Data Science with Python on the Microsoft Azure Platform

Machine Learning Introduction

Before ML can be applied, its key concepts should be discussed.

• Supervised versus Unsupervised Learning
• Machine learning methodology
• Data preparation
• Classification, regression and clustering
• Model evaluation
• Cognitive services
• Automated ML in Azure ML Services
• Working with the Azure ML Designer

Tools for Machine Learning in Azure

In this chapter we will introduce the different tools that are available to do Machine Learning in Microsoft Azure.

• Overview of Machine Learning in Azure
• Pretrained models
• Transfer Learning
• Graphical Approaches
• Coding Approaches

Getting started with Python

Since this training has no Python prerequisites, first the basics of Python are covered.

• Introducing the Python programming language
• Python environments
• Interactive development with Azure notebooks
• Variables and objects
• Common data structures: Lists, tuples, sets and dictionaries
• Functions
• Creating and using classes

Getting Started with 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 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. In this chapter, the basics of Azure Machine Learning are covered. We will discuss its components and architecture, create and connect to a workspace with Python, and see how to create compute resources.

• Why Azure Machine Learning?
• Components and Architecture of Azure Machine Learning
• Creating an Azure Machine Learning Workspace
• Setting up a Compute
• Data manipulations with Pandas

Data processing with SciPy

In data science its crucial to deal with tables: Loading, manipulating, data quality checks, … Dataframes can help out with that, and in this module the two most important Python packages for data manipulation are inspected: Numpy and Pandas.

• Numerical Python: Numpy
• Numpy data structures
• Pandas DataFrames
• Loading data with pandas
• Data manipulations with Pandas

Machine Learning 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 implemented, as well as model inspection.

• Machine learning specific data preprocessing: normalization, standardization, one-hot encoding
• Classification using decision trees, logistic regression and support vector machines
• Model tuning: working with hyper-parameters
• Building regression models with linear regression, SVM's and Neural networks
• Unsupervised learning: Clustering

Training and Deploying Models in Azure Machine Learning

In this chapter we will continue with Azure Machine Learning. Since the basics have already been covered, we can dive right into the Machine Learning part! We will see how to create experiments in which we can log the results of different attempts to train a model. We will see how to train models using different computes and pipelines. After evaluating the different models, the best model can be deployed as a webservice.

• Setting up computes and datastores
• Creating and querying experiments
• Creating and registering images
• Deploy images as web services

Getting started with 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 from a mathematical point of view this is 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