Hi ,
I have deployed a model via Jupyter notebook . This model uses a dataset to perform tasks. The dataset is frequently updated with new data. Every time the dataset is updated I have to reexecute my code in jupyter notebook to have the last version.
Is there a solution to automate this?
Hey! Yes you definitely can!
You can create a transform for your model training code from the Jupyter workspace (docs here), and then you can setup a build schedule (docs here) to run your model training whenever the dataset updates.
Accessing the data pipeline page for the jupyter transform is probably easiest by using the Explore data lineage button in the jupyter workspace after you create the transform:
Hey!
Thank you very much. It’s working !
