deploying python recipe
Hello,
I am trying to build a content-based recommendation model that is applied in a python recipe and I want to deploy it, how can I do that from the python recipe??
Also, I thought of converting it to a plugin but will that be helpful in my case?
Thanks in advance.
Best Answer
-
Alexandru Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1,226 Dataiker
Hi @Tokka
,
The API endpoint would make your Python code callable from the REST API endpoint.
So if I understand correctly, you want to be able to deploy the project bundle to the automation node via Python code. This would not be achieved via recipe typically, but instead using Python or various scenario steps already available like "Create bundle":This is possible if you want to go a custom code route. You can find code snippets here:
From this tutorial: https://knowledge.dataiku.com/latest/mlops-o16n/ci-cd/tutorial-jenkins-pipeline.html#deploy-to-prod-stage
This would be typically run as Python step in a scenario rather than a recipe.
https://downloads.dataiku.com/public/website-additional-assets/code/dss_pipeline-master.zipHope that helps!
Answers
-
Alexandru Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1,226 Dataiker
Hi @Tokka
,
By deploying, do you mean you want to be able to access this via API?
In that case, you could create a Python API endpoint :https://doc.dataiku.com/dss/latest/apinode/introduction.html#exposing-arbitrary-python-and-r-functions
If you simply want for other DSS users to reuse this model/code you could leverage Code libraries or a plugin as you mentioned