Intro to Deep Learning With TensorFlow & Keras

Overview

The abundance of data and affordable cloud scale has led to an explosion of interest in Deep Learning. Google has open sourced a library called TensorFlow which has become the de-facto standard, allowing state-of-the-art machine learning done at scale, complete with GPU-based acceleration.

This course introduces Deep Learning concepts and TensorFlow and Keras libraries to students.

What you will learn:

• Deep Learning concepts

• TensorFlow and Keras

• Create neural networks with Tensorflow and Keras

• Learn to use tools like Tensorboard to help with training neural networks

• We will build neural networks to solve the following problems

• Regression

• Classification

• Computer vision / Image analytics

Duration:

Two Days

Audience:

Developers, Data analysts, data scientists

Skill level

Introductory to Intermediate

Industry Use Cases Covered

We will study and solve some of most common industry use cases; listed below

• Finance

• Predicting loan defaults at Prosper

• Stocks analysis (time series)

• Health care

• Predicting diabetes outcome

• Customer service

• Predicting customer turnover

• Computer vision

• Various image analysis

Prerequisites

• Basic knowledge of Python language and Jupyter notebooks is assumed.
  Even if you haven’t done any Python programming, Python is such an easy language to learn quickly. We will provide Python resources.

Lab environment

• Students would need to bring their laptop. Setup instructions will be provided few days before training

Students will need the following

• A reasonably modern laptop with unrestricted connection to the Internet. Laptops with overly restrictive VPNs or firewalls may not work properly

• Chrome browser

Detailed Course Outline

Section 1: Introduction to Deep Learning

• Understanding Deep Learning use cases

• Understanding AI / Machine Learning / Deep Learning

• Data and AI

• AI vocabulary

• Hardware and software ecosystem

• Understanding types of Machine Learning (Supervised / Unsupervised / Reinforcement)

Section 2: Introducing TensorFlow

• TensorFlow intro

• TensorFlow features

• Execution graph

• TensorFlow on GPU and TPU

• TensorFlow API

• Lab: Setting up and Running TensorFlow

Section 3: Introducing Keras

• Keras Intro

• Keras concepts (models, layers)

• Using Keras API

• Lab

Section 4: Deep Learning Concepts

• Introducing Perceptrons

• Linear Perceptrons

• Activation Functions (Sigmoid, Tanh, Relu, Softmax)

• Backpropagation

• Optimizers (Gradient Descent, Adam, RMSProp)

• Loss functions for regressions and classifications

• Vanishing/exploding gradient problem

• Lab: Tensorflow playground

Section 5: Feedforward Network

• FFNN architecture

• Input layer, output layer

• Hidden layers and Deep neural networks

• Sizing neural networks

• Lab: Feedforward Neural Networks

Section 6: Computer Vision

• Introducing Convolutional Neural Networks (CNN)

• CNN architecture

• CNN concepts

• Lab: Image recognition using CNNs

About the Instructor:

Sujee Maniyam is a seasoned practitioner and founder of Elephant Scale. He teaches and consults in AI (machine learning and deep learning) and Big Data technologies (Hadoop, Spark, NoSQL and Cloud). He is an open source contributor and author of 'Hadoop illuminated' (an open-source book on Hadoop) and 'HBase Design Patterns'. Sujee is a frequent speaker at various conferences and meetups. He also advises and mentors various firms.

If you have any questions please contact: Thomas@ElephantScale.org