Production Machine Learning Systems

Production Machine Learning Systems

This course is part of multiple programs.

Instructor: Google Cloud Training

Access provided by US Postal Service

41,025 already enrolled

6 modules

Gain insight into a topic and learn the fundamentals.

1,029 reviews

Advanced level

Designed for those already in the industry

Flexible schedule

2 weeks at 10 hours a week

Learn at your own pace

87%

Most learners liked this course

6 modules

Gain insight into a topic and learn the fundamentals.

1,029 reviews

Advanced level

Designed for those already in the industry

Flexible schedule

2 weeks at 10 hours a week

Learn at your own pace

87%

Most learners liked this course

What you'll learn

Compare static versus dynamic training and inference
Manage model dependencies
Set up distributed training for fault tolerance, replication, and more
Export models for portability

Skills you'll gain

Tools you'll learn

Details to know

Shareable certificate

Add to your LinkedIn profile

Assessments

4 assignments

Taught in English

See how employees at top companies are mastering in-demand skills

Learn more about Coursera for Business

logos of Petrobras, TATA, Danone, Capgemini, P&G and L'Oreal

Build your subject-matter expertise

This course is available as part of

When you enroll in this course, you'll also be asked to select a specific program.

Learn new concepts from industry experts
Gain a foundational understanding of a subject or tool
Develop job-relevant skills with hands-on projects
Earn a shareable career certificate

There are 6 modules in this course

In this course, we dive into the components and best practices of building high-performing ML systems in production environments. We cover some of the most common considerations behind building these systems, e.g. static training, dynamic training, static inference, dynamic inference, distributed TensorFlow, and TPUs. This course is devoted to exploring the characteristics that make for a good ML system beyond its ability to make good predictions.

This module previews the topics covered in the course and how to use Qwiklabs to complete each of your labs using Google Cloud.

What's included

2 videos2 readings

This module explores what else a production ML system needs to do and how to meet those needs. You review how to make important, high-level, design decisions around training and model serving need to make in order to get the right performance profile for your model.

What's included

9 videos1 reading1 assignment1 app item1 plugin

9 videos Total 35 minutes

Architecting ML systems 2 minutes
Data extraction, analysis, and preparation 5 minutes
Model training, evaluation, and validation 2 minutes
Trained model, prediction service, and performance monitoring 3 minutes
Training design decisions 5 minutes
Serving design decisions 6 minutes
Designing from scratch 3 minutes
Using Vertex AI 9 minutes
Lab Introduction: Structured data prediction 0 minutes

1 reading Total 10 minutes

Architecting production ML systems 10 minutes

1 assignment Total 16 minutes

Architecting production ML systems 16 minutes

1 app item Total 120 minutes

Lab: Structured data prediction using Vertex AI 120 minutes

1 plugin Total 15 minutes

Accessing and completing labs 15 minutes

In this module, you learn how to recognize the ways that our model is dependent on our data, make cost-conscious engineering decisions, know when to roll back our models to earlier versions, debug the causes of observed model behavior and implement a pipeline that is immune to one type of dependency.

What's included

14 videos1 reading1 assignment3 app items

14 videos Total 45 minutes

Introduction 3 minutes
Adapting to data 3 minutes
Changing distributions 4 minutes
Lab: Adapting to data 2 minutes
Right and wrong decisions 4 minutes
System failure 2 minutes
Concept drift 9 minutes
Actions to mitigate concept drift 3 minutes
TensorFlow data validation 4 minutes
Components of TensorFlow data validation 5 minutes
Lab Introduction: Introduction to TensorFlow data validation 0 minutes
Lab Introduction: Advanced visualizations with TensorFlow data validation 1 minute
Mitigating training-serving skew through design 2 minutes
Diagnosing a production model 4 minutes

1 reading Total 10 minutes

Designing adaptable ML systems 10 minutes

1 assignment Total 14 minutes

Designing adaptable ML systems 14 minutes

3 app items Total 360 minutes

Lab: Introduction to TensorFlow data validation 120 minutes
Lab: Advanced visualizations with TensorFlow data validation 120 minutes
Lab: Vertex AI: Training and Serving a Custom Model 120 minutes

In this module, you identify performance considerations for machine learning models. Machine learning models are not all identical. For some models, you focus on improving I/O performance, and on others, you focus on squeezing out more computational speed.

What's included

14 videos1 reading1 assignment2 app items

14 videos Total 42 minutes

Introduction 1 minute
Training 6 minutes
Predictions 3 minutes
Why distributed training is needed 3 minutes
Distributed training architectures 8 minutes
TensorFlow distributed training strategies 1 minute
Mirrored strategy 3 minutes
Multi-worker mirrored strategy 4 minutes
TPU strategy 2 minutes
Parameter server strategy 2 minutes
Lab Introduction: Distributed training with Keras 0 minutes
Training on large datasets with tf.data API 5 minutes
Lab introduction: TPU-speed data pipelines 0 minutes
Inference 4 minutes

1 reading Total 10 minutes

Designing high-performance ML systems 10 minutes

1 assignment Total 8 minutes

Designing high-performance ML systems 8 minutes

2 app items Total 240 minutes

Lab: Distributed training with Keras 120 minutes
Lab: TPU-speed data pipelines 120 minutes

Understand the tools and systems available and when to leverage hybrid machine learning models.

What's included

7 videos1 reading1 assignment1 app item

7 videos Total 19 minutes

Introduction 5 minutes
Machine Learning on Hybrid Cloud 5 minutes
KubeFlow 3 minutes
Lab Introduction: Kubeflow Pipelines with AI Platform 1 minute
TensorFlow Lite 2 minutes
Optimizing TensorFlow for mobile 2 minutes
Summary 2 minutes

1 reading Total 10 minutes

Hybrid ML systems 10 minutes

1 assignment Total 8 minutes

Hybrid ML systems 8 minutes

1 app item Total 120 minutes

Running Pipelines on Vertex AI 120 minutes

This module reviews what you learned in this course.

What's included

1 video2 readings

Earn a career certificate

Add this credential to your LinkedIn profile, resume, or CV. Share it on social media and in your performance review.

Instructor

Instructor ratings

(85 ratings)

Google Cloud Training

Google Cloud

2,064 Courses 3,938,230 learners

Offered by

Google Cloud

Why people choose Coursera for their career

Felipe M.

Learner since 2018

"To be able to take courses at my own pace and rhythm has been an amazing experience. I can learn whenever it fits my schedule and mood."

Jennifer J.

Learner since 2020

"I directly applied the concepts and skills I learned from my courses to an exciting new project at work."

Larry W.

Learner since 2021

"When I need courses on topics that my university doesn't offer, Coursera is one of the best places to go."

Chaitanya A.

"Learning isn't just about being better at your job: it's so much more than that. Coursera allows me to learn without limits."

Learner reviews

5 stars
70.77%
4 stars
21.26%
3 stars
5.43%
2 stars
1.45%
1 star
1.06%

Showing 3 of 1029

Reviewed on Sep 22, 2020

Unlike pure technical courses, this one specially packs you with knowledge that you may find yourself face to. The course is really well designed and the content is crystal clear, just Awesome !

Reviewed on May 8, 2020

I got lots of new skills and I think it's a great course for ML

Reviewed on Jan 14, 2019

Very practical which was nice. Thank you for adding the Quicklabs that helped a lot.

View more reviews