Sampling

Sampling on the notebook

Add the following to your setup.py

    # given that ddataflow config usually sits on the root of the project
    # we add it to the package data manually if we want to access the config 
    # installed as a library
    py_modules=[
        "ddataflow_config",
    ],

With DBrocket

Cell 1

%pip install --upgrade pip 
%pip install ddataflow
%pip install /dbfs/temp/user/search_ranking_pipeline-1.0.1-py3-none-any.whl --force-reinstall`

Cell 2

from ddataflow_config import ddataflow
ddataflow.save_sampled_data_sources()

Then use dry_run=False when you are ready to copy.