book

Machine Learning Bookcamp

Name: Machine Learning Bookcamp
Author: Alexey Grigoriev
ISBN: 9781617296819

by Alexey Grigoriev

October 2021

Intermediate to advanced

472 pages

11h 32m

English

Manning Publications

Read now

Unlock full access

inside front cover
Machine Learning Bookcamp
Copyright
brief contents
contents
front matter
forewordprefaceacknowledgmentsabout this bookWho should read this bookHow this book is organized: a roadmapAbout the codeliveBook discussion forumOther online resourcesabout the authorabout the cover illustration
1 Introduction to machine learning
1.1 Machine learning1.1.1 Machine learning vs. rule-based systems1.1.2 When machine learning isn’t helpful1.1.3 Supervised machine learning1.2 Machine learning process1.2.1 Business understanding1.2.2 Data understanding1.2.3 Data preparation1.2.4 Modeling1.2.5 Evaluation1.2.6 Deployment1.2.7 Iterate1.3 Modeling and model validationSummary
2 Machine learning for regression
2.1 Car-price prediction project2.1.1 Downloading the dataset2.2 Exploratory data analysis2.2.1 Exploratory data analysis toolbox2.2.2 Reading and preparing data2.2.3 Target variable analysis2.2.4 Checking for missing values2.2.5 Validation framework2.3 Machine learning for regression2.3.1 Linear regression2.3.2 Training linear regression model2.4 Predicting the price2.4.1 Baseline solution2.4.2 RMSE: Evaluating model quality2.4.3 Validating the model2.4.4 Simple feature engineering2.4.5 Handling categorical variables2.4.6 Regularization2.4.7 Using the model2.5 Next steps2.5.1 Exercises2.5.2 Other projectsSummaryAnswers to exercises
3 Machine learning for classification
3.1 Churn prediction project3.1.1 Telco churn dataset3.1.2 Initial data preparation3.1.3 Exploratory data analysis3.1.4 Feature importance3.2 Feature engineering3.2.1 One-hot encoding for categorical variables3.3 Machine learning for classification3.3.1 Logistic regression3.3.2 Training logistic regression3.3.3 Model interpretation3.3.4 Using the model3.4 Next steps3.4.1 Exercises3.4.2 Other projectsSummaryAnswers to exercises
4 Evaluation metrics for classification
4.1 Evaluation metrics4.1.1 Classification accuracy4.1.2 Dummy baseline4.2 Confusion table4.2.1 Introduction to the confusion table4.2.2 Calculating the confusion table with NumPy4.2.3 Precision and recall4.3 ROC curve and AUC score4.3.1 True positive rate and false positive rate4.3.2 Evaluating a model at multiple thresholds4.3.3 Random baseline model4.3.4 The ideal model4.3.5 ROC Curve4.3.6 Area under the ROC curve (AUC)4.4 Parameter tuning4.4.1 K-fold cross-validation4.4.2 Finding best parameters4.5 Next steps4.5.1 Exercises4.5.2 Other projectsSummaryAnswers to exercises

5 Deploying machine learning models
5.1 Churn-prediction model5.1.1 Using the model5.1.2 Using Pickle to save and load the model5.2 Model serving5.2.1 Web services5.2.2 Flask5.2.3 Serving churn model with Flask5.3 Managing dependencies5.3.1 Pipenv5.3.2 Docker5.4 Deployment5.4.1 AWS Elastic Beanstalk5.5 Next steps5.5.1 Exercises5.5.2 Other projectsSummary
6 Decision trees and ensemble learning
6.1 Credit risk scoring project6.1.1 Credit scoring dataset6.1.2 Data cleaning6.1.3 Dataset preparation6.2 Decision trees6.2.1 Decision tree classifier6.2.2Decision tree learning algorithm6.2.3 Parameter tuning for decision tree6.3 Random forest6.3.1 Training a random forest6.3.2 Parameter tuning for random forest6.4 Gradient boosting6.4.1 XGBoost: Extreme gradient boosting6.4.2 Model performance monitoring6.4.3 Parameter tuning for XGBoost6.4.4 Testing the final model6.5 Next steps6.5.1 Exercises6.5.2 Other projectsSummaryAnswers to exercises
7 Neural networks and deep learning
7.1 Fashion classification7.1.1 GPU vs. CPU7.1.2 Downloading the clothing dataset7.1.3 TensorFlow and Keras7.1.4 images7.2 Convolutional neural networks7.2.1 Using a pretrained model7.2.2 Getting predictions7.3 Internals of the model7.3.1 Convolutional layers7.3.2 Dense layers7.4 Training the model7.4.1 Transfer learning7.4.2 Loading the data7.4.3 Creating the model7.4.4 Training the model7.4.5 Adjusting the learning rate7.4.6 Saving the model and checkpointing7.4.7 Adding more layers7.4.8 Regularization and dropout7.4.9 Data augmentation7.4.10 Training a larger model7.5 Using the model7.5.1 Loading the model7.5.2 Evaluating the model7.5.3 Getting the predictions7.6 Next steps7.6.1 Exercises7.6.2 Other projectsSummaryAnswers to exercises
8 Serverless deep learning
8.1 Serverless: AWS Lambda8.1.1 TensorFlow Lite8.1.2 Converting the model to TF Lite format8.1.3 Preparing the images8.1.4 Using the TensorFlow Lite model8.1.5 Code for the lambda function8.1.6 Preparing the Docker image8.1.7 Pushing the image to AWS ECR8.1.8 Creating the lambda function8.1.9 Creating the API Gateway8.2 Next steps8.2.1 Exercises8.2.2 Other projectsSummary
9 Serving models with Kubernetes and Kubeflow
9.1 Kubernetes and Kubeflow9.2 Serving models with TensorFlow Serving9.2.1 Overview of the serving architecture9.2.2 The saved_model format9.2.3 Running TensorFlow Serving locally9.2.4 Invoking the TF Serving model from Jupyter9.2.5 Creating the Gateway service9.3 Model deployment with Kubernetes9.3.1 Introduction to Kubernetes9.3.2 Creating a Kubernetes cluster on AWS9.3.3 Preparing the Docker images9.3.4 Deploying to Kubernetes9.3.5 Testing the service9.4 Model deployment with Kubeflow9.4.1 the model: Uploading it to S39.4.2 Deploying TensorFlow models with KFServing9.4.3 Accessing the model9.4.4 KFServing transformers9.4.5 Testing the transformer9.4.6 Deleting the EKS cluster9.5 Next steps9.5.1 Exercises9.5.2 Other projectsSummary
Appendix A. Preparing the environment
A.1 Installing Python and AnacondaA.1.1 Installing Python and Anaconda on LinuxA.1.2 Installing Python and Anaconda on WindowsA.1.3 Installing Python and Anaconda on macOSA.2 Running JupyterA.2.1 Running Jupyter on LinuxA.2.2 Running Jupyter on WindowsA.2.3 Running Jupyter on MacOSA.3 Installing the Kaggle CLIA.4 Accessing the source codeA.5 Installing DockerA.5.1 Installing Docker on LinuxA.5.2 Installing Docker on WindowsA.5.3 Installing Docker on MacOSA.6 Renting a server on AWSA.6.1 Registering on AWSA.6.2 Accessing billing informationA.6.3 Creating an EC2 instanceA.6.4 Connecting to the instanceA.6.5 Shutting down the instanceA.6.6 Configuring AWS CLI
Appendix B. Introduction to Python
B.1 VariablesB.1.1 Control flowB.1.2 CollectionsB.1.3 Code reusabilityB.1.4 Installing librariesB.1.5 Python programs
Appendix C. Introduction to NumPy
C.1 NumPyC.1.1 NumPy arraysC.1.2 Two-dimensional NumPy arraysC.1.3 Randomly generated arraysC.2 NumPy operationsC.2.1 Element-wise operationsC.2.2 Summarizing operationsC.2.3 SortingC.2.4 Reshaping and combiningC.2.5 Slicing and filteringC.3 Linear algebraC.3.1 MultiplicationC.3.2 Matrix inverseC.3.3 Normal equation
Appendix D. Introduction to Pandas
D.1 PandasD.1.1 DataFrameD.1.2 SeriesD.1.3 IndexD.1.4 Accessing rowsD.1.5 Splitting a DataFrameD.2 OperationsD.2.1 Element-wise operationsD.2.2 FilteringD.2.3 String operationsD.2.4 Summarizing operationsD.2.5 Missing valuesD.2.6 SortingD.2.7 Grouping
Appendix E. AWS SageMaker
E.1 AWS SageMaker NotebooksE.1.1 Increasing the GPU quota limitsE.1.2 Creating a notebook instanceE.1.3 Training a modelE.1.4 Turning off the notebook
index

Content preview from Machine Learning Bookcamp

5 Deploying machine learning models

This chapter covers

Saving models with Pickle
Serving models with Flask
Managing dependencies with Pipenv
Making the service self-contained with Docker
Deploying it to the cloud using AWS Elastic Beanstalk

As we continue to work with machine learning techniques, we’ll keep using the project we already started: churn prediction. In chapter 3, we used Scikit-learn to build a model for identifying churning customers. After that, in chapter 4, we evaluated the quality of this model and selected the best parameter C using cross-validation.

We already have a model that lives in our Jupyter Notebook. Now we need to put this model into production, so other services can use the model to make decisions based on the ...

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: 9781617296819Publisher Support Publisher Website Errata Page

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Machine Learning Bookcamp

by Alexey Grigoriev

5 Deploying machine learning models

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.