Table of contents

What is Continuous Delivery?

Continuous Integration

Continuous Delivery vs. Continuous Deployment

Machine Learning Workflow

How does Continuous Delivery help with ML challenges?

Data Management

_- _Automated Data Pipeline

Experimentation

_- _Training Code

_- _Training Process

Production Deployment

_- _Application Code

Bringing it all together

People

Conclusion

Most of the principles and practices of traditional software development can be applied to Machine Learning(ML), but certain unique ML specific challenges need to be handled differently. We discussed those unique “Challenges Deploying Machine Learning Models to Production” in the previous article. This article will look at how Continuous Delivery that has helped traditional software solve its deployment challenges be applied to Machine Learning.

#machine-learning #devops4ml #continuous-delivery #mlops #data-science