Building your own Models with Azure Machine Learning

Machine Learning Introduction

Before ML can be applied, the key concepts of machine learning need to be discussed.

• Supervised versus Unsupervised Machine Learning
• Why use Machine Learning?
• Splitting data in training, test and validation sets

Applying a machine learning methodology

When applying machine learning in an enterprise context, it's important to have a methodology to plan, document and execute the different steps in the machine learning proces.

• The 6 steps of the CRISP-DM Machine Learning Methodology
• Data Preparation
• Classification, Regression and Clustering
• Measure Model Quality
• Cross-validation and Hyper-Parameters
• Machine Learning in the Microsoft AI Stack

Setup your Machine Learning environment

Before you can start developing machine learning solutions, you must first create an environment in which you can store and run your Python code. Both local and cloud solutions are presented.

• Configuring your Environment
• Working with Python in Notebooks
• Using Environments and Notebooks in Microsoft Fabric

Prepare your data for Machine Learning

Like all data used for business intelligence, also data for machine learning should be cleansed. But machine learning data has some special requirements. This module shows how to use Python to prepare your data for machine learning.

• Exploring the scikit-learn package
• Estimators and Transformers
• One-Hot encoding for categorical data
• LabelEncoder and OrdinalEncoder
• Dealing with missing values
• Handling outliers
• Standardization and Normalization for continuous values
• Preparing free text fields
• LAB: Preparing data for Machine Learning

Build your own Machine Learning Model with scikit-learn

Many business problems can be tackled by basic machine learning techniques. In this module the most common machine learning approaches such as linear regression and random forests are used with Python. We also pay attention to model inspection.

• Classification
• Decision Trees, Logistic Regression and Support Vector Machines
• Model tuning: Working with Hyper-Parameters
• Regression Models
• Linear Regression and Neural networks
• Unsupervised learning: Clustering
• LAB: Classification and Regression with scikit-learn

Azure Machine Learning

Machine learning on a local machine and a small dataset is one thing, running this on larger datasets or more CPU-hungry techniques (or even GPUs) can become a challenge. Another problem is deploying your model: How can we easily call the resulting model from within other applications? Azure Machine Learning helps answering these questions.

• Azure Machine Learning overview
• Create an Azure Machine Learning Workspace
• Setting up Compute and Datastores
• Creating and Querying Jobs and Experiments
• Deploying and Using Models
• Creating and Registering Images
• Deploy Images as Web Services
• LAB: Building ML Models in Azure Machine Learning

Automated Machine Learning

Part of the machine learning process can be automated. Azure Automated machine learning provides a web portal and a Python API to automate data preprocessing and model training.

• Building and Deploying Automated ML models
• Calling Automated ML from Python
• LAB: Working with Automated ML

Microsoft Fabric Data Science

The Data Science experience in Microsoft Fabric provides notebooks, so all the previous Python code could be run in their. But Microsoft Fabric provides extra functionality as well: From pre-installed machine learning frameworks (SparkML, FLAML) to model management (MLFlow).

• Scale out your machine learning with SparkML
• Log your machine learning with MLFlow in Experiments
• Microsoft Fabric Model Management
• FLAML: Automated machine learning in Microsoft Fabric
• LAB: Train your machine learning model in Microsoft Fabric

Getting started with Keras for Deep Learning

From all the machine learning techniques there is one that gets popular for more challenging problems: Multiple layers of neural networks, better known as deep learning. For problems such as image recognition, speech understanding etc. this is currently the way to go. But it's from a mathematical point of view a very challenging technique. In this module the basics of deep learning are introduced.

• From Neural networks to Deep learning
• Overview of Deep Learning Frameworks
• Getting started with the Keras Framework