| This repo is under construction until 9th Jan and changes will be made|
Going Beyond Model Development - Developing ML Cloud Native Solutions
We are now able to do incredible things with machine learning packages - create powerful models, learn from vast amounts of data and evaluate our models. But now we are able to do that, expectations have been raised. Designing, building and evaluating a model is only one part of a much bigger process if you want to move models to production.
In this session we will talk through other elements that are important in the machine learning lifecycle such as distributed compute, reusable code pipelines, retraining and security. All of this is available when we look to cloud services for support - leveraging cloud computing technologies mean applying software engineering principles to machine learning projects
In this session we will cover :
- Working with Deep Learning frameworks on Azure
- Working on an end-to-end cloud native scenario
- Building reusable code pipelines
- Deploying machine learning models and retraining them
- Working as a machine learning engineering team
I will demonstrate with descriptions and demos the content and there will be a code repository containing all necessary resources for you to build your own machine learning models using Microsoft Azure Machine Learning.
- Presentation File
- MS Learn Collection for the session
- Code Files and Structure: The folders below represent the different sections of the session we will cover
Note: All code is demo code and comes with no warranty or SLA or promise to keep updated for the foreseeable future