Convolutional neural networks for image recognition in Keras and TensorFlow

Umberto Michelucci (TOELT LLC)

Monday, 14 Oct & Tuesday, 15 Oct,
9:00-17:00

Location: Hilton Meeting Room 3/4

Secondary topics: Computer Vision

Average rating:

(4.00, 1 rating)

Participants should plan to attend both days of this 2-day training course. To attend training courses, you must register for a Platinum or Training pass; does not include access to tutorials on Tuesday.

Convolutional neural networks (CNNs) are the basis of many algorithms that deal with images, from image recognition and classification to object detection. Using practical examples, Umberto Michelucci walks you through developing convolutional neural networks, using pretrained networks, and even teaching a network to paint. TensorFlow or Keras will be used for all examples.

What you'll learn, and how you can apply it

Understand CNNs and how to implement them in Python, TensorFlow, and Keras
Learn to develop CNNs, use pretrained networks like VGG16, and use them in art or practical problems like object detection in real time with an image or with a webcam

Who is this presentation for?

You're a software engineer or programmer with a background in Python who wants to better understand CNNs and how to use them in your own projects.

Level

Intermediate

Prerequisites:

Familiarity with Python, basic mathematics, modeling, and statistics

Hardware and/or installation requirements:

A laptop
Complete the instructions in the README file on the GitHub repository

NOTE: You will receive an introduction to the newly released TensorFlow 2.0 in this training.

Day 1

Morning

Brief introduction to neural networks, loss functions, and optimizers
Introduction to CNN, convolutions, and filters
Introduction to Google Colab, TensorFlow, and Keras 
Keras layers and sequential models
Development of first CNN with images (MNIST, black and white images of handwritten digits)

  LUNCH  

Afternoon

TensorFlow 2.0 and eager execution
Hardware acceleration with Keras and Google Colab
CNN with MNIST with TF 2.0
Development of a more complex CNN with the CIFAR-10 dataset (color images of several classes)

Day 2

Morning

Keras callback classes 
Keras functional APIs
Saving and loading models and weights
Example of saving and loading of weights and models with MNIST  
Transfer learning 
Using pretrained networks in Keras 
A practical application of transfer learning 
Development of a CNN with real images with transfer learning

LUNCH

Afternoon

Neural style transfer
An introduction 
Apply NST to your images
tf.data.Dataset for data management

About your instructor

Umberto Michelucci is a cofounder and the chief AI scientist at TOELT LLC, a company aiming to develop new and modern teaching, coaching, and research methods for AI to make AI technologies and research accessible to every company and everyone. He’s an expert in numerical simulation, statistics, data science, and machine learning. In addition to several years of research experience at the George Washington University (US) and the University of Augsburg (DE), he has 15 years of practical experience in the fields of data warehouse, data science, and machine learning. His last book, Applied Deep Learning—A Case-Based Approach to Understanding Deep Neural Networks, was published by Springer in 2018, and he’s working on a new book, Convolutional and Recurrent Neural Networks Theory and Applications. He’s very active in research in the field of artificial intelligence. He publishes his research results regularly in leading journals and gives regular talks at international conferences. Umberto studied physics and mathematics. Sharing is caring—for that, he is a lecturer at the ZHAW University of Applied Sciences for deep learning and neural networks theory and applications and at the HWZ University of Applied Science for big data analysis and statistics. At Helsana Versicherung AG, he’s also responsible for research and collaborations with universities in the area of AI.

Conference registration

Get the Platinum pass or the Training pass to add this course to your package.

Comments on this page are now closed.

Comments

Umberto Michelucci | Google Machine learning Developer Expert

29/08/2019 14:34 BST

Hi, I am the trainer. In case you have any question or would like to know more, feel free to get in touch with me at any time! Leave a comment in case you would like any clarification.