Production Machine Learning Systems

By Google Cloud
In this course, we will dive into the components and best practices of a high-performing ML system in production environments.
Course info
Level
Advanced
Updated
Sep 17, 2019
Duration
3h 17m
Table of contents
Welcome to the course
Getting started with GCP and Qwiklabs 4m Course Introduction 3m
Architecting Production ML Systems
The Components of an ML System 2m Lab Intro:Serving on Cloud AI Platform 2m The Components of an ML System:Data Transformation + Trainer 2m The Components of an ML System:Tuner + Model Evaluation and Validation 2m The Components of an ML System:Integrated Frontend + Storage 2m Training Design Decisions 5m The Components of an ML System:Serving 1m Serving Design Decisions 5m Designing from Scratch 3m Lab Solution:Serving on Cloud AI Platform 4m Introduction 2m Serving on Cloud AI Platform 0m The Components of an ML System:Orchestration + Workflow 4m The Components of an ML System:Data Analysis and Validation 4m
Ingesting data for Cloud-based analytics and ML
Data on Other Clouds 2m Data On-Premise 3m Demo:Automatic ETL Pipelines into GCP 4m Demo:Load data into BigQuery 5m Large Datasets 4m Existing Databases 2m Introduction 5m
Designing Adaptable ML systems
Introduction 4m Debugging a Production Model 4m Changing Distributions 3m Lab Intro:Serving ML Predictions in batch and real-time 2m Mitigating Training-Serving Skew through Design 1m Serving ML Predictions in batch and real-time 0m Summary 1m Right and Wrong Decisions 3m System Failure 2m Exercise:Adapting to Data 2m Adapting to Data 3m Lab Solution:Serving ML Predictions in batch and real-time 9m
Designing High-performance ML systems
Training 6m Parallel pipelines 6m TensorFlow Records 1m Predictions 3m Inference 4m Why distributed training? 5m Parameter Server Approach 3m Introduction 1m Native TensorFlow Operations 3m Data parallelism with All Reduce 5m Faster input pipelines 3m Distributed training architectures 6m
Hybrid ML systems
Demo:KubeFlow 22m Kubeflow: End to End 0m Summary 2m Optimizing for Mobile 6m Machine Learning on Hybrid Cloud 5m KubeFlow 3m Introduction 5m TensorFlow Lite 2m Embedded Models 3m
Course Summary
Summary 2m
Description
Course info
Level
Advanced
Updated
Sep 17, 2019
Duration
3h 17m
Description

In this course, we will dive into the components and best practices of a high-performing ML system in production environments.

About the author
Google Cloud
About the author

Build, innovate, and scale with Google Cloud Platform.

More from the author
Security Best Practices in Google Cloud
Intermediate
1h 23m
May 20, 2020
Mitigating Security Vulnerabilities on Google Cloud Platform
Intermediate
55m
May 20, 2020
Managing Security in Google Cloud Platform
Intermediate
1h 56m
May 20, 2020
More courses by Google Cloud

You're the smartest person in the room

PROVE IT

Your team won't just close the skills gap

THEY'LL HURDLE IT