Hands-on Deep Learning Workshop - Tensorflow

Hands-on Data Sciences Workshop

Advanced Deep Learning with Python, Keras & Tensorflow

“Improve your deep learning skills building complex models with Python, Keras and Tensorflow”

Curriculum

The workshop is meant to expand your skills in deep learning by exposing you to the Tensorflow and to real world case studies. You will learn:

Ø  Review of fundamental deep learning architectures (Fully Connected, Convolutional, Recurrent)

Ø  Build and train a model with pure Tensorflow

Ø  Custom architectures and loss functions 

Ø  Setting up a machine for deep learning / serving a model

Activities

The workshop is conceived to maximize the learning experience for everyone and includes 30% theory and 70% hands-on practice.

Are there any prerequisites?

Previous experience programming in Python or in other languages is advised to make best use of the workshop. Additionally, familiarity with machine learning and deep learning is necessary.

Why python?

In the last 2 years Python has become a de-facto standard in data science and is widely adopted by most major companies. Reasons for this success include:

Ø  large set of mature data science libraries => most needs covered

Ø  worldwide community of enthusiasts => get help when you need it

Ø  easy to learn, read and write => start contributing immediately

Ø  supports both functional and object oriented coding => versatile and powerful

Ø  full stack programming language => easier interaction between data scientists and software engineers

Why Tensorflow?

There are many open source Deep Learning libraries. Tensorflow is backed by Google and is quickly becoming one of the most used libraries in the fields. It has a large and growing community of users and it is versatile and easy to learn. Highlights include

Ø  largest community of developers

Ø  state of the art models and nodes

Ø  high scalability, can be distributed on many GPUs

Ø  production performance and deployment tools

Ø  very versatile and powerful for distributed high performance computing beyond neural networks

·       

Agenda

v What Is Machine Learning ?

v Limitations of Machine Learning

v Deep Learning To The Rescue

v What Is Deep Learning ?

v Deep Learning Applications

v Difference Between Machine Learning and Deep Learning

v What is TensorFlow?

v TensorFlow Data Structures

v TensorFlow Use-Case

v How Deep Learning Works?

v Single Layer Perceptron (Early Deep Learning Models)

v Single Layer Perceptron Examples

v Limitations Of Single Layer Perceptron

v Multi Layer Perceptron

v Multi Layer Perceptron Examples

v Demo – Keras

v Why Not /Feedforward Networks?

v What Is Recurrent Neural Network?

v Issues With Recurrent Neural Networks

v Vanishing And Exploding Gradient

v How To Overcome These Challenges?

v Long Short Term Memory Units

v LSTM Use-Case