Machine Learning Engineering

Paths

Expanded

Machine Learning Engineering

Authors: Abdul Rehman Yousaf, Mohammed Osman, Axel Sirota, Niraj Joshi

On completion of this path, the learner will be able to design and operationalize solutions to maximize the performance and scale of Machine Learning. These solutions will include... Read more

Design Principles for Machine Learning Framework Demystifying Machine Learning Operations (MLOps) Productionalizing Data Pipelines with Apache Airflow Principles for Data Quality Measures Evaluating tools for Machine learning operations

Pre-requisites

Machine Learning Literacy

Beginner

This section is focussed on understanding the requirements of a framework that can cope with Machine Learning applications and their tasks. You will also learn how to evaluate framework on performance, execution, and practical usage requirements.

Design Principles for Machine Learning Framework

by Abdul Rehman Yousaf

Jul 20, 2021 / 1h 23m

1h 23m

Start Course
Description

Design patterns capture best practices and solutions to common problems. In this course, Design Principles for Machine Learning Framework, you’ll learn to implement scalable data pipelines for machine learning systems. First, you’ll explore guiding principles for machine learning operations. Next, you’ll discover why you should use data pipelines to process incoming data in real-time. Finally, you'll learn how to evaluate the performance of a machine learning system. When you’re finished with this course, you’ll have the skills and knowledge of machine learning operations needed to orchestrate a scalable and powerful system.

Table of contents
  1. Course Overview
  2. Introducing Machine Learning Operations
  3. Designing Data Pipelines for Scalability and Optimizations
  4. Evaluating the Performance of a Machine Learning System

Intermediate

In this section, you will explore Machine Learning operations with focus on deploying Machine Learning models at scale with the use of Apache Airflow. One will learn to author, schedule, and monitor data pipelines through practical examples using Apache Airflow.

Demystifying Machine Learning Operations (MLOps)

by Mohammed Osman

Nov 23, 2020 / 2h 14m

2h 14m

Start Course
Description

Machine Learning is a robust science that can empower the business with unique competitive advantages to address several challenges, such as sales price prediction, customer segment classification, and product recommendation. In this course, Demystifying Machine Learning Operations (MLOps), you’ll learn to implement machine learning operations into your machine learning project. First, you’ll explore how to apply machine learning operations (MLOps) practices for your infrastructure. Next, you’ll discover how machine learning operations (MLOps) during model development. Finally, you’ll learn how to apply machine learning operations (MLOps) after model deployment. When you’re finished with this course, you’ll have the skills and knowledge of machine learning operations needed to manage the MLOps lifecycle of your project.

Table of contents
  1. Course Overview
  2. Understanding Machine Learning Operations
  3. Machine Learning Operations for Infrastructure
  4. Implementing Best Practices for Model Development
  5. Implementing Best Practices for Model Deployment

Productionalizing Data Pipelines with Apache Airflow

by Axel Sirota

Dec 9, 2020 / 2h 12m

2h 12m

Start Course
Description

Production-grade Data Pipelines are hard to get right. Even when they are done, every update is complex due to its central piece in every organization's infrastructure. In this course, Productionalizaing Data Pipelines with Apache Airflow, you’ll learn to master them using Apache Airflow. First, you’ll explore what Airflow is and how it creates Data Pipelines. Next, you’ll discover how to make your pipelines more resilient and predictable. Finally, you’ll learn how to distribute tasks with Celery and Kubernetes Executors. When you’re finished with this course, you’ll have the skills and knowledge of Apache Airflow needed to make any Data Pipelines production grade.

Table of contents
  1. Course Overview
  2. Introducing Apache Airflow
  3. Dissecting the Components of a Pipeline
  4. Demystifying Common DAGs Pitfalls
  5. Abstracting Functionality
  6. Scaling Airflow
  7. Final Thoughts

Advanced

In this section, you will focus on exploring best practices for data quality management for Machine Learning and Machine Learning operations with Kubeflow, Dask and Sagemaker.

Principles for Data Quality Measures

by Niraj Joshi

Mar 26, 2021 / 43m

43m

Start Course
Description

Data quality is an important prerequisite prior to machine learning modelling. It is of utmost importance to thoroughly assess data quality before model building. In this course, Principles for Data Quality Measures, you’ll learn to build MLOps pipelinse and explore best practices for metadata management. First, you’ll explore data discovery and cataloging. Next, you’ll discover data profiling and quality checks. Finally, you’ll learn to explore data lineage and the best metadata management practices and analyze the MLOps cycle. By the end of this course, you’ll gain a better understanding of data discovery, profiling, and metadata management of the ML Model building process.

Table of contents
  1. Course Overview
  2. Introducing Data Discovery and Cataloging
  3. Evaluating Data Quality and Profiling
  4. Tracking Data Lineage and Governance
  5. Exploring Best Practices for Metadata Management

Coming Soon

Evaluating Tools for Machine learning Operations

Coming Soon

by Pluralsight