Survey banner
The Dataiku Community is moving to a new home! Some short term disruption starting next week: LEARN MORE

Control the execution of steps in Scenario

Obireddy
Level 2
Control the execution of steps in Scenario

Hi Team,

I have created scenario to run daily wise to run the pipeline every day. in that scenario I have created 4 steps as below.

1)step 1: to get latest data available or not in table if available pick date_time of latest date

2)step 2: set the date_time as project variable.

3)step 3: based on date_time filter the data from table and keep it in test dataset(for ML model prediction).

4)step 4: run the ML model for test data.

 

Here when I run the scenario, some times latest data may not exist in table, in that case date_time variable populating as null value in project variable and  scenario failing at step 3.

 

Now I want to stop the scenario when date_time value is null and scenario should stop running instead of failing at step 3(step 1 and step 2 should run). i.e whenever data_time is null 

0 Kudos
1 Reply
Turribeach

This is easy to do by adding a conditional logic to execute your remaining scenario steps. 

See this post: https://community.dataiku.com/t5/Using-Dataiku/Conditional-execute-of-scenario-step-without-steps-fa...

More info about this technique here: https://community.dataiku.com/t5/What-s-New/Want-to-Control-the-Execution-of-Scenario-Steps-With-Con...

Personally I would use row count in your dataset as it is a built-in metric. Here is a sample on how to get the row count from a dataset metric:

toNumber(filter(parseJson(stepOutput_Compute_Metrics)['Project_Key.Dataset_Name_NP']['computed'], x, x["metricId"]=="records:COUNT_RECORDS")[0].value)

 

 

0 Kudos