SAS Viya Workbench Workshop

This repository contains all the relevant material for successfully following the SAS Viya Workbench Workshop.

Accessing the Material:

1. Create a workbench.

2. Configure your workbench in the following way.

3. Start your workbench instance.

4. Once the workbench is running, press Open Visual Studio Code

5. Open a New Terminal.

6. Type:

git clone https://github.com/fabiocerutisas/Live-SAS-WorkbenchWorkshop.git

Repository Structure

Data

It is a folder containing all the necessary data for the workshop. It is characterized by 2 folders:

• original_data: a folder containing the required 3 datasets for the workshop.
• cleaned_data: a folder containing the cleaned data as a result of the SAS Code.

Img

It is a folder containing all the relevant documentation screenshots that are embedded in both the sasnb and ipynb notebooks.

Python

It contains 3 python notebooks:

• overview_modelling.ipynb: It is a notebook containing an overview of ML Modelling in Python using both Sk-learn and Sasviya.ml.
• overview_scoring.ipynb: It imports Python ML Models created in overview_modelling.ipynb and scores data with them.
• connect_to_viya.ipynb: It registers the developed models in Model Manager.

SAS

It contains 4 folders and 7 SAS Notebooks. The folders available are:

• astore: containing the astore of the ML models that will be developed.
• format: storing the user-defined formats for different variables.
• generated_code: containing code to replicate WOE Transformations and Score new data.
• sas_tables: storing all the SAS Tables generated during the workshop.

The notebooks available are:

• 0_setup.sasnb: in which the key macro variables, functions and formats are defined.
• 1_data_import_and_cleaning.sasnb: the notebook walks through data cleaning and exploration.
• 2_data_analysis.sasnb: the notebook contains data analysis activities related to correlations and clustering.
• 3_ml_data_prep.sasnb: it prepares data for ML Modelling by applying WOE Transformations.
• 4_modelling.sasnb: creates and assess ML Models
• 5_score_new_data.sasnb: scores data with the previously created SAS ML Models.
• 6_connect_to_viya.sasnb: registers a developed model in Model Manager.