FAILED Evaluation Pipeline

Help AI Center
,
1

I got this error

Evaluate only of new_time_series 1.0 launched - Run 0c8891ad-4b90-49f8-be77-ee7dc0956c0c
Evaluate only of new_time_series 1.0 scheduled - Run 0c8891ad-4b90-49f8-be77-ee7dc0956c0c
Evaluate only of new_time_series 1.0 started - Run 0c8891ad-4b90-49f8-be77-ee7dc0956c0c
Evaluate only of new_time_series 1.0 failed - Run 0c8891ad-4b90-49f8-be77-ee7dc0956c0c
Error Details : could not execute statement; SQL [n/a]; nested exception is org.hibernate.exception.SQLGrammarException: could not execute statement

The code is still running and I got the results as expected but pipeline notified my as fail.

2

What’s the ML package you’re trying to run?

3

Hi @thuduong ,

Could you also let us know what was the configuration performed when Creating the Evaluation Pipeline ?

4

Hi, I am running a customized package of time series prediction. The training pipeline run successfully. In my ML package I dont use SQL so I dont why it has this kind of error.
the evaluation config as below

image
image1030×883 29.9 KB

5

@thuduong ,

Could you maybe cross check with the Training Pipeline datasets, if the same formatting is used for the test dataset as well. Also, Check how the Selection was done for Dataset when performed Training Pipeline.

We’ll ensure the Datasets and the configuration are in the required format first and then move for debugging in other methods.

6

The problem is that I am still able to get all results as expected. the log file show evaluation function runs to the end

image
image1366×466 34.3 KB

But dont know why pipeline is notified as failed