Quantcast

Nature Inspired Algorithms

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Nature Inspired Algorithms

Fernando Lavin
Hi everyone,

Is it possible performing " Nature Inspired Algorithms" in Weka? If yes, then how to do that? If no, what it the best way to implement such approach?

Thank you.
Fernndo

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

sujata yumnam
I would also like to know about the same!

Thanks and regards,
S

Thanks and regards
Sujata Sinha


On Tue, Jan 10, 2017 at 3:27 PM, Fernando Lavin <[hidden email]> wrote:
Hi everyone,

Is it possible performing " Nature Inspired Algorithms" in Weka? If yes, then how to do that? If no, what it the best way to implement such approach?

Thank you.
Fernndo

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html



_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

Eibe Frank-2
Administrator
Looking at the list of official packages at

  http://weka.sourceforge.net/packageMetaData/

there is

  EvolutionarySearch Attribute selection An Evolutionary Algorithm (EA) to explore the space of attributes.  
  GPAttributeGeneration Classification, Preprocessing Genetic Programming Attribute Generation
  MultiObjectiveEvolutionaryFuzzyClassifier Classification MultiObjectiveEvolutionaryFuzzyClassifier
  MultiObjectiveEvolutionarySearch Attribute selection An Multi-objective Evolutionary Algorithm (MOEA) to explore the attribute space.
  attributeSelectionSearchMethods Attribute selection Four search methods for attribute selection: ExhaustiveSearch, GeneticSearch, RandomSearch and RankSearch.
  PSOSearch Attribute selection An implementation of the Particle Swarm Optimization (PSO) algorithm to explore the space of attributes.

There are also various neural network packages, in addition to the MultilayerPerceptron classifier that comes with the WEKA distribution:

  SelfOrganizingMap Clustering Cluster data using the Kohonen's Self-Organizing Map algorithm.
  multiLayerPerceptrons Classification/regression, Preprocessing This package currently contains classes for training multilayer perceptrons with one hidden layer for classification and regression, and autoencoders.
  multilayerPerceptronCS Classification An extension of the standard MultilayerPerceptron classifier in Weka that adds context-sensitive Multiple Task Learning (csMTL)
  wekaDeeplearning4j Classification/Regression Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPU Classification/Regression Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPULibs Classification/Regression CPU native libraries for wekaDeeplearning4j
  wekaDeeplearning4jCore Classification/Regression Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPU Classification/Regression Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPULibs Classification/Regression CPU native libraries for wekaDeeplearning4j

The list of unofficial packages at

  http://weka.wikispaces.com/Unofficial+packages+for+WEKA

has some more stuff:

  Java neural network package Java (convolutional or fully-connected) neural network implementation with plugin for Weka.

Assuming you have R and the RPlugin package for WEKA installed and set up correctly, you can also access the nature-inspired classification and regression methods in R that are available via the MLR package (https://mlr-org.github.io/mlr-tutorial/release/html/integrated_learners/index.html). Finally, if Python is your friend, you can use the WekaPyScript package to implement nature-inspired algorithms that are available in Python.

The above list may not be complete. For example, I haven’t included some fuzzy methods because I’m not sure if they are “nature inspired”.

Cheers,
Eibe

> On 11/01/2017, at 3:30 AM, sujata sinha <[hidden email]> wrote:
>
> I would also like to know about the same!
>
> Thanks and regards,
> S
>
> Thanks and regards
> Sujata Sinha
>
>
> On Tue, Jan 10, 2017 at 3:27 PM, Fernando Lavin <[hidden email]> wrote:
> Hi everyone,
>
> Is it possible performing " Nature Inspired Algorithms" in Weka? If yes, then how to do that? If no, what it the best way to implement such approach?
>
> Thank you.
> Fernndo
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
>
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

Fernando Lavin

Dear Eibe,

Thank you so much for the very helpful information.

Best,
Fernando

On 12 Jan 2017 9:36 am, "Eibe Frank" <[hidden email]> wrote:
Looking at the list of official packages at

  http://weka.sourceforge.net/packageMetaData/

there is

  EvolutionarySearch                                            Attribute selection                             An Evolutionary Algorithm (EA) to explore the space of attributes.
  GPAttributeGeneration                                         Classification, Preprocessing           Genetic Programming Attribute Generation
  MultiObjectiveEvolutionaryFuzzyClassifier             Classification                                  MultiObjectiveEvolutionaryFuzzyClassifier
  MultiObjectiveEvolutionarySearch                      Attribute selection                             An Multi-objective Evolutionary Algorithm (MOEA) to explore the attribute space.
  attributeSelectionSearchMethods                               Attribute selection                             Four search methods for attribute selection: ExhaustiveSearch, GeneticSearch, RandomSearch and RankSearch.
  PSOSearch                                                             Attribute selection                             An implementation of the Particle Swarm Optimization (PSO) algorithm to explore the space of attributes.

There are also various neural network packages, in addition to the MultilayerPerceptron classifier that comes with the WEKA distribution:

  SelfOrganizingMap                             Clustering                                                              Cluster data using the Kohonen's Self-Organizing Map algorithm.
  multiLayerPerceptrons                         Classification/regression, Preprocessing                This package currently contains classes for training multilayer perceptrons with one hidden layer for classification and regression, and autoencoders.
  multilayerPerceptronCS                        Classification                                                          An extension of the standard MultilayerPerceptron classifier in Weka that adds context-sensitive Multiple Task Learning (csMTL)
  wekaDeeplearning4j                            Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPU                 Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPULibs             Classification/Regression                                       CPU native libraries for wekaDeeplearning4j
  wekaDeeplearning4jCore                        Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPU                 Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPULibs             Classification/Regression                                       CPU native libraries for wekaDeeplearning4j

The list of unofficial packages at

  http://weka.wikispaces.com/Unofficial+packages+for+WEKA

has some more stuff:

  Java neural network package   Java (convolutional or fully-connected) neural network implementation with plugin for Weka.

Assuming you have R and the RPlugin package for WEKA installed and set up correctly, you can also access the nature-inspired classification and regression methods in R that are available via the MLR package (https://mlr-org.github.io/mlr-tutorial/release/html/integrated_learners/index.html). Finally, if Python is your friend, you can use the WekaPyScript package to implement nature-inspired algorithms that are available in Python.

The above list may not be complete. For example, I haven’t included some fuzzy methods because I’m not sure if they are “nature inspired”.

Cheers,
Eibe

> On 11/01/2017, at 3:30 AM, sujata sinha <[hidden email]> wrote:
>
> I would also like to know about the same!
>
> Thanks and regards,
> S
>
> Thanks and regards
> Sujata Sinha
>
>
> On Tue, Jan 10, 2017 at 3:27 PM, Fernando Lavin <[hidden email]> wrote:
> Hi everyone,
>
> Is it possible performing " Nature Inspired Algorithms" in Weka? If yes, then how to do that? If no, what it the best way to implement such approach?
>
> Thank you.
> Fernndo
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
>
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

sujata yumnam
Dear EIbe,
Thanks a lot from my side too!

Regards,
Sujata

Thanks and regards
Sujata Sinha


On Thu, Jan 12, 2017 at 11:08 AM, Fernando Lavin <[hidden email]> wrote:

Dear Eibe,

Thank you so much for the very helpful information.

Best,
Fernando

On 12 Jan 2017 9:36 am, "Eibe Frank" <[hidden email]> wrote:
Looking at the list of official packages at

  http://weka.sourceforge.net/packageMetaData/

there is

  EvolutionarySearch                                            Attribute selection                             An Evolutionary Algorithm (EA) to explore the space of attributes.
  GPAttributeGeneration                                         Classification, Preprocessing           Genetic Programming Attribute Generation
  MultiObjectiveEvolutionaryFuzzyClassifier             Classification                                  MultiObjectiveEvolutionaryFuzzyClassifier
  MultiObjectiveEvolutionarySearch                      Attribute selection                             An Multi-objective Evolutionary Algorithm (MOEA) to explore the attribute space.
  attributeSelectionSearchMethods                               Attribute selection                             Four search methods for attribute selection: ExhaustiveSearch, GeneticSearch, RandomSearch and RankSearch.
  PSOSearch                                                             Attribute selection                             An implementation of the Particle Swarm Optimization (PSO) algorithm to explore the space of attributes.

There are also various neural network packages, in addition to the MultilayerPerceptron classifier that comes with the WEKA distribution:

  SelfOrganizingMap                             Clustering                                                              Cluster data using the Kohonen's Self-Organizing Map algorithm.
  multiLayerPerceptrons                         Classification/regression, Preprocessing                This package currently contains classes for training multilayer perceptrons with one hidden layer for classification and regression, and autoencoders.
  multilayerPerceptronCS                        Classification                                                          An extension of the standard MultilayerPerceptron classifier in Weka that adds context-sensitive Multiple Task Learning (csMTL)
  wekaDeeplearning4j                            Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPU                 Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPULibs             Classification/Regression                                       CPU native libraries for wekaDeeplearning4j
  wekaDeeplearning4jCore                        Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPU                 Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPULibs             Classification/Regression                                       CPU native libraries for wekaDeeplearning4j

The list of unofficial packages at

  http://weka.wikispaces.com/Unofficial+packages+for+WEKA

has some more stuff:

  Java neural network package   Java (convolutional or fully-connected) neural network implementation with plugin for Weka.

Assuming you have R and the RPlugin package for WEKA installed and set up correctly, you can also access the nature-inspired classification and regression methods in R that are available via the MLR package (https://mlr-org.github.io/mlr-tutorial/release/html/integrated_learners/index.html). Finally, if Python is your friend, you can use the WekaPyScript package to implement nature-inspired algorithms that are available in Python.

The above list may not be complete. For example, I haven’t included some fuzzy methods because I’m not sure if they are “nature inspired”.

Cheers,
Eibe

> On 11/01/2017, at 3:30 AM, sujata sinha <[hidden email]> wrote:
>
> I would also like to know about the same!
>
> Thanks and regards,
> S
>
> Thanks and regards
> Sujata Sinha
>
>
> On Tue, Jan 10, 2017 at 3:27 PM, Fernando Lavin <[hidden email]> wrote:
> Hi everyone,
>
> Is it possible performing " Nature Inspired Algorithms" in Weka? If yes, then how to do that? If no, what it the best way to implement such approach?
>
> Thank you.
> Fernndo
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
>
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html



_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

Michael Hall
On 12 Jan 2017 9:36 am, "Eibe Frank" <[hidden email]> wrote:
Looking at the list of official packages at

  http://weka.sourceforge.net/packageMetaData/

Using the search command in my HalfPipe application I think you could find most of these with…

weka.pkgmgr.search "genetic | evolution | neural | natural | Swarm”

The argument is a regular expression. Which I’m not a regular, pun sort of intended, user of.

For a sort of shameless plug of HalfPipe...
I am working on an updated version of the HalfPipe, java shell, application to include scheduling (Quartz) functionality.

Where I see this might be useful for Weka would possibly be periodic updates of data. 
This version will include the Weka 3.8 jar and will be compiled java 8.

The scheduling also should enhance the application for periodic monitoring functionality or periodic running of scripts.
I have also recently added groovy as one way to script, although I haven’t done that much with it.

The current full set of weka. commands is…

weka.cmds
alias.weka.pkgmgr=us.hall.weka.WekaPkgmgrInvoke
alias.weka.run=weka.Run
alias.weka.pkgmgr.check=us.hall.weka.WekaPackageCheck
alias.weka.cmds=set alias.weka.*
alias.weka.version=us.hall.weka.WekaVersion
alias.weka.log=cat ${user.home}/wekafiles/weka.log
alias.weka.pkgmgr.search=us.hall.weka.WekaPackageInfoGrep

The package search I have found useful from time to time. The weka.log command can also be a quick and easy way to see that. 
Updating the package version check command is also on the to-do list to have a nicer gui implementation.
The only thing that might do more easily than Weka is update all out-dated packages at once. I’m not sure Weka has that functionality.

Michael Hall



_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

Keith Roy
In reply to this post by Eibe Frank-2
Dear Eibe, 

1- In your opinion, how powerful are "Nature Inspired Algorithms" comparing to the algorithms that are availabe already in Weka?

2- Do you have any advice about applying Nature Inspired Algorithms?

3- What are the situations/cases you believe that Nature Inspired Algorithms can produce useful results?

Thanks in advance.
Keith

On Thu, Jan 12, 2017 at 9:36 AM, Eibe Frank <[hidden email]> wrote:
Looking at the list of official packages at

  http://weka.sourceforge.net/packageMetaData/

there is

  EvolutionarySearch                                            Attribute selection                             An Evolutionary Algorithm (EA) to explore the space of attributes.
  GPAttributeGeneration                                         Classification, Preprocessing           Genetic Programming Attribute Generation
  MultiObjectiveEvolutionaryFuzzyClassifier             Classification                                  MultiObjectiveEvolutionaryFuzzyClassifier
  MultiObjectiveEvolutionarySearch                      Attribute selection                             An Multi-objective Evolutionary Algorithm (MOEA) to explore the attribute space.
  attributeSelectionSearchMethods                               Attribute selection                             Four search methods for attribute selection: ExhaustiveSearch, GeneticSearch, RandomSearch and RankSearch.
  PSOSearch                                                             Attribute selection                             An implementation of the Particle Swarm Optimization (PSO) algorithm to explore the space of attributes.

There are also various neural network packages, in addition to the MultilayerPerceptron classifier that comes with the WEKA distribution:

  SelfOrganizingMap                             Clustering                                                              Cluster data using the Kohonen's Self-Organizing Map algorithm.
  multiLayerPerceptrons                         Classification/regression, Preprocessing                This package currently contains classes for training multilayer perceptrons with one hidden layer for classification and regression, and autoencoders.
  multilayerPerceptronCS                        Classification                                                          An extension of the standard MultilayerPerceptron classifier in Weka that adds context-sensitive Multiple Task Learning (csMTL)
  wekaDeeplearning4j                            Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPU                 Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPULibs             Classification/Regression                                       CPU native libraries for wekaDeeplearning4j
  wekaDeeplearning4jCore                        Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPU                 Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPULibs             Classification/Regression                                       CPU native libraries for wekaDeeplearning4j

The list of unofficial packages at

  http://weka.wikispaces.com/Unofficial+packages+for+WEKA

has some more stuff:

  Java neural network package   Java (convolutional or fully-connected) neural network implementation with plugin for Weka.

Assuming you have R and the RPlugin package for WEKA installed and set up correctly, you can also access the nature-inspired classification and regression methods in R that are available via the MLR package (https://mlr-org.github.io/mlr-tutorial/release/html/integrated_learners/index.html). Finally, if Python is your friend, you can use the WekaPyScript package to implement nature-inspired algorithms that are available in Python.

The above list may not be complete. For example, I haven’t included some fuzzy methods because I’m not sure if they are “nature inspired”.

Cheers,
Eibe

> On 11/01/2017, at 3:30 AM, sujata sinha <[hidden email]> wrote:
>
> I would also like to know about the same!
>
> Thanks and regards,
> S
>
> Thanks and regards
> Sujata Sinha
>
>
> On Tue, Jan 10, 2017 at 3:27 PM, Fernando Lavin <[hidden email]> wrote:
> Hi everyone,
>
> Is it possible performing " Nature Inspired Algorithms" in Weka? If yes, then how to do that? If no, what it the best way to implement such approach?
>
> Thank you.
> Fernndo
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
>
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html


_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

George Dombi

Hi Gang,

The concept of nature inspired algorithms makes me think of plants and their life cycles.

On a non-mathematical topic, you might enjoy the book Thinking Like a Plant, by Craig Holdrege

https://www.amazon.com/Thinking-Like-Plant-Living-Science/dp/1584201436

Bye for now,

George


On 01/12/2017 06:17 AM, Keith Roy wrote:
Dear Eibe, 

1- In your opinion, how powerful are "Nature Inspired Algorithms" comparing to the algorithms that are availabe already in Weka?

2- Do you have any advice about applying Nature Inspired Algorithms?

3- What are the situations/cases you believe that Nature Inspired Algorithms can produce useful results?

Thanks in advance.
Keith

On Thu, Jan 12, 2017 at 9:36 AM, Eibe Frank <[hidden email]> wrote:
Looking at the list of official packages at

  http://weka.sourceforge.net/packageMetaData/

there is

  EvolutionarySearch                                            Attribute selection                             An Evolutionary Algorithm (EA) to explore the space of attributes.
  GPAttributeGeneration                                         Classification, Preprocessing           Genetic Programming Attribute Generation
  MultiObjectiveEvolutionaryFuzzyClassifier             Classification                                  MultiObjectiveEvolutionaryFuzzyClassifier
  MultiObjectiveEvolutionarySearch                      Attribute selection                             An Multi-objective Evolutionary Algorithm (MOEA) to explore the attribute space.
  attributeSelectionSearchMethods                               Attribute selection                             Four search methods for attribute selection: ExhaustiveSearch, GeneticSearch, RandomSearch and RankSearch.
  PSOSearch                                                             Attribute selection                             An implementation of the Particle Swarm Optimization (PSO) algorithm to explore the space of attributes.

There are also various neural network packages, in addition to the MultilayerPerceptron classifier that comes with the WEKA distribution:

  SelfOrganizingMap                             Clustering                                                              Cluster data using the Kohonen's Self-Organizing Map algorithm.
  multiLayerPerceptrons                         Classification/regression, Preprocessing                This package currently contains classes for training multilayer perceptrons with one hidden layer for classification and regression, and autoencoders.
  multilayerPerceptronCS                        Classification                                                          An extension of the standard MultilayerPerceptron classifier in Weka that adds context-sensitive Multiple Task Learning (csMTL)
  wekaDeeplearning4j                            Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPU                 Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jCPULibs             Classification/Regression                                       CPU native libraries for wekaDeeplearning4j
  wekaDeeplearning4jCore                        Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPU                 Classification/Regression                                       Weka wrappers for Deeplearning4j
  wekaDeeplearning4jGPULibs             Classification/Regression                                       CPU native libraries for wekaDeeplearning4j

The list of unofficial packages at

  http://weka.wikispaces.com/Unofficial+packages+for+WEKA

has some more stuff:

  Java neural network package   Java (convolutional or fully-connected) neural network implementation with plugin for Weka.

Assuming you have R and the RPlugin package for WEKA installed and set up correctly, you can also access the nature-inspired classification and regression methods in R that are available via the MLR package (https://mlr-org.github.io/mlr-tutorial/release/html/integrated_learners/index.html). Finally, if Python is your friend, you can use the WekaPyScript package to implement nature-inspired algorithms that are available in Python.

The above list may not be complete. For example, I haven’t included some fuzzy methods because I’m not sure if they are “nature inspired”.

Cheers,
Eibe

> On 11/01/2017, at 3:30 AM, sujata sinha <[hidden email]> wrote:
>
> I would also like to know about the same!
>
> Thanks and regards,
> S
>
> Thanks and regards
> Sujata Sinha
>
>
> On Tue, Jan 10, 2017 at 3:27 PM, Fernando Lavin <[hidden email]> wrote:
> Hi everyone,
>
> Is it possible performing " Nature Inspired Algorithms" in Weka? If yes, then how to do that? If no, what it the best way to implement such approach?
>
> Thank you.
> Fernndo
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
>
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html



_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html


_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

Eibe Frank-3
In reply to this post by Keith Roy
On Fri, Jan 13, 2017 at 12:17 AM, Keith Roy <[hidden email]> wrote:
 
1- In your opinion, how powerful are "Nature Inspired Algorithms" comparing to the algorithms that are availabe already in Weka?

Deep neural networks are clearly very powerful in some domains. There is some support for them in WEKA now through the wekaDeeplearning4j packages.
 
2- Do you have any advice about applying Nature Inspired Algorithms?

 No, I don't have much experience with them.

3- What are the situations/cases you believe that Nature Inspired Algorithms can produce useful results?

You mean better results than other techniques? Deep neural networks have been shown to perform significantly better than other existing methods on problems such as image classification but I'm not so sure about nature-inspired optimization algorithms applied to machine learning (e.g., evolutionary methods, etc.).

Cheers,
Eibe

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Nature Inspired Algorithms

Keith Roy

Many thanks Eibe for your help. That's what I was looking for.

Cheers,
Keith

On 13 Jan 2017 6:16 am, "Eibe Frank" <[hidden email]> wrote:
On Fri, Jan 13, 2017 at 12:17 AM, Keith Roy <[hidden email]> wrote:
 
1- In your opinion, how powerful are "Nature Inspired Algorithms" comparing to the algorithms that are availabe already in Weka?

Deep neural networks are clearly very powerful in some domains. There is some support for them in WEKA now through the wekaDeeplearning4j packages.
 
2- Do you have any advice about applying Nature Inspired Algorithms?

 No, I don't have much experience with them.

3- What are the situations/cases you believe that Nature Inspired Algorithms can produce useful results?

You mean better results than other techniques? Deep neural networks have been shown to perform significantly better than other existing methods on problems such as image classification but I'm not so sure about nature-inspired optimization algorithms applied to machine learning (e.g., evolutionary methods, etc.).

Cheers,
Eibe

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html


_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Loading...