Do you know the History of Data Science? READ MORE

Advance Designer Certificate Exam Step 9

Solved!
timchen
Level 1
Advance Designer Certificate Exam Step 9

Hello, 

I am working on creating the flow on Dataiku Advance Designer Certificate Assessment.

I am having an issue with redispatch

From the hands-on tutorial, there is an option for Redispatch:

Screen Shot 2021-08-28 at 14.21.56.png

However, in the flow I am creating, I didn't see this option

Screen Shot 2021-08-28 at 14.23.02.png

Did I miss any steps or I am thinking to the wrong step.

Can you please guide me how to solve step #9?

(I have confirmed from step 8 I have 67 rows in the dataset 2010)

I am on version 9.0.2

Thank you! 

Best,

Tim

0 Kudos
1 Solution
AlexT
Dataiker
Dataiker

Hi Tim,

The Redispatch option will show up only after you add partitions to the output dataset. The steps would be :

1. Create the Sync recipe, go back to the flow. 

2. Edit the output dataset partitions go to Settings -> Partitioning -> Activate Partitioning, Add the required dimension and Save

3.  Got back the Sync recipe and the Redispatch option should appear.

 

 

View solution in original post

4 Replies
AlexT
Dataiker
Dataiker

Hi Tim,

The Redispatch option will show up only after you add partitions to the output dataset. The steps would be :

1. Create the Sync recipe, go back to the flow. 

2. Edit the output dataset partitions go to Settings -> Partitioning -> Activate Partitioning, Add the required dimension and Save

3.  Got back the Sync recipe and the Redispatch option should appear.

 

 

View solution in original post

timchen
Level 1
Author

Hello Alex,

Thanks for getting back to me. Can you give me some hint about the pattern and name.

I don't know how to get the setting align with the requirement.

Thank you!  

0 Kudos
AlexT
Dataiker
Dataiker

Hi Tim,

The instructions for step 9 mention :

Using a Sync recipe, redispatch the data in the Online_Retail_to_2010 dataset into partitions given by the stock_code (a discrete identifier). That is, create partitions for each stock code.

So you should end up with a dataset partitioned by the discrete identifierr stock_code. Hope this helps!

0 Kudos
AlexT
Dataiker
Dataiker

Additionally, there is an example here: https://knowledge.dataiku.com/latest/kb/data-prep/partitions/partitioning-redispatch.html that may be helpful in understanding this concept. 

0 Kudos
Labels (1)