**Guaranteed to Run**

Course Objectives

Leading corporations such as Google, Amazon, and Facebook are betting their futures on AI. In this 2-day workshop, you will learn everything from major players in the AI ecosystem, neural network basics to deploying Python models into a production environment.

Course Prerequisites

Attendees should be familiar with developing in Java or Python. No prior knowledge of Deeplearning4j is required.

Course Outline

Day 1: Introduction to Deep Learning Theory and Ecosystem Players

• Introduction to Deep Learning and Neural Networks

• Introduction to Tensorflow and Theano

• Introduction to Keras

• Demonstration of Neural Network

• NumPY/SciKit Learn basics

• Building a basic Neural Network

• Neural Network internals

• Activation Functions

• Backpropagation

• Loss Functions

• Weight Initialization

• Data Normalization/Standardization

Day 2: Model Tuning, Deployment, and Scaling

• Deep Network Topologies

• Feed Forward

• Convolutional

• Recurrent

• Tuning

• Overfitting

• Learning Rate

• Adaptive Learning Rates

• Dropout

• Regularization

• Advanced topics

• Import Keras into Deeplearning4j for production

• Model Import

• Transfer Learning

• Model Serializer

• Deploying onto the public cloud (AWS, Azure)

• Overview of the ecosystem

• Distributed Training on Spark

• Integration with Hadoop cluster

Why This Matters?

Importing and deploying prototype models to production is important for two reasons:

1. Students and researchers often rely on Python to experiment with deep learning models. As they migrate to large corporations, they bring their existing tools with them, creating friction between data scientists (Python) and data engineers (Java).

2. Many corporations don’t have enough data to train neural networks. Because of this, they are unable to take advantage of deep learning in their businesses.

Deeplearning4j bridges these gaps by allowing data scientists to import pre-trained Python models into a production environment (which use Java) via Keras.

Who Should Attend?

This course is designed for data scientists and data engineers who need to interact with big data infrastructure.

What You Will Achieve

At the conclusion of this workshop, you will acquire:

1. All the knowledge necessary to succeed in corporate data science environments.
2. A copy of our Deep Learning Textbook.
3. A certificate of completion.

Instructor Bio

Tom Hanlon currently leads training and workshops at Skymind.

He has been an instructor in groundbreaking technologies since 2001. When the LAMP revolution transformed web development, Tom was on the front lines teaching MySQL. When the big data revolution transformed the way data was handled and stored, Tom was at Cloudera and Hortonworks teaching Hadoop to Fortune 500 data engineering teams. He has trained teams at Blizzard, Bloomberg and more.

About Skymind

Skymind is the company behind Deeplearning4j, the only commercial-grade, open-source, distributed deep-learning library written for Java and Scala. Integrated with Hadoop and Spark, DL4J is specifically designed to run in business environments on distributed GPUs and CPUs.