Model tree cpu.arff

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

Model tree cpu.arff

AndreiA-3
You can help me!!
Observing LM1 and LM2, as we can evaluate that the results were good? I
don't understand what wants to say the percentil.  How is he calculated?

The correlation coefficient is loud, even so, the% of LM1 it is very
low. Can we say that in this example the results were good?

LMMAX <= 14000 : LM1 (141/4.178%)
LMMAX >  14000 : LM2 (68/50.073%)
Correlation coefficient                  0.9766
Correlation coefficient                  0.9766
Mean absolute error                     13.6917
Root mean squared error                 35.3003
Relative absolute error                 15.6194 %
Root relative squared error             22.8092 %
Total Number of Instances              209


Thank you!!




_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Reply | Threaded
Open this post in threaded view
|

Re: Model tree cpu.arff

Paul m
To interpret your question - your asking what is the meaning of
"Relative absolute error " is and whether the results you posted are
'good' or not.

What seems to be a good explanation can be found here :
http://grb.mnsu.edu/grbts/doc/manual/Error_Measurements.html

Something more weka related, look here :
https://list.scms.waikato.ac.nz/mailman/htdig/wekalist/2004-August/002817.html

/paul

On 7/3/05, Andreia Vieira <[hidden email]> wrote:

> You can help me!!
> Observing LM1 and LM2, as we can evaluate that the results were good? I
> don't understand what wants to say the percentil.  How is he calculated?
>
> The correlation coefficient is loud, even so, the% of LM1 it is very
> low. Can we say that in this example the results were good?
>
> LMMAX <= 14000 : LM1 (141/4.178%)
> LMMAX >  14000 : LM2 (68/50.073%)
> Correlation coefficient                  0.9766
> Correlation coefficient                  0.9766
> Mean absolute error                     13.6917
> Root mean squared error                 35.3003
> Relative absolute error                 15.6194 %
> Root relative squared error             22.8092 %
> Total Number of Instances              209
>
>
> Thank you!!
>
>
>
>
> _______________________________________________
> Wekalist mailing list
> [hidden email]
> https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
>

_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist