what are our options with Dataiku for using .whl file present on our own repository? or local path?
What are our options with Dataiku for using .whl files presently stored on our own repository? Can a .whl be a plugin? or is it better to reference the .whl in the code environment?
Operating system used: Almalinux
Answers
-
Turribeach Dataiku DSS Core Designer, Neuron, Dataiku DSS Adv Designer, Registered, Neuron 2023 Posts: 2,467 NeuronSimply add your custom Python repository in the Dataiku configuration and then you can install your custom packages in your code environments normally.
-
Grixis PartnerApplicant, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 105 ✭✭✭✭✭✭Hey,
I have never got how to manage correctly wheels from misc or add "extra options for pip install". So I do it from ‘resources’ section of each env, which allows you to manage your whl files directly in the ressource's environment code directory.
And I add a initialization python script which browses paths and executes the install pip of .whl files from the current code env, /resources/python/code_env_name etc.
I don't even if it's right (as a good practice).