book

The Deep Learning with PyTorch Workshop

Name: The Deep Learning with PyTorch Workshop
ISBN: 9781838989217

by Hyatt Saleh, Tim Hoolihan, Learnkart Technology Private Limited, Anuj Shah, Nahar Singh, Subhash Sundaravadivelu

July 2020

Intermediate to advanced

330 pages

6h 5m

English

Packt Publishing

Read now

Unlock full access

The Deep Learning with PyTorch Workshop
Preface
About the BookAudienceAbout the ChaptersConventionsCode PresentationHardware RequirementsSoftware Requirements Setting up Your EnvironmentInstalling Python on Windows and macOSInstalling Python on LinuxInstalling pipInstalling PyTorch Installing LibrariesOpening a Jupyter NotebookAccessing the Code Files
1. Introduction to Deep Learning and PyTorch
IntroductionWhy Deep Learning?Applications of Deep LearningIntroduction to PyTorch GPUs in PyTorchWhat Are Tensors?Exercise 1.01: Creating Tensors of Different Ranks Using PyTorchAdvantages of Using PyTorchDisadvantages of Using PyTorchKey Elements of PyTorchThe PyTorch autograd LibraryThe PyTorch nn ModuleExercise 1.02: Defining a Single-Layer ArchitectureThe PyTorch optim PackageExercise 1.03: Training a Neural NetworkActivity 1.01: Creating a Single-Layer Neural NetworkSummary
2. Building Blocks of Neural Networks
IntroductionIntroduction to Neural NetworksWhat Are Neural Networks?Exercise 2.01: Performing the Calculations of a PerceptronMulti-Layer PerceptronThe Learning Process of a Neural NetworkForward PropagationThe Calculation of Loss FunctionsBackward PropagationGradient DescentAdvantages and DisadvantagesAdvantagesDisadvantagesIntroduction to Artificial Neural NetworksIntroduction to Convolutional Neural NetworksIntroduction to Recurrent Neural NetworksData PreparationDealing with Messy DataExercise 2.02: Dealing with Messy DataData RescalingExercise 2.03: Rescaling DataSplitting the DataExercise 2.04: Splitting a DatasetDisadvantages of Failing to Prepare Your DataActivity 2.01: Performing Data PreparationBuilding a Deep Neural NetworkExercise 2.05: Building a Deep Neural Network Using PyTorchActivity 2.02: Developing a Deep Learning Solution for a Regression ProblemSummary
3. A Classification Problem Using DNN
IntroductionProblem DefinitionDeep Learning in BankingExploring the DatasetData PreparationBuilding the ModelANNs for Classification TasksA Good ArchitecturePyTorch Custom ModulesExercise 3.01: Defining a Model's Architecture Using Custom ModulesDefining the Loss Function and Training the ModelActivity 3.01: Building an ANNDealing with an Underfitted or Overfitted ModelError AnalysisExercise 3.02: Performing Error AnalysisActivity 3.02: Improving a Model's PerformanceDeploying Your ModelSaving and Loading Your ModelPyTorch for Production in C++Building an APIExercise 3.03: Creating a Web APIActivity 3.03: Making Use of Your ModelSummary
4. Convolutional Neural Networks
IntroductionBuilding a CNNWhy Are CNNs Used for Image Processing?The Image as InputApplications of CNNsClassificationLocalizationDetectionSegmentationThe Building Blocks of CNNsConvolutional LayersExercise 4.01: Calculating the Output Shape of a Convolutional LayerPooling LayersExercise 4.02: Calculating the Output Shape of a Set of Convolutional and Pooling LayersFully Connected LayersSide Note – Downloading Datasets from PyTorchActivity 4.01: Building a CNN for an Image Classification ProblemData AugmentationData Augmentation with PyTorchActivity 4.02: Implementing Data AugmentationBatch NormalizationBatch Normalization with PyTorchActivity 4.03: Implementing Batch NormalizationSummary
5. Style Transfer
IntroductionStyle TransferHow Does It Work?Implementation of Style Transfer Using the VGG-19 Network ArchitectureInputs – Loading and DisplayingExercise 5.01: Loading and Displaying ImagesLoading the ModelExercise 5.02: Loading a Pre-Trained Model in PyTorchExtracting the FeaturesExercise 5.03: Setting Up the Feature Extraction ProcessThe Optimization Algorithm, Losses, and Parameter UpdateContent LossStyle LossTotal LossExercise 5.04: Creating the Target ImageActivity 5.01: Performing Style TransferSummary
6. Analyzing the Sequence of Data with RNNs
IntroductionRecurrent Neural NetworksApplications of RNNsHow Do RNNs Work?Input and Targets for Sequenced DataExercise 6.01: Creating the Input and Target Variables for a Sequenced Data ProblemRNNs in PyTorchActivity 6.01: Using a Simple RNN for a Time Series PredictionLong Short-Term Memory NetworksApplications of LSTM NetworksHow Do LSTM Networks Work?LSTM Networks in PyTorchPreprocessing the Input DataNumbered LabelsGenerating the BatchesOne-Hot EncodingExercise 6.02: Preprocessing the Input Data and Creating a One-Hot MatrixBuilding the ArchitectureTraining the ModelPerforming PredictionsActivity 6.02: Text Generation with LSTM NetworksNatural Language ProcessingSentiment AnalysisSentiment Analysis in PyTorchPreprocessing the Input DataBuilding the ArchitectureTraining the ModelActivity 6.03: Performing NLP for Sentiment AnalysisSummary
Appendix
1. Introduction to Deep Learning and PyTorchActivity 1.01: Creating a Single-Layer Neural NetworkSolution2. Building Blocks of Neural NetworksActivity 2.01: Performing Data PreparationSolutionActivity 2.02: Developing a Deep Learning Solution for a Regression ProblemSolution3. A Classification Problem Using DNNsActivity 3.01: Building an ANNActivity 3.02: Improving a Model's PerformanceActivity 3.03: Making Use of Your ModelSolution4. Convolutional Neural NetworksActivity 4.01: Building a CNN for an Image Classification ProblemSolutionActivity 4.02: Implementing Data AugmentationSolutionActivity 4.03: Implementing Batch NormalizationSolution5. Style TransferActivity 5.01: Performing Style TransferSolution6. Analyzing the Sequence of Data with RNNsActivity 6.01: Using a Simple RNN for a Time Series PredictionSolutionActivity 6.02: Text Generation with LSTM NetworksSolutionActivity 6.03: Performing NLP for Sentiment AnalysisSolution

Content preview from The Deep Learning with PyTorch Workshop

Appendix

1. Introduction to Deep Learning and PyTorch

Activity 1.01: Creating a Single-Layer Neural Network

Solution

Import the required libraries, including pandas, for importing a CSV file:
import pandas as pd
import torch
import torch.nn as nn
import matplotlib.pyplot as plt
Read the CSV file containing the dataset:
data = pd.read_csv("SomervilleHappinessSurvey2015.csv")
Separate the input features from the target. Note that the target is located in the first column of the CSV file. Convert the values into tensors, making sure the values are converted into floats:
x = torch.tensor(data.iloc[:,1:].values).float()
y = torch.tensor(data.iloc[:,:1].values).float()
Define the architecture of the model and store it in a variable named model. Remember ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9781838989217

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

The Deep Learning with PyTorch Workshop

by Hyatt Saleh, Tim Hoolihan, Learnkart Technology Private Limited, Anuj Shah, Nahar Singh, Subhash Sundaravadivelu

Appendix

1. Introduction to Deep Learning and PyTorch