Master Deep Learning with TensorFlow 2.0 in Python [2019]

Video description

Build deep learning algorithms with TensorFlow 2.0, dive into neural networks, and apply your skills in a business case.

About This Video

  • Build deep learning algorithms from scratch in Python using NumPy and TensorFlow
  • Get hands-on with deep learning and machine learning
  • Understand the math behind deep learning algorithms

In Detail

Data scientists, machine learning engineers, and AI researchers all have their own skillsets. However, there is a quality they all have in common. They are all masters of deep learning.

We often hear about artificial intelligence, self-driving cars, and algorithmic magic at Google, Facebook, and Amazon. All these are related to deep learning. And more specifically, it is usually deep neural networks, the single algorithm that is responsible for them all.

In this course, you’ll gain useful insights into deep learning. You’ll start with the basics and then progress toward building a deep learning algorithm. The course will help you learn easily as it programs everything in Python and explains each line of code clearly. All this will help you move on to the more complex topics easily.

You'll get familiar with TensorFlow and NumPy, two tools that are essential for creating and understanding deep learning algorithms. You'll also explore layers, along with their building blocks and activations – sigmoid, tanh, ReLU, Softmax, and more.

As you progress, you’ll understand the backpropagation process. You'll be able to spot and prevent overfitting, one of the biggest issues in machine and deep learning. The course will then guide you through state-of-the-art initialization methods. Later, you'll learn how to build deep neural networks using real data, implemented by companies in the real world, along with templates.

By the end of this course, you will have developed the skills you need to advance in your data science career and confidently build deep learning algorithms.

Downloading the example code for this course: You can download the example code files for this course on GitHub at the following link: https://github.com/PacktPublishing/Master-Deep-Learning-with-TensorFlow-2.0-in-Python-2019. If you require support please email: customercare@packt.com

Table of contents

  1. Chapter 1 : Welcome! Course introduction
    1. Meet your instructors and why you should study machine learning? 00:06:55
    2. What does the course cover? 00:03:08
  2. Chapter 2 : Introduction to neural networks
    1. Introduction to neural networks 00:04:09
    2. Training the model 00:02:55
    3. Types of machine learning 00:03:43
    4. The linear model 00:03:09
    5. The linear model. Multiple inputs 00:02:25
    6. The linear model. Multiple inputs and multiple outputs 00:04:23
    7. Graphical representation 00:01:47
    8. The objective function 00:01:27
    9. L2-norm loss 00:02:04
    10. Cross-entropy loss 00:03:56
    11. One parameter gradient descent 00:06:34
    12. N-parameter gradient descent 00:06:09
  3. Chapter 3 : Setting up the working environment
    1. Setting up the environment - An introduction - Do not skip, please! 00:00:44
    2. Why Python and why Jupyter? 00:04:54
    3. Installing Anaconda 00:03:03
    4. The Jupyter dashboard - part 1 00:02:27
    5. The Jupyter dashboard - part 2 00:05:14
    6. Installing TensorFlow 2 00:05:31
  4. Chapter 4 : Minimal example - your first machine learning algorithm
    1. Minimal example - part 1 00:03:07
    2. Minimal example - part 2 00:04:59
    3. Minimal example - part 3 00:03:26
    4. Minimal example - part 4 00:07:47
  5. Chapter 5 : TensorFlow - An introduction
    1. TensorFlow outline 00:03:28
    2. TensorFlow 2 intro 00:02:33
    3. A Note on Coding in TensorFlow 00:00:58
    4. Types of file formats in TensorFlow and data handling 00:02:35
    5. Model layout - inputs, outputs, targets, weights, biases, optimizer and loss 00:05:48
    6. Interpreting the result and extracting the weights and bias 00:04:10
    7. Customizing your model 00:02:52
  6. Chapter 6 : Going deeper: Introduction to deep neural networks
    1. Layers 00:01:53
    2. What is a deep net? 00:02:18
    3. Understanding deep nets in depth 00:04:58
    4. Why do we need non-linearities? 00:03:00
    5. Activation functions 00:03:37
    6. Softmax activation 00:03:24
    7. Backpropagation 00:03:12
    8. Backpropagation - visual representation 00:03:02
  7. Chapter 7 : Overfitting
    1. Underfitting and overfitting 00:03:51
    2. Underfitting and overfitting - classification 00:01:52
    3. Training and validation 00:03:23
    4. Training, validation, and test 00:02:31
    5. N-fold cross validation 00:03:07
    6. Early stopping 00:04:55
  8. Chapter 8 : Initialization
    1. Initialization - Introduction 00:02:32
    2. Types of simple initializations 00:02:47
    3. Xavier initialization 00:02:46
  9. Chapter 9 : Gradient descent and learning rates
    1. Stochastic gradient descent 00:03:25
    2. Gradient descent pitfalls 00:02:02
    3. Momentum 00:02:30
    4. Learning rate schedules 00:04:26
    5. Learning rate schedules. A picture 00:01:33
    6. Adaptive learning rate schedules 00:04:08
    7. Adaptive moment estimation 00:02:39
  10. Chapter 10 : Preprocessing
    1. Preprocessing introduction 00:02:52
    2. Basic preprocessing 00:01:18
    3. Standardization 00:04:31
    4. Dealing with categorical data 00:02:15
    5. One-hot and binary encoding 00:03:39
  11. Chapter 11 : The MNIST example
    1. The dataset 00:02:25
    2. How to tackle the MNIST 00:02:44
    3. Importing the relevant packages and load the data 00:02:11
    4. Preprocess the data - create a validation dataset and scale the data 00:04:44
    5. Preprocess the data - shuffle and batch the data 00:06:30
    6. Outline the model 00:04:54
    7. Select the loss and the optimizer 00:02:05
    8. Learning 00:05:38
    9. Testing the model 00:03:56
  12. Chapter 12 : Business case
    1. Exploring the dataset and identifying predictors 00:07:54
    2. Outlining the business case solution 00:01:32
    3. Balancing the dataset 00:03:39
    4. Preprocessing the data 00:11:32
    5. Load the preprocessed data 00:03:24
    6. Learning and interpreting the result 00:04:15
    7. Setting an early stopping mechanism 00:05:02
    8. Testing the model 00:01:24
  13. Chapter 13 : Conclusion
    1. See how much you have learned 00:03:41
    2. What's further out there in the machine and deep learning world 00:01:47
    3. An overview of CNNs 00:04:56
    4. An overview of RNNs 00:02:50
    5. An overview of non-NN approaches 00:03:53

Product information

  • Title: Master Deep Learning with TensorFlow 2.0 in Python [2019]
  • Author(s): 365 Careers
  • Release date: July 2019
  • Publisher(s): Packt Publishing
  • ISBN: 9781839218163