Home Page Icon
Home Page
Table of Contents for
IBM PowerAI: Deep Learning Unleashed on IBM Power Systems Servers
Close
IBM PowerAI: Deep Learning Unleashed on IBM Power Systems Servers
by Richard Wale, Shota Tsukamoto, Chris Parsons, Alfonso Jara, Bruno C. Faria, Bing
IBM PowerAI: Deep Learning Unleashed on IBM Power Systems Servers
Front cover
Notices
Trademarks
Preface
Authors
Now you can become a published author, too!
Comments welcome
Stay connected to IBM Redbooks
Chapter 1. Introduction to artificial intelligence and deep learning
1.1 Deep learning
1.1.1 Artificial intelligence milestones and the development of deep learning
1.2 Neural networks overview
1.2.1 A brief history about neural networks
1.2.2 Why neural networks are an important subject
1.2.3 Types of neural networks and their usage
1.2.4 Neural network architectures
1.2.5 Difference between a classical and deep neural networks
1.2.6 Neural networks versus classical machine learning algorithms
1.3 Deep learning frameworks
1.3.1 Most popular deep learning frameworks
1.3.2 A final word on deep learning frameworks
Chapter 2. Introduction and overview of IBM PowerAI
2.1 What is IBM PowerAI
2.1.1 Contents of IBM PowerAI (IBM PowerAI Release 4)
2.1.2 Minimum hardware requirement for IBM PowerAI
2.2 Why IBM PowerAI simplifies adoption of deep learning
2.3 IBM unique capabilities
2.3.1 NVLink and NVLink 2.0
2.3.2 Power AI Distributed Deep Learning
2.3.3 Large Model Support
2.4 Extra integrations that are available for IBM PowerAI
2.4.1 IBM Data Science Experience
2.4.2 IBM PowerAI Vision (technology preview)
2.4.3 IBM Spectrum Conductor Deep Learning Impact
Chapter 3. IBM PowerAI components
3.1 IBM PowerAI components
3.1.1 IBM PowerAI support and extra services from IBM
3.1.2 IBM Power Systems for deep learning
3.1.3 Linux on Power for deep learning
3.1.4 NVIDIA GPUs
3.1.5 NVIDIA components
3.1.6 NVIDIA drivers
3.1.7 IBM PowerAI deep learning package
3.1.8 Libraries
3.1.9 Frameworks
3.1.10 Other software and functions
3.2 IBM PowerAI compatibility matrix
Chapter 4. Deploying IBM PowerAI
4.1 IBM PowerAI V1.4 setup guide
4.1.1 About this chapter
4.1.2 Preparing to install IBM PowerAI V1.4
4.1.3 IBM Power System S822LC for High Performance Computing initial setup
4.1.4 Installing Ubuntu 16.04.x
4.1.5 Installing IBM PowerAI V1.4
4.2 Testing IBM PowerAI V1.4
4.2.1 First test: TensorFlow test program
4.2.2 Utilization of a multilayer perceptron on a sample data set
4.2.3 Using Caffe with MNIST
4.2.4 Using Caffe with TensorFlow
4.3 Setting up IBM PowerAI V1.5.0 on a POWER L822SC for High Performance Computing server
4.3.1 Deep learning software packages
4.3.2 System setup
4.3.3 Installing the deep learning frameworks
4.3.4 Tuning recommendations
4.3.5 Getting started with machine learning and deep learning frameworks
4.3.6 Uninstalling machine learning and deep learning frameworks
4.4 IBM PowerAI V1.5.0 setup guide for POWER AC922
4.4.1 Deep learning software packages
4.4.2 System setup
4.4.3 Installing the deep learning frameworks
4.4.4 Tuning recommendations
4.4.5 Getting started with machine learning and deep learning frameworks
4.4.6 Uninstalling machine learning and deep learning frameworks
Chapter 5. Working with data and creating models in IBM PowerAI
5.1 Knowing your requirements and data
5.2 Why is it so important to prepare your data
5.3 Sentiment analysis by using TensorFlow on IBM PowerAI
5.3.1 Example data set
5.3.2 How the code is structured
5.3.3 Data preparation
5.3.4 Model creation
5.3.5 Using the model
5.3.6 Running the code
5.4 Word suggestions by using long and short term memory on TensorFlow
5.4.1 Our data set
5.4.2 Overall structure of the code
5.4.3 Data preparation
5.4.4 Model creation
5.4.5 Training
5.4.6 Using the model
5.4.7 Running the code
5.4.8 Final considerations
Chapter 6. Introduction to IBM Spectrum Conductor Deep Learning Impact
6.1 Definitions, acronyms, buzzwords, and abbreviations
6.2 Benefits of IBM Spectrum Conductor Deep Learning Impact
6.3 Key features of Deep Learning Impact
6.3.1 Parallel data set processing
6.3.2 Monitoring and Optimization for one training model
6.3.3 Hyperparameter optimization and search
6.3.4 IBM Fabric for distributed training
6.3.5 IBM Fabric and auto-scaling
6.3.6 DLI inference model
6.3.7 Supporting a shared multi-tenant infrastructure
6.4 DLI deployment
6.4.1 Deployment consideration
6.4.2 DLI single-node mode
6.4.3 DLI cluster without a high availability function
6.4.4 DLI cluster with a high availability function
6.4.5 Binary files installation for the high availability enabled cluster
6.4.6 A DLI cluster with a high availability function installation guide
6.5 Master node crashed when a workload is running
6.6 Introduction to DLI graphic user interface
6.6.1 Data set management
6.6.2 Model management
6.6.3 Deep learning activity monitor and debug management
6.7 Supported deep learning network and training engine in DLI
6.7.1 Deep learning network samples
6.7.2 Integrating with a customer’s network in DLI
6.8 Use case: Using a Caffe Cifar-10 network with DLI
6.8.1 Data preparation
6.8.2 Data set import
6.8.3 Model creation
6.8.4 Model training
6.8.5 Model validation
6.8.6 Model tuning
6.8.7 Model prediction
6.8.8 Training model weight file management
Chapter 7. Case scenarios: Using IBM PowerAI
7.1 Use case one: Bare metal environment
7.1.1 Customer requirements
7.1.2 IBM solution
7.1.3 Benefits
7.2 Use case two: Multitenant environment
7.2.1 Customer requirements
7.2.2 IBM solution
7.2.3 Benefits
7.3 Use case three: High-performance computing environment
7.3.1 Customer requirements
7.3.2 IBM solution
7.3.3 Benefits
7.4 Conclusion
Appendix A. Sentiment analysis code
Sentiment analysis with TensorFlow
How the code is organized
Sentiment analysis code
Model and training
Using the model
Appendix B. Problem determination tools
Logs and configuration data gathering tools
Troubleshooting pointers for Linux on Power
Solving a RAID failure
Related publications
IBM Redbooks
Online resources
Help from IBM
Back cover
Search in book...
Toggle Font Controls
Playlists
Add To
Create new playlist
Name your new playlist
Playlist description (optional)
Cancel
Create playlist
Sign In
Email address
Password
Forgot Password?
Create account
Login
or
Continue with Facebook
Continue with Google
Sign Up
Full Name
Email address
Confirm Email Address
Password
Login
Create account
or
Continue with Facebook
Continue with Google
Prev
Previous Chapter
Related publications
Next
Next Chapter
Back cover
IBM PowerAI: Deep Learning Unleashed on IBM Power Systems Servers
Add Highlight
No Comment
..................Content has been hidden....................
You can't read the all page of ebook, please click
here
login for view all page.
Day Mode
Cloud Mode
Night Mode
Reset