dipanjans / hands-on-transfer-learning-with-python

Deep learning simplified by transferring prior learning using the Python deep learning ecosystem

Hands-On Transfer Learning with Python

Implement advanced deep learning and neural network models using Tensorflow and Keras

With the world moving towards digitalization and automation, as a technologist/programmer it is important to keep oneself updated and learn how to leverage these tools and techniques. "Hands-On Transfer Learning with Python", is an attempt to help practitioners get acquainted with and equipped to use these advancements in their respective domains. This book is structured broadly into three sections:

• Deep learning foundations
• Essentials of transfer learning
• Transfer learning case studies

This repository contains all the code, notebooks and examples used in this book. We will also be adding bonus content here from time to time. So keep watching this space!

Get the book

About the book

Transfer learning is a machine learning (ML) technique where knowledge gained during the training of one set of ML problems can be used to train other similar types of problems. The purpose of this book is two-fold. We focus on detailed coverage of deep learning and transfer learning, comparing and contrasting the two with easy-to-follow concepts and examples. The second area of focus will be on real-world examples and research problems using tensorflow, keras, and the Python ecosystem with hands-on examples.

The book starts with core essential concepts of ML and deep learning, followed by some depictions and coverage of important deep learning architectures, such as CNNs, DNNs, RNNs, LSTMs, and capsule networks. Our focus then shifts to transfer learning concepts and pretrained state of the art networks such as VGG, Inception, and ResNet. We also learn how these systems can be leveraged to improve performance of our deep learning models. Finally, we focus on a multitude of real-world case studies and problems in areas such as computer vision, audio analysis, and natural language processing (NLP). By the end of this book, you will be all ready to implement both deep learning and transfer learning principles in your own systems.

^{Edition: 1st   Pages: 438   Language: English
Book Title: Hands-On Transfer Learning with Python   Publisher: Packt
Copyright: Sarkar, Bali & Ghosh   ISBN 13: 9781788831307}

Contents

• Part I: Deep learning foundations
  • Chapter 1: Machine Learning Fundamentals Basics
  • Chapter 2: Deep Learning Essentials Basics
  • Chapter 3: Understanding Deep Learning Architectures Basics
• Part II: Essentials of transfer learning
  • Chapter 4: Transfer Learning Fundamentals Basics
  • Chapter 5: Unleashing the Power of Transfer Learning
• Part III: Transfer learning case studies
  • Chapter 6: Image Recognition and Classification
  • Chapter 7: Text Document Categorization
  • Chapter 8: Audio Identification and Classification
  • Chapter 9: Deep Dream
  • Chapter 10: Style Transfer
  • Chapter 11: Automated Image Caption Generator
  • Chapter 12: Image Colorization

Key Features:

• Build deep learning models with transfer learning principles in Python
• Implement transfer learning to solve real-world research problems
• Perform complex operations such as image captioning neural style transfer

What You Will Learn:

• Set up your own DL environment with graphics processing unit (GPU) and Cloud support
• Delve into transfer learning principles with ML and DL models
• Explore various DL architectures, including CNN, LSTM, and capsule networks
• Learn about data and network representation and loss functions
• Get to grips with models and strategies in transfer learning
• Walk through potential challenges in building complex transfer learning models from scratch
• Explore real-world research problems related to computer vision and audio analysis
• Understand how transfer learning can be leveraged in NLP

Audience

Hands-On Transfer Learning with Python is for data scientists, ML engineers, analysts, and developers with an interest in data and applying state-of-the-art transfer learning methodologies to solve tough real-world problems. Basic proficiency in ML and Python is required.

Acknowledgements

TBA

hands-on-transfer-learning-with-python's Issues

training data where does it live?

I'm confused in your blog post and here, you talk about training data and getting it from the data page. Where does the zip file actually live? Sorry to be so dense :-)

get_metrics

get_metrics - I had assumed that it would be in the missing cnn_utils - but now that file has been added and get_metrics not to be found.

DId a quick GitHub search - too many get_metrics - looked at first 15 pages of listings and did not see that tensorflow or keras had this.

I assume this line was the source for the 85% and 0.85 F1 score in the book.

memory error while training model

Transfer Learning for a Regression Problem

Hi Dipanjan,
Thank you for this great book about Transfer Learning. It gave me a good overview understanding of transfer Learning. I am currently working on a regression problem (7 predictors and 1 response variable) to model some restrictor parameters in a flow distribution system. The flow system is segmented into 7 zones, some of which are simple and share some similarities, but some of the zones are complex and distinct. It was suggested to use transfer Learning to model at least the zones which are similar to each other, as available data is limited. However in my research I find transfer Learning is mostly applied to classification tasks. My basic questions are:
1.) Is it possible to use transfer Learning for a regression problem where we have to predict a numerical value? Will there be any reasonable benefits in sparse data situations?
2.) Can you please point to some reliable resources (or chapters in your book) which could help in applying transfer Learning to my regression problem?

No "display_confusion_matrix_pretty" in model_evaluation_utils.py

As the tittle, just "display_confusion_matrix" there. Anybody know the solutions?

Chapter 6 - where is cnn_utils

Trying to follow along with the book - had an interest in CNN so I started with the CIFAR-10.
This code finds no library

• import cnn_utils as utils. It's used at the end of the cifar10 transfer learning based classifier. I cannot find it in my anaconda distribution - cant find it searching anaconda to add, and does not seem to be in the git for the book.
  Where is it?

I want to create my own pre trained model, is there any guidance for it in this book?

I know how to create a cnn model and test it. By using that i am getting 73% accuracy on a multi time series data. i have some other similar data, i want to train my model on one data and then apply transfer learning . Is there any guidance in this book related this. ?
I dont have pretrained model available

Ch05 - Unleash the Power of Transfer Learning / Basic CNN model

Hi,
I used this book's code to learn in Ch05, Basic CNN Model.ipynb, after running these codes, I found my validate_acc always 50%, and it didn't change any more.
I guess the following code may have to modify:
"
model.compile(loss='binary_crossentropy',
optimizer=optimizers.RMSprop(),
metrics=['accuracy'])
"
Another optimizer could help to get a better accuracy?
I am a starter.