(Spark-ml) java.util.NosuchElementException: key not found exception on doing prediction and computing test error.

[neha nihal <ne...@gmail.com>]

Hi,

I am using Apache spark 2.0.2 randomforest ml (standalone mode) for text
classification. TF-IDF feature extractor is also used. The training part
runs without any issues and returns 100% accuracy. But when I am trying to
do prediction using trained model and compute test error, it fails with
java.util.NosuchElementException: key not found exception.
Any help will be much appreciated.

Thanks & Regards

(Spark-ml) java.util.NosuchElementException: key not found exception on doing prediction and computing test error.

[neha nihal <ne...@gmail.com>]

Thanks. Its working now. My test data had some labels which were not there
in training set.

On Wednesday, June 28, 2017, Pralabh Kumar <pralabhkumar@gmail.com
<javascript:_e(%7B%7D,'cvml','pralabhkumar@gmail.com');>> wrote:

> Hi Neha
>
> This generally occurred when , you training data set have some value of
> categorical variable ,which in not there in your testing data. For e.g you
> have column DAYS ,with value M,T,W in training data . But when your test
> data contains F ,then it say no key found exception .  Please look into
> this  , and if that's not the case ,then Could you please share your code
> ,and training/testing data for better understanding.
>
> Regards
> Pralabh Kumar
>
> On Wed, Jun 28, 2017 at 11:45 AM, neha nihal <ne...@gmail.com>
> wrote:
>
>>
>> Hi,
>>
>> I am using Apache spark 2.0.2 randomforest ml (standalone mode) for text
>> classification. TF-IDF feature extractor is also used. The training part
>> runs without any issues and returns 100% accuracy. But when I am trying to
>> do prediction using trained model and compute test error, it fails with
>> java.util.NosuchElementException: key not found exception.
>> Any help will be much appreciated.
>>
>> Thanks
>>
>>
>

Re: (Spark-ml) java.util.NosuchElementException: key not found exception on doing prediction and computing test error.

[Pralabh Kumar <pr...@gmail.com>]

Hi Neha

This generally occurred when , you training data set have some value of
categorical variable ,which in not there in your testing data. For e.g you
have column DAYS ,with value M,T,W in training data . But when your test
data contains F ,then it say no key found exception .  Please look into
this  , and if that's not the case ,then Could you please share your code
,and training/testing data for better understanding.

Regards
Pralabh Kumar

On Wed, Jun 28, 2017 at 11:45 AM, neha nihal <ne...@gmail.com> wrote:

>
> Hi,
>
> I am using Apache spark 2.0.2 randomforest ml (standalone mode) for text
> classification. TF-IDF feature extractor is also used. The training part
> runs without any issues and returns 100% accuracy. But when I am trying to
> do prediction using trained model and compute test error, it fails with
> java.util.NosuchElementException: key not found exception.
> Any help will be much appreciated.
>
> Thanks
>
>

Fwd: (Spark-ml) java.util.NosuchElementException: key not found exception on doing prediction and computing test error.

[neha nihal <ne...@gmail.com>]

Hi,

I am using Apache spark 2.0.2 randomforest ml (standalone mode) for text
classification. TF-IDF feature extractor is also used. The training part
runs without any issues and returns 100% accuracy. But when I am trying to
do prediction using trained model and compute test error, it fails with
java.util.NosuchElementException: key not found exception.
Any help will be much appreciated.

Thanks