"(Normal ?) bug : log all criteria / Optimization of cluster model"

lionelderkrikor

你好,

It's to report a bug when the parameterlog all criteriais checked for the optimization of a cluster model (Kmeans) .

When the process is executed, RapidMiner raise the following error :

java.lang.ArrayIndexOutOfBoundsException

When RM create theOptimize Parametersresults, each row has in theory a different length - (length(row(i+1)) = length(row(i)) + 1 -

because for each row , RM addAvg. within centroid distance_cluster_i.So when RM try to create the second row, it raise an error

because the dimensions of the table change.

I hope it is understanble. Here the process :

<运营商激活= " true " class = "过程”兼容ibility="8.0.001" expanded="true" name="Process">



https://archive.ics.uci.edu/ml/machine-learning-databases/00292/Wholesale customers data.csv"/>





<列出关键="annotations">



<列出关键="data_set_meta_data_information">

















<列出关键="parameters">

What is your opinion about that, do you think it deserves "Product feedback" ?

Regards,

Lionel

lionelderkrikor

Hi again,

Little update : the problem concernlog all criteria, so theLoop parametersoperator is concerned too.

Regards,

Lionel

Telcontar120

I would report this as a bug in the Product Feedback board.

sgenzer

moving this thread to Product Feedback.

Scott

jczogalla

Hi Lionel,

we are looking into this. We will keep you updated here!

Regards

Jan

sgenzer

jczogalla

Hi Lionel,

we temporarily fixed this for the Beta by throwing an appropriate user error and are working on a permanent solution

Regards

Jan

lionelderkrikor

Hi Jan,

Thanks you for your feedback,

Regards,

Lionel

sgenzer

@jczogalla- mark this resolved or still investigating?

jczogalla

@sgenzer- Leave it as investigating please.