The labs constitute a series of tasks that develop a data science solution starting with data analysis and culminating with deploying a trained model to the Azure Machine Learning service.
The following is a summary of the lab objectives for each module:
Lab 1 - Train a Classification Model in Python using scikit-learn
In this lab, the students will train a classification model using Python in an Azure Notebook. The model will predict what type of bicycle a customer is most likely to buy. Some exploratory data analysis and feature engineering will be required.
Lab 2 - Train and Deploy a Model using Azure Machine Learning service
In this lab, the students will train, register, deploy, and test, the model same model they developed in lab 1. Emphasis is on using the Azure Machine Learning service Python SDK.
Lab 3 - Use AutoML and Hyperdrive to Automate Machine
In this lab, students will use AutoML and HyperDrive to select the best performing machine learning classification model and determine the optimal hyperparameter values. The goal is to see if they can get a model that performs better than the one they trained manually.