java-ml-support — Java ML support mailinglist

[Java-ML-support] Java-ml Capability Question

[Oscar B. <oba...@um...>]

Hello,

Would someone please tell me if Java-ml is able to carry out a
multi-variate regression exercuse by receiving a dataset/file containing my
one dependent variable and three independent variables and subsequently
tell me what the mathematical relationship/equation is that describes how
to obtain the dependent variable from the three independent variables?

I looked through the list (http://java-ml.sourceforge.net/api/0.1.7/) that
has all the component packages/classes for Java-ml, but no descriptor of
any of the items listed appeared to describe the multivariate regression I
am trying to do.  Thanks.

Oscar

Oscar Bastidas, Ph.D.
Postdoctoral Research Associate
University of Minnesota

[Java-ML-support] About particle swarm optimization

[javed k. <jav...@gm...>]

Hello

How can I select some features using particle swarm optimization in java
code? I used it in WEKA GUI but it did not improve the prediction's
accuracy.

Regards

[Java-ML-support] About kmeans

[Carlos R. <rom...@gm...>]

Hi. I'm currently using javaml, specifically the kmeans algorithm. I need
to cluster several pieces of information. I'm using the constructor passing
the number of clusters and the number of iterations as parameters so I can
alter the parameters (mainly the iterations) and watch the changes that may
happen. I have a doubt concerning the end of the algorithm, does the kmeans
in this package always reach the maximum number of iterations or does it
stop when it reaches certain stability within the clusters? For example if
I set the iterations to 200 but around the 100 there are no changes it
keeps going, or it stops when it detects that it's not changin?

Thanks in advance

Re: [Java-ML-support] pull requests

[Thomas A. <th...@ab...>]

Hi Luis,

Development is pretty sleepy, but not quite dead yet. And there are some 
development on the horizon.

The right place for pull requests would be here:
https://github.com/AbeelLab/javaml

best,
Thomas


On 3/3/2016 5:29 PM, Luis Veci wrote:
> The JavaML source code in sourceforge hasn't been updated in a long time.
> Is there any active development on github or somewhere else?
>
> If I would like to submit a bug fix how should this be done?
> Thanks
>

[Java-ML-support] pull requests

[Luis V. <lv...@ar...>]

The JavaML source code in sourceforge hasn't been updated in a long time.
Is there any active development on github or somewhere else?

If I would like to submit a bug fix how should this be done?
Thanks

-- 
Luis Veci
Software Engineering Lead
Array System Computing Inc
www.array.ca
Toronto Canada
(416) 736-0900 ext 298

[Java-ML-support] Feature Scoring for Regression Datasets

[Sumit S. <sum...@gm...>]

Hi

Is there a function to do feature scoring for regression datasets, where
the class label is a real number rather than the conventional binary
classification scenario?

I tried on a synthetic random dataset but all score values are zero.

Pl help, here is what I tried.

public static void main(String arr[]){
Dataset dataset = new DefaultDataset();

for (int i=1; i<=10; i++){
Instance tmpInstance = InstanceTools.randomInstance(25);
tmpInstance.setClassValue(5);
System.out.println(tmpInstance.value(0)+","+tmpInstance.toString());
dataset.add(tmpInstance);
}

System.out.println("Samples="+dataset.size());
GainRatio ga = new GainRatio();
ga.build(dataset);
System.out.println("Features="+ga.noAttributes());
for (int i = 0; i < ga.noAttributes(); i++)
System.out.println(ga.score(i));
}


Sumit

Re: [Java-ML-support] Java-ml data types

[Marcin <ma...@ga...>]

Thank you for answer.
I did that (encoding), but I would like algorithms from WEKA to 
distinguish numerical from nominal attributes, cause it becomes highly 
important there, do you know how to do it?

btw. I just get null pointer from
"Object predictedClassValue = knn.classify(inst);"
any idea how to check what went wrongin this case?

Best regards
Marcin Witkowski

W dniu 2015-09-04 o 13:23, Thomas Abeel pisze:
> There are currently only numerical attributes. You can easily encode
> nominal attributes either numerically or binary.
>
> best,
> Thomas
>
> On 9/4/2015 11:44 AM, Marcin wrote:
>> Hi all,
>>
>> I am usinf Java-ml for last two days, and I come across the problem of
>> data types. I did not find information whether Java-ml treats attributes
>> as nominal or numerical and how to force him to treat them as nominal.
>>
>> Thanks for all the help.
>>
>> Best Regards
>> Marcin Witkowski
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Java-ml-support mailing list
>> Jav...@li...
>> https://lists.sourceforge.net/lists/listinfo/java-ml-support
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support
>

Re: [Java-ML-support] Java-ml data types

[Thomas A. <th...@ab...>]

There are currently only numerical attributes. You can easily encode 
nominal attributes either numerically or binary.

best,
Thomas

On 9/4/2015 11:44 AM, Marcin wrote:
> Hi all,
>
> I am usinf Java-ml for last two days, and I come across the problem of
> data types. I did not find information whether Java-ml treats attributes
> as nominal or numerical and how to force him to treat them as nominal.
>
> Thanks for all the help.
>
> Best Regards
> Marcin Witkowski
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support

[Java-ML-support] Java-ml data types

[Marcin <ma...@ga...>]

Hi all,

I am usinf Java-ml for last two days, and I come across the problem of 
data types. I did not find information whether Java-ml treats attributes 
as nominal or numerical and how to force him to treat them as nominal.

Thanks for all the help.

Best Regards
Marcin Witkowski

Re: [Java-ML-support] On LibSVM wrapper

[Thomas A. <th...@ab...>]

What's the goal of accessing the model? If it's for reuse, you can 
serialize the entire LibSVM object. If it's for something else, the 
answer is no, the model is a private variable within the wrapper and 
right now there is no accessor method.

I'm happy to include it if you send a merge request on github.

cheers,
Thomas
On 7/16/2014 4:12 AM, PhD - Wang Jingyi wrote:
> The last email was blocked, so I send again. :)
> On 16 Jul, 2014, at 3:38 pm, Wang Jingyi-sutd 
> <jin...@my... 
> <mailto:jin...@my...>> wrote:
>
>> Hi:
>>
>> I'm using your LibSVM wrapper class.
>>
>> After building the classifier, is there a way for me to access the 
>> result model obtained?
>>
>> *Regards,*
>> *Jingyi*
>
> The above message may contain confidential and/or proprietary 
> information that is exempt from disclosure under applicable law and is 
> intended for receipt and use solely by the addressee(s) named above. 
> If you are not the intended recipient, you are hereby notified that 
> any use, dissemination, distribution, or copying of this message, or 
> any attachment, is strictly prohibited. If you have received this 
> email in error, please inform the sender immediately by reply e-mail 
> or telephone, reversing the charge if necessary. Please delete the 
> message thereafter. Thank you.
>
>
>
> ------------------------------------------------------------------------------
> Want fast and easy access to all the code in your enterprise? Index and
> search up to 200,000 lines of code with a free copy of Black Duck
> Code Sight - the same software that powers the world's largest code
> search on Ohloh, the Black Duck Open Hub! Try it now.
> http://p.sf.net/sfu/bds
>
>
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support

Re: [Java-ML-support] On LibSVM wrapper

[PhD - W. J. <jin...@my...>]

The last email was blocked, so I send again. :)
On 16 Jul, 2014, at 3:38 pm, Wang Jingyi－sutd <jin...@my...<mailto:jin...@my...>> wrote:

​Hi:

I'm using your LibSVM wrapper class.

After building the classifier, is there a way for me to access the result model obtained?

Regards,
Jingyi


The above message may contain confidential and/or proprietary information that is exempt from disclosure under applicable law and is intended for receipt and use solely by the addressee(s) named above. If you are not the intended recipient, you are hereby notified that any use, dissemination, distribution, or copying of this message, or any attachment, is strictly prohibited. If you have received this email in error, please inform the sender immediately by reply e-mail or telephone, reversing the charge if necessary. Please delete the message thereafter. Thank you.

[Java-ML-support] How to use a custom distance measure for K-medoids clustering ??

[Khirod K. N. <khi...@gm...>]

I have to use the K-medoids clustering for my project and I am having
trouble using a simple implementation. So far I have done this

import net.sf.javaml.core.*;
import net.sf.javaml.clustering.*;
import net.sf.javaml.distance.*;

public class ClusteringTest {
    public static void main(String args[]) {
        Dataset data = new DefaultDataset();

        double[] val = {1, 2, 7};
        double[] gval = {1, 2, 3};

        CustomDist de = new CustomDist();

        for (int i = 0; i < val.length; i++) {
            Instance instance = new SparseInstance(2);
            instance.put(1, val[i]);
            instance.put(2, gval[i]);

            data.add(instance);
        }


        // Test Clustering
        Clusterer km = new KMedoids(3, 10, de);
        Dataset[] clusters = km.cluster(data);

        System.out.println(clusters[0]);
        System.out.println(clusters[1]);
    }
}

class CustomDist implements DistanceMeasure {
    public boolean compare(double x, double y) {
        return x < y;
    }

    public double getMaxValue() {
        return Math.sqrt(40);
    }

    public double getMinValue() {
        return Math.sqrt(2);
    }

    public double measure(Instance x, Instance y) {
        // Use Euclidean Distance (for test only)
        double x1 = Math.abs(x.value(1) - y.value(1)) * Math.abs(x.value(1)
- y.value(1));
        double y1 = Math.abs(x.value(2) - y.value(2)) * Math.abs(x.value(2)
- y.value(2));

        return Math.sqrt(x1 + y1);
    }
}

Is there something wrong with my ClusteringTest Class??

Re: [Java-ML-support] Question about AIC and BIC scores

[Thomas A. <th...@ab...>]

Okay.

If it's a case of "it depends", I'll make it a parameter that users have 
to specify when constructing the metric.

I've committed a patch to that effect to the repository.

cheers,
Thomas
On 11/22/2013 11:45 AM, Sergio Pulido wrote:
> Hi Tomas,
>
> I found this:
> http://books.nips.cc/papers/files/nips16/NIPS2003_AA36.pdf
> http://www.autonlab.org/autonweb/14661/version/3/part/5/data/pelleg-xmeans.pdf?branch=main&language=en
>
> They say that actually k (the number of free parameters) is bigger 
> than the number of clusters - 1 as was my initial thought. It also 
> depends on dimensionality, k should be:
>
> The papers do not agree in what should be the numbers of free 
> parameters (I just looked at them very fast so I am probably wrong or 
> not seeing something obvious)
>
> The first papers says: "p = k(d + 1) is the number of parameters in 
> the model"
> double k = clusters.lenght * (clusters[0].get(0).noAttributes() + 1);
>
> the second says: "the number of free parameters p is simply the sum of 
> k - 1 class probabilities, M.K centroid coordinates, and one variance 
> estimate"
> double k = (clusters.lenght - 1) + 
> clusters.lenght*(clusters[0].get(0).noAttributes() + 1);
>
> I think the second is the best:
> 1. It put additional penalties for the number of clusters but not 
> greatly (just an aditional clusters.lenght - 1). It count as a 
> parameter the existence of each cluster (?)
> 2. It have the -1... in the other one I do not see the bound parameter 
> anywhere.
>
> The problem I see with this is that that is the calculation of free 
> parameters for a KMeans clustering, but not necessary for other types 
> of clustering algorithms with different assumptions, K-means assumes a 
> Gaussian circular distribution.
>
> Anyhow, it should be better than just k = 1 :)
>
> Cheers,
> Sergio.
>
>
>
>
>
>
> On Fri, Nov 22, 2013 at 3:18 PM, Thomas Abeel <th...@ab... 
> <mailto:th...@ab...>> wrote:
>
>     Hi Sergio,
>
>     This has been a while. This may be a bug. Can you point me to any
>     literature reference that says the number of free parameters =
>     clusters.length-1?
>
>     cheers,
>     Thomas
>
>     On 11/12/2013 9:08 AM, Sergio Pulido wrote:
>>     Hello all,
>>
>>     I see that the implementation of AICScore and BICScore uses a k=1
>>     as the number of free parameters
>>
>>     // number of free parameters K
>>     double k = 1;
>>
>>     Can you explain me why the number of free parameters is not the
>>     number of clusters -1?
>>
>>     // number of free parameters K
>>     double k = clusters.length - 1;
>>
>>     Thank you a lot,
>>     Sergio.
>>
>>
>>     ------------------------------------------------------------------------------
>>     Shape the Mobile Experience: Free Subscription
>>     Software experts and developers: Be at the forefront of tech innovation.
>>     Intel(R) Software Adrenaline delivers strategic insight and game-changing
>>     conversations that shape the rapidly evolving mobile landscape. Sign up now.
>>     http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk
>>
>>
>>     _______________________________________________
>>     Java-ml-support mailing list
>>     Jav...@li...  <mailto:Jav...@li...>
>>     https://lists.sourceforge.net/lists/listinfo/java-ml-support
>
>
>     ------------------------------------------------------------------------------
>     Shape the Mobile Experience: Free Subscription
>     Software experts and developers: Be at the forefront of tech
>     innovation.
>     Intel(R) Software Adrenaline delivers strategic insight and
>     game-changing
>     conversations that shape the rapidly evolving mobile landscape.
>     Sign up now.
>     http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk
>     _______________________________________________
>     Java-ml-support mailing list
>     Jav...@li...
>     <mailto:Jav...@li...>
>     https://lists.sourceforge.net/lists/listinfo/java-ml-support
>
>
>
>
> ------------------------------------------------------------------------------
> Shape the Mobile Experience: Free Subscription
> Software experts and developers: Be at the forefront of tech innovation.
> Intel(R) Software Adrenaline delivers strategic insight and game-changing
> conversations that shape the rapidly evolving mobile landscape. Sign up now.
> http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk
>
>
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support

Re: [Java-ML-support] Question about AIC and BIC scores

[Sergio P. <spu...@gm...>]

Hi Tomas,

I found this:
http://books.nips.cc/papers/files/nips16/NIPS2003_AA36.pdf
http://www.autonlab.org/autonweb/14661/version/3/part/5/data/pelleg-xmeans.pdf?branch=main&language=en

They say that actually k (the number of free parameters) is bigger than the
number of clusters - 1 as was my initial thought. It also depends on
dimensionality, k should be:

The papers do not agree in what should be the numbers of free parameters (I
just looked at them very fast so I am probably wrong or not seeing
something obvious)

The first papers says: "p = k(d + 1) is the number of parameters in the
model"
double k = clusters.lenght * (clusters[0].get(0).noAttributes() + 1);

the second says: "the number of free parameters p is simply the sum of k -
1 class probabilities, M.K centroid coordinates, and one variance estimate"
double k = (clusters.lenght - 1) +
clusters.lenght*(clusters[0].get(0).noAttributes() + 1);

I think the second is the best:
1. It put additional penalties for the number of clusters but not greatly
(just an aditional clusters.lenght - 1). It count as a parameter the
existence of each cluster (?)
2. It have the -1... in the other one I do not see the bound parameter
anywhere.

The problem I see with this is that that is the calculation of free
parameters for a KMeans clustering, but not necessary for other types of
clustering algorithms with different assumptions, K-means assumes a
Gaussian circular distribution.

Anyhow, it should be better than just k = 1 :)

Cheers,
Sergio.






On Fri, Nov 22, 2013 at 3:18 PM, Thomas Abeel <th...@ab...> wrote:

>  Hi Sergio,
>
> This has been a while. This may be a bug. Can you point me to any
> literature reference that says the number of free parameters =
> clusters.length-1?
>
> cheers,
> Thomas
>
> On 11/12/2013 9:08 AM, Sergio Pulido wrote:
>
> Hello all,
>
>  I see that the implementation of AICScore and BICScore uses a k=1 as the
> number of free parameters
>
>  // number of free parameters K
> double k = 1;
>
>  Can you explain me why the number of free parameters is not the number
> of clusters -1?
>
>  // number of free parameters K
> double k = clusters.length - 1;
>
>  Thank you a lot,
> Sergio.
>
>
> ------------------------------------------------------------------------------
> Shape the Mobile Experience: Free Subscription
> Software experts and developers: Be at the forefront of tech innovation.
> Intel(R) Software Adrenaline delivers strategic insight and game-changing
> conversations that shape the rapidly evolving mobile landscape. Sign up now. http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk
>
>
>
> _______________________________________________
> Java-ml-support mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/java-ml-support
>
>
>
>
> ------------------------------------------------------------------------------
> Shape the Mobile Experience: Free Subscription
> Software experts and developers: Be at the forefront of tech innovation.
> Intel(R) Software Adrenaline delivers strategic insight and game-changing
> conversations that shape the rapidly evolving mobile landscape. Sign up
> now.
> http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support
>
>

Re: [Java-ML-support] Question about AIC and BIC scores

[Thomas A. <th...@ab...>]

Hi Sergio,

This has been a while. This may be a bug. Can you point me to any 
literature reference that says the number of free parameters = 
clusters.length-1?

cheers,
Thomas
On 11/12/2013 9:08 AM, Sergio Pulido wrote:
> Hello all,
>
> I see that the implementation of AICScore and BICScore uses a k=1 as 
> the number of free parameters
>
> // number of free parameters K
> double k = 1;
>
> Can you explain me why the number of free parameters is not the number 
> of clusters -1?
>
> // number of free parameters K
> double k = clusters.length - 1;
>
> Thank you a lot,
> Sergio.
>
>
> ------------------------------------------------------------------------------
> Shape the Mobile Experience: Free Subscription
> Software experts and developers: Be at the forefront of tech innovation.
> Intel(R) Software Adrenaline delivers strategic insight and game-changing
> conversations that shape the rapidly evolving mobile landscape. Sign up now.
> http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk
>
>
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support

[Java-ML-support] MCL

[Rafael J. <ra...@eb...>]

Hello,

I am looking for some help. I am using MCL to group protein interactions.
MCL helps me to get a better graphical representation of interactions. Here
some examples ...

http://www.heartproteome.org/copa/interactome_mouse_proteasome.aspx
http://www.heartproteome.org/copa/interactome_human_mitochondria.aspx

So far I am using the MCL command line tool in the simplest way ...

mcl input.txt --abc -o output.txt

This is an input example ...

Prot.A Prot.B Score
P37173 P07200 0.5248886
Q9Y6C2 P07200 0.40116468
Q15750 Q15750 0.40116468
...

All my code is in Java and I would like to be more independent of the
command line, is there a way to do the same thing with Java-ML?

Thanks in advance for your help,
Regards,
Rafael.

[Java-ML-support] Question about AIC and BIC scores

[Sergio P. <spu...@gm...>]

Hello all,

I see that the implementation of AICScore and BICScore uses a k=1 as the
number of free parameters

// number of free parameters K
double k = 1;

Can you explain me why the number of free parameters is not the number of
clusters -1?

// number of free parameters K
double k = clusters.length - 1;

Thank you a lot,
Sergio.

[Java-ML-support] \\ Can I still get the source code for research purpose?

[Qin Z. M. Z. <q.z...@uq...>]

                           Hi ,  I am trying to use the javaml which was developed by your team, however, there are something I can figure out so I want study the source code to understand . when I install the plugin and import SVN, it always cannot success without any error information.

                           Could you please give me some guidance ?

                         Thank you !!!

Re: [Java-ML-support] How to save a model file for later use

[Thomas A. <ta...@br...>]

You can simply serialize the classifier object after buildClassifier and 
it will retain it's state. Just deserialize it back when you need it.

cheers,
Thomas

On 8/10/2013 1:41 AM, Ritesh wrote:
> H
> ​i,
>
> I was ​wondering if there is some way to save something like a model
> file on the disk after we build a classifier from a dataset. The API
> documentation lists the methods "buildClassifier" which basically trains
> and builds any classifier and "classify" which classifies an instance
> using this classifier. So effectively it means I have to build a
> classifier every time I want to classify some instances which takes up
> lot of resource. Is there a way to save the classifier after
> "buildClassifier" which could be loaded into the memory any time we want
> to classify some instance?
>
> Regards,
>
> Ritesh
>
>
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite!
> It's a free troubleshooting tool designed for production.
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
>
>
>
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support
>

[Java-ML-support] How to save a model file for later use

[Ritesh <rit...@gm...>]

H
​i,

I was ​wondering if there is some way to save something like a model file
on the disk after we build a classifier from a dataset. The API
documentation lists the methods "buildClassifier" which basically trains
and builds any classifier and "classify" which classifies an instance using
this classifier. So effectively it means I have to build a classifier every
time I want to classify some instances which takes up lot of resource. Is
there a way to save the classifier after "buildClassifier" which could be
loaded into the memory any time we want to classify some instance?

Regards,

Ritesh

Re: [Java-ML-support] instance key, how to set?

[Thomas A. <th...@ab...>]

Have you tried the documentation on java-ml.sf.net?

At the moment there is no way to use keys with instances, everything is 
index based.

If you think the javadocs need to be expanded, we would be very grateful 
if you can expand them and submit the patches and I'll be happy to apply 
them to the code-base.


cheers,
Thomas

On 8/3/2013 3:51 AM, טל דניאל wrote:
> It isn't clear from the documentation how to set a key for an instance
> (row) in a dataset.
> Sometimes, one has to find a sparseInstance, by key, in order to add
> values to it, but it isn't clear how to accomplish this.
>
> Also, the word class in the documentation is confusing, and I think
> should be changed to classification.
>
> I tried to figure out how to use filters but didn't get it from the
> javadocs.
>
> Any tutotial out there? can someone help me understand if Java-ML
> instances can be used with key, and then values of instance? the getID
> seems to through a random number.
>
> As a last resort, I can create a HashMap that maps each userID, for an
> instance row number in the dataset, but I want to know if JAva-ML
> already supports that simple action.
>
> Thanks,
> Tal
>
>
>
> ------------------------------------------------------------------------------
> Get your SQL database under version control now!
> Version control is standard for application code, but databases havent
> caught up. So what steps can you take to put your SQL databases under
> version control? Why should you start doing it? Read more to find out.
> http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
>
>
>
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support
>

[Java-ML-support] instance key, how to set?

[טל ד. <tal...@gm...>]

It isn't clear from the documentation how to set a key for an instance
(row) in a dataset.
Sometimes, one has to find a sparseInstance, by key, in order to add values
to it, but it isn't clear how to accomplish this.

Also, the word class in the documentation is confusing, and I think should
be changed to classification.

I tried to figure out how to use filters but didn't get it from the
javadocs.

Any tutotial out there? can someone help me understand if Java-ML instances
can be used with key, and then values of instance? the getID seems to
through a random number.

As a last resort, I can create a HashMap that maps each userID, for an
instance row number in the dataset, but I want to know if JAva-ML already
supports that simple action.

Thanks,
Tal

[Java-ML-support] transforming dataset into a different dataset structure

[טל ד. <tal...@gm...>]

Hi, I'd like to transform a raw dataset, loaded from file (e.g. user item
rating),
into a new dataset with a different structure (e.g. user item1rating
item2rating ...)

how can I accomplish this? I tried using indexOf, and use the first field
value as an identifier, on the raw dataset, but it doesn't work.

Any idea how to batch ratings with the same userID in the raw dataset, into
a single instance in the new dataset?

Thanks,
Tal

Re: [Java-ML-support] Weka Bridge

[Thomas A. <ta...@br...>]

You can specific most options with either setters or during construction 
of the weka clusterer.

cheers,
Thomas

On 7/23/2013 4:57 PM, Luis Fdo Guzmán Nateras wrote:
> Is there any way to specify the options for the weka clusterers through
> the WekaClusterer Class in JML?
>
> For example the XMeans Weka class has all these options:
>
> Valid options are:
>
>   -I <num>
>    maximum number of overall iterations
>    (default 1).
>
>   -M <num>
>    maximum number of iterations in the kMeans loop in
>    the Improve-Parameter part
>    (default 1000).
>
>   -J <num>
>    maximum number of iterations in the kMeans loop
>    for the splitted centroids in the Improve-Structure part
>    (default 1000).
>
>   -L <num>
>    minimum number of clusters
>    (default 2).
>
>   -H <num>
>    maximum number of clusters
>    (default 4).
>
>   -B <value>
>    distance value for binary attributes
>    (default 1.0).
>
>   -use-kdtree
>    Uses the KDTree internally
>    (default no).
>
>   -K <KDTree class specification>
>    Full class name of KDTree class to use, followed
>    by scheme options.
>    eg: "weka.core.neighboursearch.kdtrees.KDTree -P"
>    (default no KDTree class used).
>
>   -C <value>
>    cutoff factor, takes the given percentage of the splitted
>    centroids if none of the children win
>    (default 0.0).
>
>   -D <distance function class specification>
>    Full class name of Distance function class to use, followed
>    by scheme options.
>    (default weka.core.EuclideanDistance).
>
>   -N <file name>
>    file to read starting centers from (ARFF format).
>
>   -O <file name>
>    file to write centers to (ARFF format).
>
>   -U <int>
>    The debug level.
>    (default 0)
>
>   -Y <file name>
>    The debug vectors file.
>
>   -S <num>
>    Random number seed.
>    (default 10)
>
> However in the tutorial there isn't a way to specify these options.
>
> I would really appreciate any help in this matter.
>
>
> Luis Guzman
>
>
>
> ------------------------------------------------------------------------------
> See everything from the browser to the database with AppDynamics
> Get end-to-end visibility with application monitoring from AppDynamics
> Isolate bottlenecks and diagnose root cause in seconds.
> Start your free trial of AppDynamics Pro today!
> http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
>
>
>
> _______________________________________________
> Java-ml-support mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/java-ml-support
>

[Java-ML-support] Weka Bridge

[Luis F. G. N. <mef...@ho...>]

Is there any way to specify the options for the weka clusterers through the WekaClusterer Class in JML?
For example the XMeans Weka class has all these options:
Valid options are: -I <num>
  maximum number of overall iterations
  (default 1). -M <num>
  maximum number of iterations in the kMeans loop in
  the Improve-Parameter part 
  (default 1000). -J <num>
  maximum number of iterations in the kMeans loop
  for the splitted centroids in the Improve-Structure part 
  (default 1000). -L <num>
  minimum number of clusters
  (default 2). -H <num>
  maximum number of clusters
  (default 4). -B <value>
  distance value for binary attributes
  (default 1.0). -use-kdtree
  Uses the KDTree internally
  (default no). -K <KDTree class specification>
  Full class name of KDTree class to use, followed
  by scheme options.
  eg: "weka.core.neighboursearch.kdtrees.KDTree -P"
  (default no KDTree class used). -C <value>
  cutoff factor, takes the given percentage of the splitted 
  centroids if none of the children win
  (default 0.0). -D <distance function class specification>
  Full class name of Distance function class to use, followed
  by scheme options.
  (default weka.core.EuclideanDistance). -N <file name>
  file to read starting centers from (ARFF format). -O <file name>
  file to write centers to (ARFF format). -U <int>
  The debug level.
  (default 0) -Y <file name>
  The debug vectors file. -S <num>
  Random number seed.
  (default 10)However in the tutorial there isn't a way to specify these options.I would really appreciate any help in this matter.
Luis Guzman