Convolutional Neural Networks with TensorFlow in Python

Curriculum

• Introduction to the course Free

  2 Lessons 10 Min

  In this first section, we discuss the topics covered throughout this course along with the numerous benefits of learning about Convolutional Neural Networks.

  What does the course cover? Free Why CNNs? Free
• Kernels Free

  5 Lessons 17 Min

  Here, we introduce the concept of image kernels and transformations. We will explain how they work through convolution, which forms the basis for CNNs.

  Introduction to image kernels Free How do image transformations work? Free Kernels as matrices Free Convolution - applying kernels Free Edge handling Free
• CNN Introduction

  4 Lessons 23 Min

  This is where we have an in-depth discussion of the Convolutional Neural Networks: you will understand the motivation and fundamental strength of this type of network and learn more about the concepts and layers that make it work – feature maps and pooling layers. Finally, you will discover how the dimensions change in such a network.

  CNNs motivation Feature maps Pooling and Stride Dimensions
• Neural network techniques (revision)

  3 Lessons 9 Min

  In this section, we quickly revise the main concepts of neural networks in general. These include activation functions, overfitting and early stopping, and optimizers. This part is intended only as a reference and is not a substitute for a full course on the basics of Machine Learning.

  Activation functions Overfitting and early stopping Optimizers
• Setting up the environment

  2 Lessons 2 Min

  In this part of the course, you will learn how to set up Python 3 and load up Jupyter. We’ll also show you what the Anaconda Prompt is and how you can use it to download and import new modules.

  Setting up the environment - Do not skip, please! Installing the packages
• CNN assembling - MNIST

  5 Lessons 45 Min

  In this section, you will put theory into practice. You will develop a simple CNN architecture, implement it from scratch, and train the model on the MNIST example.

  Road plan A simple CNN architecture Preprocessing the data Building and training the CNN Testing the trained CNN
• Tensorboard: Visualization tool for TensorFlow

  3 Lessons 36 Min

  After training a model to classify digits, we turn our attention to tracking and visualizing different metrics. The ability to analyze your models is crucial in Machine Learning, and here, we will lay the groundwork which makes that possible through TensorBoard.

  Tensorboard on the MNIST example Confusion matrix and visualizing it with Tensorboard Using Tensorboard to tune hyperparameters
• Common techniques for better performance of neural networks

  5 Lessons 19 Min

  This section introduces 3 crucial concepts for improving the performance of neural networks in general. We will discuss theoretically L2 regularization and weight decay, Dropout, and Data Augmentation, which you will practice in the next section.

  Introduction Regularization L2 Regularization and weight decay Dropout Data augmentation
• A practical project: Labelling fashion items

  13 Lessons 79 Min

  So far, you have already gained some practice with CNNs in a somewhat sterile environment. However, the real world is much messier than the MNIST example. That’s why, in this section, we have prepared a huge practical example inspired by tasks you are very likely to do in your job. You will practice classifying clothes and other fashion items with multiple labels. Here, you will see why the neat results from the MNIST example are almost unattainable and gain valuable insight into the procedure of undertaking such a project.

  Introduction to the problem The objective and the images Converting images to arrays Getting started with the code concepts Primary classification task - Part 1 Primary classification task - Part 2 Primary classification task - Part 3 Trousers and Jeans - discussion of approaches Trousers and Jeans - All Trousers and Jeans - Gender + Type Trousers and Jeans - comparing the methods L2 regularization and Dropout Data augmentation - Shoes All
• Understanding CNNs

  1 Lesson 7 Min

  In this section, we briefly discuss what aspects of neural networks you need to be careful with, illustrating them with real examples of models that exhibit unexpected behavior.

  Unexpected failures
• Popular CNN architectures

  5 Lessons 19 Min

  This section is dedicated to exploring the most influential and breakthrough Convolutional Neural Networks of the past - AlexNet, VGG, GoogLeNet, and ResNet - and explaining the concepts behind their architectures. We will also look at the Computer Vision competition ILSVRC and its development over the years.

  Introduction - the ILSVRC challenge AlexNet - CNN success VGG - more layers GoogleNet - computational efficiency ResNet - revolution of depth