Running and Monitoring Training Scripts in Azure Machine Learning
This lesson looks at the process of converting notebooks into scripts and then running those scripts from the command line. This involves converting notebook cells to a script, preparing code for production, and using the Python SDK to create a command job. Jobs’ parameters, metrics, artifacts, and models can be monitored and logged with MLFlow. This lesson looks at the ways this is achieved and how to view the logged elements via scripts and Azure Machine Learning Studio.
Learning Objectives
- Understand how to convert a Jupyter Notebook into a script
- Practice running a script as a command
- Use MLflow to log model training
Intended Audience
- Students wanting to know how to automate aspects of model training
- Students preparing for the DP-100: Designing and Implementing a Data Science Solution on Azure exam
Prerequisites
- Familiarity with data science concepts
- Some exposure to Python
Running a Script as a Command demo commands
rm -r azure-ml-labs -f
https://github.com/MicrosoftLearn/mslearn-azure-ml.git azure-mil-labs
cd azuer-ml-labs/Labs/08
./setup.sh
pip uninstall azure-ai-ml
pip install azure-ai-ml
git clone https://github.com/MicrosoftLearn/mslearn-azure-ml.git azure-mil-labs
Hallam is a software architect with over 20 years experience across a wide range of industries. He began his software career as a Delphi/Interbase disciple but changed his allegiance to Microsoft with its deep and broad ecosystem. While Hallam has designed and crafted custom software utilizing web, mobile and desktop technologies, good quality reliable data is the key to a successful solution. The challenge of quickly turning data into useful information for digestion by humans and machines has led Hallam to specialize in database design and process automation. Showing customers how leverage new technology to change and improve their business processes is one of the key drivers keeping Hallam coming back to the keyboard.