Book Image

Hands-On Transfer Learning with Python

By : Dipanjan Sarkar, Raghav Bali, Tamoghna Ghosh
Book Image

Hands-On Transfer Learning with Python

By: Dipanjan Sarkar, Raghav Bali, Tamoghna Ghosh

Overview of this book

Transfer learning is a machine learning (ML) technique where knowledge gained during training a set of problems can be used to solve other similar problems. The purpose of this book is two-fold; firstly, we focus on detailed coverage of deep learning (DL) and transfer learning, comparing and contrasting the two with easy-to-follow concepts and examples. The second area of focus is real-world examples and research problems using TensorFlow, Keras, and the Python ecosystem with hands-on examples. The book starts with the key essential concepts of ML and DL, followed by depiction and coverage of important DL architectures such as convolutional neural networks (CNNs), deep neural networks (DNNs), recurrent neural networks (RNNs), long short-term memory (LSTM), and capsule networks. Our focus then shifts to transfer learning concepts, such as model freezing, fine-tuning, pre-trained models including VGG, inception, ResNet, and how these systems perform better than DL models with practical examples. In the concluding chapters, we will focus on a multitude of real-world case studies and problems associated with areas such as computer vision, audio analysis and natural language processing (NLP). By the end of this book, you will be able to implement both DL and transfer learning principles in your own systems.
Table of Contents (14 chapters)

Building a coloring deep neural network

It's time to build a coloring deep neural network or colornet. As discussed in the previous section, if we utilize an alternate colorspace, such as LAB (or YUV), we can transform the colorization task into a mathematical transformation. The transformation is as follows:

Mathematical formulations and creativity are fine, but where are the images to learn these transformations? Deep learning networks are data hungry, but luckily, we have a huge collection of diverse images from various open source datasets. For the purposes of this chapter, we will rely on a few sample images from ImageNet itself. Since ImageNet is a huge dataset, we have randomly selected a few color images for our problem statement. In later sections, we will discuss why we selected this subset and a few of its nuances.

We relied upon the image-extraction utility...