Production Machine Learning Systems

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

Build, innovate, and scale with Google Cloud Platform.

More from the author
Essential Cloud Infrastructure: Foundation
Intermediate
3h 17m
Oct 30, 2020
Elastic Google Cloud Infrastructure: Scaling and Automation
Intermediate
2h 58m
Oct 30, 2020
Getting Started with Google Kubernetes Engine
Intermediate
2h 2m
Oct 29, 2020
More courses by Authored 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