Coding the whole machine learning pipeline with custom python scripts
![pafj](https://us.v-cdn.net/6038231/uploads/Dataiku/nAvatar2.png)
Hi Dataiku Community,
We have an existing ML model sitting in our VM and we would like to move the entire pipeline (preprocessing, training, scoring, etc. which involves multiple python scripts including a main.py script to run all the remaining scripts in order) into the dataiku for automaton purposes. However, we don't want to use the visual interface. What is the best way to accomplish this approach?
I went through the knowledge tutorial materials, but I was not able to find anything that shows how to customize the entire pipeline.
Thanks
Answers
-
rolaelsebai Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1 Dataiker
Hi
. Thank you for reaching out pafj. This is a common use case! In short, you will be able to bring in the existing ML model into Dataiku. We'll be reaching out directly to understand what you are currently doing in greater detail to determine the best way to transfer such workloads into Dataiku.