Operation directory
The operation directory contains all the logic and configuration to operationalize your data science project.
Execution folder
This folder contains all the code that manages the core scripts as, for instance, sending the training to a remote target, creating data science pipelines, using automl, etc. Basically, everything functionalities that is not core data science scripts.
Tests
This folder contains all the different tests to be run in the CI/CD pipeline: data tests, integration tests, unit tests, etc
Monitoring
This folder contains all the logic to monitor the artifacts: the models performance, model interpretability, data drifts, logging, model prediction logs.