predicted value vs actual value in Weka

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

predicted value vs actual value in Weka

asadbtk
We get the MAE, RMSE etc values when we predict values in regression. But how can we get the individual values of output variable like we do in r language as : fitted(model)

Thanks

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
To subscribe, unsubscribe, etc., visit 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
|

Re: predicted value vs actual value in Weka

Eibe Frank-3
Which user interface are you using to work with WEKA? In the Explorer, there is an option available under "More options..." in the Classify tab that triggers output of predicted and actual values.

A similar option is available at the command-line. Use -h together with java weka.Run <name of learning scheme> to get a list of the available command-line options.

Cheers,
Eibe

On Tue, Jan 1, 2019 at 12:37 AM javed khan <[hidden email]> wrote:
We get the MAE, RMSE etc values when we predict values in regression. But how can we get the individual values of output variable like we do in r language as : fitted(model)

Thanks
_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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
|

Re: predicted value vs actual value in Weka

asadbtk
Hello Elbe, thanks a lot for your response.

I am using the Explorer interface. Can we get all the predicted and actual values via Explorer?

Best regards

On Tue, Jan 1, 2019 at 2:31 AM Eibe Frank <[hidden email]> wrote:
Which user interface are you using to work with WEKA? In the Explorer, there is an option available under "More options..." in the Classify tab that triggers output of predicted and actual values.

A similar option is available at the command-line. Use -h together with java weka.Run <name of learning scheme> to get a list of the available command-line options.

Cheers,
Eibe

On Tue, Jan 1, 2019 at 12:37 AM javed khan <[hidden email]> wrote:
We get the MAE, RMSE etc values when we predict values in regression. But how can we get the individual values of output variable like we do in r language as : fitted(model)

Thanks
_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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
|

Re: predicted value vs actual value in Weka

Eibe Frank-3
Yes, in the Classify panel under “More options...”, you can turn this on, e.g., by switching to PlainText output.

Cheers,
Eibe

On Tue, 1 Jan 2019 at 11:03 PM, javed khan <[hidden email]> wrote:
Hello Elbe, thanks a lot for your response.

I am using the Explorer interface. Can we get all the predicted and actual values via Explorer?

Best regards

On Tue, Jan 1, 2019 at 2:31 AM Eibe Frank <[hidden email]> wrote:
Which user interface are you using to work with WEKA? In the Explorer, there is an option available under "More options..." in the Classify tab that triggers output of predicted and actual values.

A similar option is available at the command-line. Use -h together with java weka.Run <name of learning scheme> to get a list of the available command-line options.

Cheers,
Eibe

On Tue, Jan 1, 2019 at 12:37 AM javed khan <[hidden email]> wrote:
We get the MAE, RMSE etc values when we predict values in regression. But how can we get the individual values of output variable like we do in r language as : fitted(model)

Thanks
_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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
|

Re: predicted value vs actual value in Weka

asadbtk
Thanks a lot Eibe for your cooperation. 

Best regards


On Tue, Jan 1, 2019 at 12:17 PM Eibe Frank <[hidden email]> wrote:
Yes, in the Classify panel under “More options...”, you can turn this on, e.g., by switching to PlainText output.

Cheers,
Eibe

On Tue, 1 Jan 2019 at 11:03 PM, javed khan <[hidden email]> wrote:
Hello Elbe, thanks a lot for your response.

I am using the Explorer interface. Can we get all the predicted and actual values via Explorer?

Best regards

On Tue, Jan 1, 2019 at 2:31 AM Eibe Frank <[hidden email]> wrote:
Which user interface are you using to work with WEKA? In the Explorer, there is an option available under "More options..." in the Classify tab that triggers output of predicted and actual values.

A similar option is available at the command-line. Use -h together with java weka.Run <name of learning scheme> to get a list of the available command-line options.

Cheers,
Eibe

On Tue, Jan 1, 2019 at 12:37 AM javed khan <[hidden email]> wrote:
We get the MAE, RMSE etc values when we predict values in regression. But how can we get the individual values of output variable like we do in r language as : fitted(model)

Thanks
_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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
|

Re: predicted value vs actual value in Weka

asadbtk
Hello sir, 

I have a question regarding the meta heuristic feature selection algorithms, if they are written by third party or belongs to the Waikato University. Second, how can we implement the hybrid approach via weka explorer for features selections. Is it like we select the features first using one algorithm, say genetic algorithm and then execute the selected features again with another algorithm, like ant search? 

Best regards 

On Tuesday, January 1, 2019, javed khan <[hidden email]> wrote:
Thanks a lot Eibe for your cooperation. 

Best regards


On Tue, Jan 1, 2019 at 12:17 PM Eibe Frank <[hidden email]> wrote:
Yes, in the Classify panel under “More options...”, you can turn this on, e.g., by switching to PlainText output.

Cheers,
Eibe

On Tue, 1 Jan 2019 at 11:03 PM, javed khan <[hidden email]> wrote:
Hello Elbe, thanks a lot for your response.

I am using the Explorer interface. Can we get all the predicted and actual values via Explorer?

Best regards

On Tue, Jan 1, 2019 at 2:31 AM Eibe Frank <[hidden email]> wrote:
Which user interface are you using to work with WEKA? In the Explorer, there is an option available under "More options..." in the Classify tab that triggers output of predicted and actual values.

A similar option is available at the command-line. Use -h together with java weka.Run <name of learning scheme> to get a list of the available command-line options.

Cheers,
Eibe

On Tue, Jan 1, 2019 at 12:37 AM javed khan <[hidden email]> wrote:
We get the MAE, RMSE etc values when we predict values in regression. But how can we get the individual values of output variable like we do in r language as : fitted(model)

Thanks
_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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
|

Re: predicted value vs actual value in Weka

Eibe Frank-3
Sure. For example, you can nest one AttributeSelectedClassifier into another one to achieve this. Or you can use the MultiFilter to combine two AttributeSelection filters.

Cheers,
Eibe

On Sun, 6 Jan 2019 at 12:11 AM, javed khan <[hidden email]> wrote:
Hello sir, 

I have a question regarding the meta heuristic feature selection algorithms, if they are written by third party or belongs to the Waikato University. Second, how can we implement the hybrid approach via weka explorer for features selections. Is it like we select the features first using one algorithm, say genetic algorithm and then execute the selected features again with another algorithm, like ant search? 

Best regards 

On Tuesday, January 1, 2019, javed khan <[hidden email]> wrote:
Thanks a lot Eibe for your cooperation. 

Best regards


On Tue, Jan 1, 2019 at 12:17 PM Eibe Frank <[hidden email]> wrote:
Yes, in the Classify panel under “More options...”, you can turn this on, e.g., by switching to PlainText output.

Cheers,
Eibe

On Tue, 1 Jan 2019 at 11:03 PM, javed khan <[hidden email]> wrote:
Hello Elbe, thanks a lot for your response.

I am using the Explorer interface. Can we get all the predicted and actual values via Explorer?

Best regards

On Tue, Jan 1, 2019 at 2:31 AM Eibe Frank <[hidden email]> wrote:
Which user interface are you using to work with WEKA? In the Explorer, there is an option available under "More options..." in the Classify tab that triggers output of predicted and actual values.

A similar option is available at the command-line. Use -h together with java weka.Run <name of learning scheme> to get a list of the available command-line options.

Cheers,
Eibe

On Tue, Jan 1, 2019 at 12:37 AM javed khan <[hidden email]> wrote:
We get the MAE, RMSE etc values when we predict values in regression. But how can we get the individual values of output variable like we do in r language as : fitted(model)

Thanks
_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit 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]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html