you can also specify additional packages on the decorator@PipelineDecorator.component(..., packages=["tqdm>=2.1", "scikit-learn"]) def step_one(...): # code here
Oh, is your pipeline code a part of a git repository ?
OK, I got it by modifying the .conf file and putting the credentials on node
Nice! 🙂
Hi DepressedFox45
Basically move the import into the function, it will automatically detect the package.@PipelineDecorator.component(...) def step_one(...): import sklearn import pandas as pd # stuff
Make sense ?
I now I'm logged in the github repo with ClearML, but how can I link on the Pipeline step?
Hmm, this means the step should have included the git repo itself, which means the code should have been able to import the .py
Can you see the link to the git repository on the Pipeline step Task ?
Make sense AgitatedDove14 . But I'm triyng to import a .py in a folder, how can I do that?
OK, I got it by modifying the .conf file and putting the credentials on node