Re: [Dev] Sample Execution Summary -ML

2015-08-26 Thread Nirmal Fernando
Thanks a lot Fazlan! @Mano please fix these issues on samples asap.

On Wed, Aug 26, 2015 at 7:13 PM, Fazlan Nazeem fazl...@wso2.com wrote:

 Hi Nirmal,

 I have run the samples with the latest build and some of the samples
 cannot be successfully executed. Error messages for each failed sample is
 same in the first glance, but I have included(the links) the stack trace
 just in case if they are different.

 *failed samples*

 ---default---

- lasso-regression http://pastie.org/10376887
- linear-regression http://pastie.org/10376932

 --tuned--

- linear-regression http://pastie.org/10376981
- lasso-regression http://pastie.org/10376987
http://pastie.org/10376987
- ridge-regression http://pastie.org/10376992


 I haven't tested the *logistic-regression-with-hdfs *sample yet (
 because It needs a hadoop server running. Will do it)


 --
 Fazlan Nazeem

 *Software Engineer*

 *WSO2 Inc*
 Mobile : +94772338839
 %2B94%20%280%29%20773%20451194
 fazl...@wso2.com




-- 

Thanks  regards,
Nirmal

Team Lead - WSO2 Machine Learner
Associate Technical Lead - Data Technologies Team, WSO2 Inc.
Mobile: +94715779733
Blog: http://nirmalfdo.blogspot.com/
___
Dev mailing list
Dev@wso2.org
http://wso2.org/cgi-bin/mailman/listinfo/dev


Re: [Dev] Sample Execution Summary -ML

2015-08-26 Thread Manorama Perera
Thanks Fazlan for reporting this issue. This issue is due to the
improvements done in feature type (categorical/numerical) identification
algorithm.

In the above mentioned samples, wine-quality dataset [1] has been used.
According to the new algorithm, the response variable quality, is
identified as a categorical feature. As all these are numerical prediction
algorithms, and require a numerical variable as the response variable, this
exception[2] is thrown.

I've fixed it in [3]

[1] https://archive.ics.uci.edu/ml/datasets/Wine+Quality
[2] org.wso2.carbon.ml.core.exceptions.MLModelBuilderException: An error
occurred while building supervised machine learning model: Categorical
attribute quality cannot be used as the response variable of the Numerical
Prediction algorithm:
[3] https://github.com/wso2/product-ml/pull/228

Thanks.

On Wed, Aug 26, 2015 at 8:26 PM, Nirmal Fernando nir...@wso2.com wrote:

 Thanks a lot Fazlan! @Mano please fix these issues on samples asap.

 On Wed, Aug 26, 2015 at 7:13 PM, Fazlan Nazeem fazl...@wso2.com wrote:

 Hi Nirmal,

 I have run the samples with the latest build and some of the samples
 cannot be successfully executed. Error messages for each failed sample is
 same in the first glance, but I have included(the links) the stack trace
 just in case if they are different.

 *failed samples*

 ---default---

- lasso-regression http://pastie.org/10376887
- linear-regression http://pastie.org/10376932

 --tuned--

- linear-regression http://pastie.org/10376981
- lasso-regression http://pastie.org/10376987
http://pastie.org/10376987
- ridge-regression http://pastie.org/10376992


 I haven't tested the *logistic-regression-with-hdfs *sample yet (
 because It needs a hadoop server running. Will do it)


 --
 Fazlan Nazeem

 *Software Engineer*

 *WSO2 Inc*
 Mobile : +94772338839
 %2B94%20%280%29%20773%20451194
 fazl...@wso2.com




 --

 Thanks  regards,
 Nirmal

 Team Lead - WSO2 Machine Learner
 Associate Technical Lead - Data Technologies Team, WSO2 Inc.
 Mobile: +94715779733
 Blog: http://nirmalfdo.blogspot.com/





-- 
Manorama Perera
Software Engineer
WSO2, Inc.;  http://wso2.com/
Mobile : +94716436216
___
Dev mailing list
Dev@wso2.org
http://wso2.org/cgi-bin/mailman/listinfo/dev