Example ML pipeline for DDB master. The data needed to run this pipeline can be found at:
https://www.kaggle.com/jeanmidev/smart-meters-in-london
Please use this repository as a template for your funda product.
IMPORTANT: the source code should reside in modules consisting of classes and functions (no top level code) in a folder at the top level (see the forecasting folder above). This folder should contain an empty init.py file so that python recognizes it as a package. Moreover, the setup.py file should point to this folder in the packages() line (see template). With this setup, you can create a conda environment with
conda create --name myenvironment python==3.8.5
and then do (from within your package folder):
conda activate myenvironment
pip install -e .
This then allows you to import modules from the package folder inside your run/run.py script, which should be the script implementing the pipeline steps (i.e., the script called from command line).