'm running gplab(100,50) with exp_x.txt as the input and exp_y.txt as the output, MATLAB will show me some parameters in the end of the running. but it didn't show the equation result. how can i get the equation result? i'm sorry that i'm a newbie. ---- really need a help on this, Rosanny Sihombing |
From: Wasif A. <mwa...@ya...> - 2008-04-06 16:53:36
|
Hi Sara, Thanks for taking time out. I do not set 'realmaxlevel' prior to the run, so it is empty by default and should be filled in automatically. I run gplab for 200 generations with number of individuals kept constant to 30 in each generation. In generation 194, it displays "increasing maximum depth to 20", but I have 'dynamiclevel' set to '0', so I guess it should not have dynamic maximum tree limit on. Here are the parameters: maxresources: [] functions: {6x2 cell} terminals: {} numvars: [] autovars: 1 initpoptype: 'rampedinit' expected: 'rank85' elitism: 'replace' survival: 'fixedpopsize' dynamicresources: '0' resourcespopsize: 'steady' resourcesfitness: 'normal' periode: 1 ajout: 'M1' sampling: 'lexictour' drawperspin: [] calcfitness: 'regfitness' adjustfitness: [] precision: 12 lowerisbetter: 1 gengap: [] savetofile: 'never' operatorprobstype: 'fixed' initialprobstype: 'fixed' reproduction: 0.1000 operatornames: {'crossover' 'mutation'} operatornparents: [2 1] operatornchildren: [2 1] initialfixedprobs: [0.8000 0.1000] initialvarprobs: [] savedir: [] savename: [] datafilex: [] datafiley: [] usetestdata: 0 testdatafilex: [] testdatafiley: [] files2data: 'xy2inout' numbackgen: 3 percentback: 0.2500 percentchange: 0.2500 adaptinterval: [] adaptwindowsize: [] smalldifference: [] minprob: 0.1000 tournamentsize: [] depthnodes: '1' fixedlevel: 1 dynamiclevel: '0' inicmaxlevel: [] inicdynlevel: [] realmaxlevel: [] veryheavy: 0 calccomplexity: 0 calcdiversity: {} hits: [100 0] output: 'normal' graphics: {} keepevalssize: [] filters: {} Please let me know if you have any findings. Best Regards, Wasif Afzal Sara Silva <sa...@de...> wrote: Hello, I would need to see other parameters related to the limits, like 'realmaxlevel'. Something is definitely wrong. Can you send me the list of all your parameters? Thanks. Sara Wasif Afzal wrote: > Hi All, > > I am running a simple symbolic regression problem in GPLAB. I have two > files, one with input data and other with output data, which I use to > start the GPLAB run. > > Before startting the run, I have set the following parameters: > > fixedlevel=1 > dynamiclevel='0' > depthnodes= '1' > > The above setting means that I have the {'strictdepth'} filter on, which > should reject an individual deeper than the maximum allowed depth, which > is determined by the parameter realmaxlevel and equals 17 (for > depthnodes='1'). > > I run GPLAB for 100 generations and 150 population size which remains > fixed in all generations. > > During several of the generations, I observe that there are individuals > whose depths are greater than 17 and are being displayed by GPLAB. In > fact, I see that at the end of the run, the best so far individual has > depth 31. My expectation was that GPLAB should reject individuals having > depth greater than 17, based on the parameter settings. > > Please let me know if anybody knows the reason behind such a behavior or > if I have done a mistake in understanding parameter settings. > > Best Regards, > Wasif > > > > > ------------------------------------------------------------------------ > Yahoo! for Good helps you make a difference > > > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > Check out the new SourceForge.net Marketplace. > It's the best place to buy or sell services for > just about anything Open Source. > http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace > > > ------------------------------------------------------------------------ > > _______________________________________________ > gplab-users mailing list > gpl...@li... > https://lists.sourceforge.net/lists/listinfo/gplab-users ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Register now and save $200. Hurry, offer ends at 11:59 p.m., Monday, April 7! Use priority code J8TLD2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ gplab-users mailing list gpl...@li... https://lists.sourceforge.net/lists/listinfo/gplab-users --------------------------------- Yahoo! for Good helps you make a difference |
From: Sara S. <sa...@de...> - 2008-04-06 15:36:03
|
Hello, I would need to see other parameters related to the limits, like 'realmaxlevel'. Something is definitely wrong. Can you send me the list of all your parameters? Thanks. Sara Wasif Afzal wrote: > Hi All, > > I am running a simple symbolic regression problem in GPLAB. I have two > files, one with input data and other with output data, which I use to > start the GPLAB run. > > Before startting the run, I have set the following parameters: > > fixedlevel=1 > dynamiclevel='0' > depthnodes= '1' > > The above setting means that I have the {'strictdepth'} filter on, which > should reject an individual deeper than the maximum allowed depth, which > is determined by the parameter realmaxlevel and equals 17 (for > depthnodes='1'). > > I run GPLAB for 100 generations and 150 population size which remains > fixed in all generations. > > During several of the generations, I observe that there are individuals > whose depths are greater than 17 and are being displayed by GPLAB. In > fact, I see that at the end of the run, the best so far individual has > depth 31. My expectation was that GPLAB should reject individuals having > depth greater than 17, based on the parameter settings. > > Please let me know if anybody knows the reason behind such a behavior or > if I have done a mistake in understanding parameter settings. > > Best Regards, > Wasif > > > > > ------------------------------------------------------------------------ > Yahoo! for Good helps you make a difference > <http://us.rd.yahoo.com/mailuk/taglines/isp/control/*http://us.rd.yahoo.com/evt=51947/*http://uk.promotions.yahoo.com/forgood/> > > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > Check out the new SourceForge.net Marketplace. > It's the best place to buy or sell services for > just about anything Open Source. > http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace > > > ------------------------------------------------------------------------ > > _______________________________________________ > gplab-users mailing list > gpl...@li... > https://lists.sourceforge.net/lists/listinfo/gplab-users |
From: Wasif A. <mwa...@ya...> - 2008-04-04 15:31:39
|
Hi All, I am running a simple symbolic regression problem in GPLAB. I have two files, one with input data and other with output data, which I use to start the GPLAB run. Before startting the run, I have set the following parameters: fixedlevel=1 dynamiclevel='0' depthnodes= '1' The above setting means that I have the {'strictdepth'} filter on, which should reject an individual deeper than the maximum allowed depth, which is determined by the parameter realmaxlevel and equals 17 (for depthnodes='1'). I run GPLAB for 100 generations and 150 population size which remains fixed in all generations. During several of the generations, I observe that there are individuals whose depths are greater than 17 and are being displayed by GPLAB. In fact, I see that at the end of the run, the best so far individual has depth 31. My expectation was that GPLAB should reject individuals having depth greater than 17, based on the parameter settings. Please let me know if anybody knows the reason behind such a behavior or if I have done a mistake in understanding parameter settings. Best Regards, Wasif --------------------------------- Yahoo! for Good helps you make a difference |
From: Sara S. <sa...@de...> - 2008-03-17 16:29:52
|
Dear Santiago, I read you email twice and I still don't understand one thing: are these problems specifically related to GPLAB, or is it GP in general? It seems to me that the fitness function may be what's preventing GPLAB from finding usable solutions. I am not familiar with the pendulum application, but if you give a high fitness (10^10) to unacceptable (unstable) solutions, then you must guarantee that a low fitness is only given to good solutions. The fitness function is what drives the evolution - it tells the search process where to go. If you believe your problems are not directly related to GPLAB, but to the search process in general, may I suggest you ask for suggestions in the GP mailing list: http://tech.groups.yahoo.com/group/genetic_programming/ Having said so, I may also suggest that you try several different depth/size limits for bloat control in GPLAB. I believe you may already be familiar with them. I wonder if you are using very low limits that are not allowing GP to move to better searching areas? Just an idea. I am not familiar with simulink or your application, so there is not much else I can do to help... Good luck! Sara Santiago Remolina wrote: > Hello GPLAB users, > > This semester I am working on by bachelor's thesis on the Hamburg > University of Technology. My project is to find a stabilizing and swing > up controller for a double inverted pendulum, which consists basically > on a pendulum hanging from a rotating arm using Genetic Programming. > After researching a little, I found out that GPLAB is the best or maybe > the only matlab tool to use GP. Unfortunately after two months, in which > most of my time was spent in getting to understand GPLAB and to know the > different options, I haven't been able to find any usable controller. I > have tried changing almos every parameter, and even using some > "techniques" that I have come up with. Like changing the methods used in > the sampling, elitism and expected parameters. I have also tried > combining the best individuals of different runs, and starting GPLAB > with them (to have a larger variety in my population). I changed the way > GPLAB calculates the fitness of an individual, by integrating the error > signal of the two angles describing the position of the arm and pedulum > that a simulink model gives as an output. The definition of the fitness > function has been one of the biggest problems that I have had. Since I > have been giving unstable functions a very bad fitness (e.a. 10^10), > GPLAB thinks that controllers whose final value is zero ( cos(u[1] - > cos(u[1]) ), and therefore don't make my system unstable, are good > controllers. This problem not only happens with "zero" controllers but > also with controller whose output isvery small, and therefore act almost > like "zero" controllers. > > Anyway, my question would be if anyone has also used GPLAB interacting > with simulink, if they had any recomendations. And if anyone had some > suggestions of how I could play with the different parameters that > GPLAB offers to get better results, or other techniques. I have many > other questions but these would be the ones at the top of my list, so I > hope I could explain the problem that I have, and would appreciate it if > I got any answers. > > Santiago > > ------------------------------------------------------------------------ > Shed those extra pounds with MSN and The Biggest Loser! Learn more. > <http://biggestloser.msn.com/> > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > > ------------------------------------------------------------------------ > > _______________________________________________ > gplab-users mailing list > gpl...@li... > https://lists.sourceforge.net/lists/listinfo/gplab-users |
From: Santiago R. <san...@ho...> - 2008-03-15 14:14:26
|
Hello GPLAB users, This semester I am working on by bachelor's thesis on the Hamburg University of Technology. My project is to find a stabilizing and swing up controller for a double inverted pendulum, which consists basically on a pendulum hanging from a rotating arm using Genetic Programming. After researching a little, I found out that GPLAB is the best or maybe the only matlab tool to use GP. Unfortunately after two months, in which most of my time was spent in getting to understand GPLAB and to know the different options, I haven't been able to find any usable controller. I have tried changing almos every parameter, and even using some "techniques" that I have come up with. Like changing the methods used in the sampling, elitism and expected parameters. I have also tried combining the best individuals of different runs, and starting GPLAB with them (to have a larger variety in my population). I changed the way GPLAB calculates the fitness of an individual, by integrating the error signal of the two angles describing the position of the arm and pedulum that a simulink model gives as an output. The definition of the fitness function has been one of the biggest problems that I have had. Since I have been giving unstable functions a very bad fitness (e.a. 10^10), GPLAB thinks that controllers whose final value is zero ( cos(u[1] - cos(u[1]) ), and therefore don't make my system unstable, are good controllers. This problem not only happens with "zero" controllers but also with controller whose output isvery small, and therefore act almost like "zero" controllers. Anyway, my question would be if anyone has also used GPLAB interacting with simulink, if they had any recomendations. And if anyone had some suggestions of how I could play with the different parameters that GPLAB offers to get better results, or other techniques. I have many other questions but these would be the ones at the top of my list, so I hope I could explain the problem that I have, and would appreciate it if I got any answers. Santiago _________________________________________________________________ Shed those extra pounds with MSN and The Biggest Loser! http://biggestloser.msn.com/ |
From: Sara S. <sa...@de...> - 2008-03-01 18:10:55
|
Hi Josh, GPLAB does not currently implement strong type checking. I may have some (unofficial) code that implements that feature. I will look for it (it may take a while because my hard drive is always a mess :)) and then send you whatever I find. Be prepared for unpolished and undocumented coding, though... Sara Josh Eads wrote: > Hi, > > On my current project I need the ability to only allow certain functions > in the GP tree to be able to take other functions or terminals as input > (strongly typed). I have used GPLab so far for this project and have > not seen any functionality to do this and have not found anything in the > documentation either. I was wondering if there is any known method for > enabling strongly typed functions or anyone that has experience with > this issue before I work on adding the feature myself. > > --Josh > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > gplab-users mailing list > gpl...@li... > https://lists.sourceforge.net/lists/listinfo/gplab-users > > |
From: Josh E. <jm...@ms...> - 2008-02-26 21:02:56
|
Hi, On my current project I need the ability to only allow certain functions in the GP tree to be able to take other functions or terminals as input (strongly typed). I have used GPLab so far for this project and have not seen any functionality to do this and have not found anything in the documentation either. I was wondering if there is any known method for enabling strongly typed functions or anyone that has experience with this issue before I work on adding the feature myself. --Josh |
From: Sara S. <sa...@de...> - 2007-07-11 23:06:16
|
Dear James, Sorry for the delay in replying. I am currently writing my PhD thesis so I am VERY busy. But I will try to help. You say that GPLAB does converge. Does it mean that it reaches an optimal solution according to your fitness function? Also, I cannot imagine what type of problem you are having in interpreting the final tree. Can you give an example? Cheers, Sara James Griffin wrote: > I am currently having problems to get a useful output from the GPlab. I > realise this is due to my understanding of the problem and wondered if > anyone could help me get started on this. I have some useful ideas and > already classified my data using fuzzy clustering. To that, I would like > to classify the data using GP as I have worked on this for the last 4 > months and don’t really want to say it can’t classify the data when in > fact it’s down to my understanding and not the strategy behind GP. > The problem: > > If I have 30 cases of data - for argument sake, the 1st ten represent > classification phenomenon 1, the next ten (11-20) represent > classification phenomenon 2 and lastly, the last ten (21-30), represent > classification phenomenon 3. In short I wish to segregate classes 1 to 3 > using ‘if,’ ‘else,’ ‘if greater than,’ and ‘less than rules.’ I use two > text inputs one with the data (120*30 matrix) and one with the targets 1 > to 3. I have got useful rules from tree regression software (Matlab) > however not GP, I believe it’s my fitness function (standard GP toolbox) > and terminals are currently set to nil. The functions are as stated > above (implemented my own ‘min,’ ‘max,’ kurtosis,’ ‘stddev,’ although > these just confused matters. > > The GP lab is converging however I am having problems in interpreting > the output string/tree for classifying the 3 classes of data (it would > be useful to get an output similar to the NN GP output against desired > output. Each of the 30 cases are 120 pts in size and each of the classes > are significantly different from the other classes. I.e. first class has > first 30 pts larger than the 2nd batch and the 3rd batch has 60 -90 pts > both larger than the other classes. > > Based on this problem I would like to use the fitness measure to > continually update a correct classification and penalize a > misclassification. Perhaps a figure of merit could be associated to > ensure the best individual is found. Can anyone help me getting started > with such problem? I.e what functions and terminals do they think are > appropriate (I am looking at le, gr and myif), not sure what terminals > to use ??? The fitness function should look if the rules/tree segregates > the data giving the correct classification .....to this end, I believe > something similar to the ant demo is required but not as complex. > > I would like to express my thanks to Sara for producing such a user > friendly and powerful tool box and once I get into this in greater > detail I would like to share some ideas I have for enabling your GP > toolbox to handle n-dimensionally huge data ??? Although automating may > take some effort and that's of course if you haven't already done this! > Again I would be grateful for any help any of the GP community can give > as I have been mucking around with the toolbox for 2 months now and I > can't seem to replicate John Koza/Daniel Howards GP classification ideas > (they classify imagery data - although mine is mechanical measurements > and Acoustic Emission and a lot smaller in size due to n-dimensional > reduction). > > _________________________________________________________________ > Win tickets to the sold out Live Earth concert! > http://liveearth.uk.msn.com > > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > > > ------------------------------------------------------------------------ > > _______________________________________________ > gplab-users mailing list > gpl...@li... > https://lists.sourceforge.net/lists/listinfo/gplab-users |
From: James G. <jam...@ho...> - 2007-07-07 19:27:53
|
I am currently having problems to get a useful output from the GPlab. I realise this is due to my understanding of the problem and wondered if anyone could help me get started on this. I have some useful ideas and already classified my data using fuzzy clustering. To that, I would like to classify the data using GP as I have worked on this for the last 4 months and dont really want to say it cant classify the data when in fact its down to my understanding and not the strategy behind GP. The problem: If I have 30 cases of data - for argument sake, the 1st ten represent classification phenomenon 1, the next ten (11-20) represent classification phenomenon 2 and lastly, the last ten (21-30), represent classification phenomenon 3. In short I wish to segregate classes 1 to 3 using if, else, if greater than, and less than rules. I use two text inputs one with the data (120*30 matrix) and one with the targets 1 to 3. I have got useful rules from tree regression software (Matlab) however not GP, I believe its my fitness function (standard GP toolbox) and terminals are currently set to nil. The functions are as stated above (implemented my own min, max, kurtosis, stddev, although these just confused matters. The GP lab is converging however I am having problems in interpreting the output string/tree for classifying the 3 classes of data (it would be useful to get an output similar to the NN GP output against desired output. Each of the 30 cases are 120 pts in size and each of the classes are significantly different from the other classes. I.e. first class has first 30 pts larger than the 2nd batch and the 3rd batch has 60 -90 pts both larger than the other classes. Based on this problem I would like to use the fitness measure to continually update a correct classification and penalize a misclassification. Perhaps a figure of merit could be associated to ensure the best individual is found. Can anyone help me getting started with such problem? I.e what functions and terminals do they think are appropriate (I am looking at le, gr and myif), not sure what terminals to use ??? The fitness function should look if the rules/tree segregates the data giving the correct classification .....to this end, I believe something similar to the ant demo is required but not as complex. I would like to express my thanks to Sara for producing such a user friendly and powerful tool box and once I get into this in greater detail I would like to share some ideas I have for enabling your GP toolbox to handle n-dimensionally huge data ??? Although automating may take some effort and that's of course if you haven't already done this! Again I would be grateful for any help any of the GP community can give as I have been mucking around with the toolbox for 2 months now and I can't seem to replicate John Koza/Daniel Howards GP classification ideas (they classify imagery data - although mine is mechanical measurements and Acoustic Emission and a lot smaller in size due to n-dimensional reduction). _________________________________________________________________ The next generation of Hotmail is here! http://www.newhotmail.co.uk/ |
From: James G. <jam...@ho...> - 2007-07-07 19:20:57
|
I am currently having problems to get a useful output from the GPlab. I realise this is due to my understanding of the problem and wondered if anyone could help me get started on this. I have some useful ideas and already classified my data using fuzzy clustering. To that, I would like to classify the data using GP as I have worked on this for the last 4 months and dont really want to say it cant classify the data when in fact its down to my understanding and not the strategy behind GP. The problem: If I have 30 cases of data - for argument sake, the 1st ten represent classification phenomenon 1, the next ten (11-20) represent classification phenomenon 2 and lastly, the last ten (21-30), represent classification phenomenon 3. In short I wish to segregate classes 1 to 3 using if, else, if greater than, and less than rules. I use two text inputs one with the data (120*30 matrix) and one with the targets 1 to 3. I have got useful rules from tree regression software (Matlab) however not GP, I believe its my fitness function (standard GP toolbox) and terminals are currently set to nil. The functions are as stated above (implemented my own min, max, kurtosis, stddev, although these just confused matters. The GP lab is converging however I am having problems in interpreting the output string/tree for classifying the 3 classes of data (it would be useful to get an output similar to the NN GP output against desired output. Each of the 30 cases are 120 pts in size and each of the classes are significantly different from the other classes. I.e. first class has first 30 pts larger than the 2nd batch and the 3rd batch has 60 -90 pts both larger than the other classes. Based on this problem I would like to use the fitness measure to continually update a correct classification and penalize a misclassification. Perhaps a figure of merit could be associated to ensure the best individual is found. Can anyone help me getting started with such problem? I.e what functions and terminals do they think are appropriate (I am looking at le, gr and myif), not sure what terminals to use ??? The fitness function should look if the rules/tree segregates the data giving the correct classification .....to this end, I believe something similar to the ant demo is required but not as complex. I would like to express my thanks to Sara for producing such a user friendly and powerful tool box and once I get into this in greater detail I would like to share some ideas I have for enabling your GP toolbox to handle n-dimensionally huge data ??? Although automating may take some effort and that's of course if you haven't already done this! Again I would be grateful for any help any of the GP community can give as I have been mucking around with the toolbox for 2 months now and I can't seem to replicate John Koza/Daniel Howards GP classification ideas (they classify imagery data - although mine is mechanical measurements and Acoustic Emission and a lot smaller in size due to n-dimensional reduction). _________________________________________________________________ Win tickets to the sold out Live Earth concert! http://liveearth.uk.msn.com |
From: Sara S. <sa...@de...> - 2007-04-27 23:53:35
|
Dear GPLAB users, GPLAB version 3 is now available for download at http://gplab.sourceforge.net/download.html As always, comments and suggestions are welcome. Thanks! Cheers, Sara Silva |
From: Sara S. <sa...@de...> - 2006-10-26 15:32:54
|
Dear GPLAB users, Just when you thought this new version was finally bug-free... voilà! There were error messages whenever a test data set was used. Two updated files solve the problem. You only need them if you are using GPLAB 2.1. You can access the bug fixes directly here: http://prdownloads.sourceforge.net/gplab/gplab.2.1.bugfixes.20061024.zip?download Or you can visit the GPLAB download page: http://gplab.sourceforge.net/download.html Cheers, Sara |
From: Sara S. <sa...@de...> - 2006-10-08 01:00:00
|
Dear GPLAB users, A new release of GPLAB is available for download here: http://gplab.sourceforge.net/download.html Any comments and suggestions are always welcome. Thanks! Cheers, Sara Silva http://eden.dei.uc.pt/~sara/ PS - If you wonder why it took me so long to upload a new release, check my hobbies page... :) |
From: chinmay <chi...@gm...> - 2006-06-15 03:15:12
|
Hi ! Your friend chinmay from Surat India, just invited you to his online photo albums and journals at Zorpia.com. So what is Zorpia? It is an online community that allows you to upload unlimited amount of photos, write journals and make friends. We also have a variety of skins in store for you so that you can customize your homepage freely. Join now for free! Please click the following link to join Zorpia: http://www.zorpia.com/cgi/signup.cgi?invitation_key=00075fa6c41feab7e6dd5639170570cc&referral=chinmaydesai This message was delivered with the chinmay's initiation. If you wish to discontinue receiving invitations from us, please click the following link: http://www.zorpia.com/cgi/signup.cgi?action=optout&email=gpl...@li... |
From: selva k. <sel...@ya...> - 2006-02-10 07:19:57
|
hi everyone, Now i will explain my problem in detail and somebody plz give me a clear visualisation as early as possible. I wrote a MATLAB CODE to calculate mean and stddeviation which are the inputs(terminals) to GPLAB CODE, on moving window fashion which slides through entire image. This is done by a FOR loop (for window to slide entire image) . i read in GPLAB MANUAL that these inputs must be stored in a file. Now my doubt is ,for each execution of the main FOR LOOP the values must be stored in file. I need a clear visualisation for this procedure. just for trial i stored arbitrary values in a file and ran a GP program . it says number of columns must be same. i cant understand this .plz explian. If possible plz send some basic source codes of GPLAB related to this. Iam awaiting your reply . plz respond soon. yours, selva --------------------------------- Yahoo! Mail Use Photomail to share photos without annoying attachments. |
From: Sara S. <sa...@de...> - 2006-02-07 18:12:13
|
hi there, sorry if i take a long time to reply now - i had a baby very recently and now i am on maternity leave. i will keep responding, though, but may take longer. to selva kumar, your question is related to using matlab, not gplab. sorry, i don't have enough spare time to do this for you. you may want to read matlab's user's manual. you can also request help in the matlab newsgroup: comp.soft-sys.matlab good luck! sara selva kumar wrote: > i have the following questions anybody please help me: > > 1: i need to give mean & stddeviation of moving window as input to GP, > for that i must save them in seperate file . how can i accomplish > this in MATLAB > > from > selvakumar > > ------------------------------------------------------------------------ > Brings words and photos together (easily) with > PhotoMail > <http://us.rd.yahoo.com/mail_us/taglines/PMDEF3/*http://photomail.mail.yahoo.com> > - it's free and works with Yahoo! Mail. |
From: Sara S. <sa...@de...> - 2006-02-07 18:03:57
|
hi again, everything you need is explained in gplab's user's manual, available here: http://prdownloads.sourceforge.net/gplab/gplab.manual.2.pdf?download please let me know if you have any specific questions. cheers, sara selva kumar wrote: > hello > > iam a beginner of gplab. would you please help with some basic gplab > source codes like > 1: giving input > 2: setting parameters > 3. evaluating fitness etc > > it wold be very helpful for me. plz send as early as possible > > > from > &nb sp; > > selvakumar > > ------------------------------------------------------------------------ > Brings words and photos together (easily) with > PhotoMail > <http://us.rd.yahoo.com/mail_us/taglines/PMDEF3/*http://photomail.mail.yahoo.com> > - it's free and works with Yahoo! Mail. |
From: selva k. <sel...@ya...> - 2006-02-06 05:13:08
|
hello iam a beginner of gplab. would you please help with some basic gplab source codes like 1: giving input 2: setting parameters 3. evaluating fitness etc it wold be very helpful for me. plz send as early as possible from selvakumar --------------------------------- Brings words and photos together (easily) with PhotoMail - it's free and works with Yahoo! Mail. |
From: selva k. <sel...@ya...> - 2006-02-06 05:11:50
|
i have the following questions anybody please help me: 1: i need to give mean & stddeviation of moving window as input to GP, for that i must save them in seperate file . how can i accomplish this in MATLAB from selvakumar --------------------------------- Brings words and photos together (easily) with PhotoMail - it's free and works with Yahoo! Mail. |
From: chinmay d. <chi...@gm...> - 2005-08-19 05:48:01
|
Dear receiver , I the sender of this e-mail is research scholer at IIT=20 Powai(Indian Institute of Technology),INDIA.We are working on the project i= n=20 which Genetic programming will use as predicition tool. Sir i wanted to kno= w=20 which function shold use to create a separate file of GP predicted data. Sir, this is thw crusial junction of our project so please do reply me. Thanking you, with regrads.=20 --=20 Mr.Chinmay K. Desai, Lecturer Mechnical Engineering Department at C.K.P.CET,SURAT, 11,Suruchi Society,Near Jamana Nagar Society,Near St'Xavier's High school, Ghod-dod Road,Surat-395001,State-Gujarat,INDIA. |
From: Sara S. <sa...@de...> - 2005-07-13 13:49:31
|
Hi again, I'm not sure if you are following the latest developments on the problem. Here is the MATLAB newsgroup thread link: http://newsreader.mathworks.com/WebX?14@LK1Ca8Bf0b4@.ef0d8ca I think there may be a solution in sight... :) Cheers, Sara Sara Silva wrote: > > > Aurélien Boffy wrote: > >> ... >> So, it is us who thank you for your wonderful library and long live >> GPLab! > > > Well, thank you! > I am glad GPLAB has been useful to you, not just a source of problems. :) > >> Concerning the problem with the "eval" function: >> >> Someone has already replied to your message in the MATLAB newsgroup >> and has >> suggested to pass strings for the function names. >> We had already tried this idea and that works. >> ... > > > Yes, it works for the artificial ant problem, but it ruins the > evaluation of simple symbolic regression problems, or boolean problems, > or any other problems that use MATLAB's built-in functions. > > For example: > s1='plus(plus(2,2),times(3,1))' > s2='plus(''plus(''2'',''2'')'',''times(''3'',''1'')'')' > > s2 is how s1 would have to be coded if using strings for the function > names. While s1 evaluates nicely and fast, without the need to code > anything else, s2 requires that plus and times are rewritten and > overloaded so they can be evaluated. Otherwise it doesn't work. > > >> eval(s1) > ans = 7 > > >> eval(s2) > ??? Error: ")" expected, "numeric value" found. > > So you understand why this is not a good solution either. The string > must be used as it is built, with no further modifications. If that is > not possible, I would rather perform the evaluation using the tree > instead of the string, as you did. I think it may be possible to use it > while avoiding any "hard-coding" of the function names (so that new > functions can be added one day without causing problems). > > I am going to try a few other things (I am not expecting any miracle > answers from the MATLAB newsgroup) and if I conclude it is not possible > to use the string without causing problems, I will code the evaluation > using the tree (something using both your code and my code). This way > only the ant evaluation will be afected, although I may need to add a > few arguments to other related functions. > > Boy, this has made me realize a really important feature of MATLAB that > I was totally unaware of! > > Thanks for all your help and suggestions. Do you want to become GPLAB > developers? :-) > > Sara > > PS - Sorry to all the mailing list subscribers that are not interested > in the artificial ant. But others may be interested in the developments > of this problem. > > > > > > ------------------------------------------------------- > This SF.Net email is sponsored by the 'Do More With Dual!' webinar > happening > July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual > core and dual graphics technology at this free one hour event hosted by > HP, AMD, and NVIDIA. To register visit http://www.hp.com/go/dualwebinar > _______________________________________________ > gplab-users mailing list > gpl...@li... > https://lists.sourceforge.net/lists/listinfo/gplab-users > > |