O'Reilly logo

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

AWS SageMaker, Machine Learning and AI with Python

Video Description

Learn about cloud-based machine learning algorithms and how to integrate them with your applications

About This Video

This course is focused on three aspects:

  • The core of the machine learning process is the algorithm itself.
  • Gaining an intuitive understanding of the algorithm, how does it find the solution, and what knobs are essential to tweak for a successful career in this field. That is where we will focus first.
  • Once we build the model, how do we know if it is good or bad? Or If we want to compare two different models, how do we decide which one to pick? We will look at industry standard metrics and powerful visualization tools that AWS provides to assess the goodness of a model.
  • The third aspect and most exciting part of model development is putting the prediction capability in the hands of the users, validate how they are using it and identify what needs to be refined.
  • There is a whole section in this course dedicated to the integration of machine learning models with your application. You will walk through several integration and security options.

In Detail

This course is designed to make you an expert in AWS machine learning and it teaches you how to convert your cool ideas into highly scalable products in a matter of days.The biggest challenge for a data science professional is how to convert the proof-of-concept models into actual products that your customers can use. There are several courses on machine learning that teach you how to build models in R, Python, Matlab and so forth. However, converting a model into a scalable solution and integrating with your existing application requires a lot of effort and development. The real success of your ideas and concepts depends on how soon you can put the capabilities in the hands of your customers.With the AWS Machine Learning service, you can easily conduct experiments and test your concepts. Once you are happy, you can instantly scale them to support millions of requests. No separate development work is needed.

Table of Contents

  1. Chapter 1 : Introduction
    1. Introduction 00:02:37
    2. Root Account Setup and Billing Dashboard Overview 00:03:41
    3. Enable Access to Billing Data for IAM Users 00:05:15
    4. Create Users Required For the Course 00:14:52
    5. AWS Command Line Interface Tool Setup and Summary 00:05:08
    6. Six Advantages of Cloud Computing 00:06:08
    7. AWS Global Infrastructure Overview 00:06:23
  2. Chapter 2 : SageMaker Service Introduction
    1. SageMaker Overview 00:02:20
    2. Compute Instance Families and Pricing 00:03:03
    3. Algorithms and Data Formats Supported For Training and Inference 00:01:55
  3. Chapter 3 : A.2 XGBoost
    1. XGBoost - Introduction and Comparison with Other Approaches 00:07:14
    2. Demo 1: S3 Bucket Setup 00:02:53
    3. Demo 2: Setup Notebook Instance on SageMaker 00:06:48
    4. Demo 3: Source Code and Data Setup 00:05:21
    5. Demo 4: Create Files in SageMaker Data Formats and Save Files to S3 00:07:57
    6. Demo 5: Working with XGBoost - Linear Regression Straight Line Fit 00:12:36
    7. Demo 6: XGBoost Example with Quadratic Fit 00:04:09
    8. Demo 7: Kaggle Bike Rental Data Setup, Exploration and Preparation 00:11:53
    9. Demo 8: Kaggle Bike Rental Model Version 1 00:10:46
    10. Demo 9: Kaggle Bike Rental Model Version 2 00:04:43
    11. Demo 10: Kaggle Bike Rental Model Version 3 00:04:02
    12. Demo 11: Training on SageMaker Cloud - Kaggle Bike Rental Model Version 3 00:13:55
    13. Demo 12: Invoking SageMaker Model Endpoints for Real Time Predictions 00:05:10
    14. Demo 13: Invoking SageMaker Model Endpoints from Client Outside of AWS 00:03:33
    15. XGBoost Hyper Parameter Tuning 00:05:54
    16. Demo 14: XGBoost Multi-Class Classification Iris Data 00:09:26
    17. Demo 15: XGBoost Binary Classifier for Diabetes Prediction 00:04:57
    18. Demo 16: XGBoost Binary Classifier for Edible Mushroom Prediction 00:04:23
    19. Summary – XGBoost 00:02:12
  4. Chapter 4 : A.3. SageMaker - Principal Component Analysis (PCA)
    1. Introduction PCA and SageMaker PCA 00:05:49
    2. PCA Demo Source Code setup 00:01:57
    3. Demo 1: PCA with Random Dataset 00:03:29
    4. Demo 2: PCA with Correlated Dataset 00:05:26
    5. Demo 3.1: PCA with Kaggle Bike Sharing - Overview and Normalization 00:03:52
    6. Demo 3.2: PCA Local Model with Kaggle Bike Train 00:03:31
    7. Demo 3.3: PCA training with SageMaker 00:04:23
    8. Demo 3.4: PCA Projection with SageMaker 00:02:42
    9. Summary 00:01:23
  5. Chapter 5 : A.4. Factorization Machines For Recommender Systems and Click Rate Prediction
    1. Introduction to Factorization Machines 00:05:59
    2. Demo - Movie Recommender Data Preparation 00:12:05
    3. Demo - Movie Recommender Model Training 00:05:35
    4. Demo - Movie Predictions by User 00:07:10
  6. Chapter 6 : AWS Machine Learning Service
    1. Python Development Environment and Boto3 Setup 00:03:28
    2. Project Source Code and Data Setup 00:04:10
    3. Lab: Intro to Python Jupyter Notebook Environment, Pandas, Matplotlib 00:13:24
    4. Lab: AWS S3 Bucket Setup and Configure Security 00:08:03
    5. Summary 00:01:18
    6. Machine Learning Terminology 00:03:39
    7. Data Types supported by AWS Machine Learning 00:02:56
    8. Linear Regression Introduction 00:05:42
    9. Binary Classification Introduction 00:04:09
    10. Multiclass Classification Introduction 00:03:03
    11. Data Visualization - Linear, Log, Quadratic and More 00:07:51
  7. Chapter 7 : Linear Regression
    1. Lab: Linear Model, Squared Error Loss Function, Stochastic Gradient Descent 00:15:47
    2. Lab: Linear Regression for complex shapes 00:05:09
    3. Summary 00:01:59
  8. Chapter 8 : AWS - Linear Regression Models
    1. Lab: Simple Training Data 00:06:35
    2. Lab: Datasource 00:12:20
    3. Lab: Train Model with default recipe 00:04:17
    4. Concept - How to evaluate regression model accuracy? 00:04:52
    5. Lab: Evaluate predictive quality of the trained model 00:10:52
    6. Lab: Review Default Recipe Settings Used to Train model 00:02:28
    7. Lab: Train Model with Custom Recipe and Review Performance 00:07:05
    8. Model Performance Summary and Conclusion 00:02:24
  9. Chapter 9 : Adding Features to Improve Model
    1. Lab: Quadratic Fit Training Data 00:06:32
    2. Lab: Under fitting with Linear Features 00:15:43
    3. Lab: Normal Fit with Quadratic Features 00:08:11
    4. Summary 00:01:28
  10. Chapter 10 : Normalization
    1. Lab: Impact of Features with Different Magnitude 00:11:26
    2. Concept: Normalization to smoothen magnitude differences 00:04:48
    3. Lab: Train Model with Feature Normalization 00:06:26
    4. Summary 00:01:34
  11. Chapter 11 : Adding Complex Features
    1. Lab: Prepare Training Data 00:03:11
    2. Lab: Adding Complex Features 00:02:16
    3. Lab: Train Model with Higher Order Features 00:07:20
    4. Lab: Performance of Model with Degree 1 Features 00:02:21
    5. Lab: Performance of Model with Degree 4 Features 00:01:52
    6. Lab: Performance of Model with Degree 15 Features 00:01:22
    7. Summary 00:01:47
  12. Chapter 12 : Kaggle Bike Hourly Rental Prediction
    1. Review Kaggle Bike Train Problem and Dataset 00:11:41
    2. Lab: Train Model to Predict Hourly Rental 00:04:22
    3. Lab: Evaluate Prediction Quality 00:07:34
    4. Linear Regression Wrap up and Summary 00:01:54
  13. Chapter 13 : Logistic Regression
    1. Binary Classification - Logistic Regression, Loss Function, Optimization 00:08:47
    2. Lab: Binary Classification Approach 00:07:39
    3. True Positive, True Negative, False Positive and False Negative 00:07:44
    4. Lab: Logistic Optimization Objectives 00:03:55
    5. Lab: Logistic Cost Function 00:03:16
    6. Lab: Cost Example 00:03:51
    7. Optimizing Weights 00:04:28
    8. Summary 00:03:08
  14. Chapter 14 : Onset of Diabetes Prediction
    1. Problem Objective, Input Data and Strategy 00:09:09
    2. Lab: Prepare For Training 00:02:59
    3. Lab: Training a Classification Model 00:03:32
    4. Concept: Classification Metrics 00:05:15
    5. Concept: Classification Insights with AWS Histograms 00:05:13
    6. Concept: AUC Metric 00:02:14
    7. Lab: Review Diabetes Model Performance 00:06:33
    8. Lab: Cutoff Threshold Interactive Testing 00:02:31
    9. Lab: Evaluating Prediction Quality with Additional Dataset 00:07:02
    10. Lab: Batch Prediction and Compute Metrics 00:07:51
    11. Summary 00:01:51
  15. Chapter 15 : Multiclass Classifiers using Multinomial Logistic Regression
    1. Lab: Iris Classification 00:08:03
    2. Lab: Train Classifier with Default and Custom Recipe 00:07:30
    3. Concept: Evaluating Predictive Quality of Multiclass Classifiers 00:02:32
    4. Concept: Confusion Matrix to Evaluating Predictive Quality 00:05:20
    5. Lab: Evaluate Performance of Iris Classifiers using Default Recipe 00:04:23
    6. Lab: Evaluate Performance of Iris Classifiers using Custom Recipe 00:03:22
    7. Lab: Batch Prediction and Computing Metrics using Python Code 00:08:52
    8. Summary 00:03:05
  16. Chapter 16 : Text Based Classification with AWS Twitter Dataset
    1. AWS Twitter Feed Classification for Customer Service 00:05:28
    2. Lab: Train, Evaluate Model and Assess Predictive Quality 00:08:44
    3. Lab: Interactive Prediction with AWS 00:02:56
    4. Logistic Regression Summary 00:00:53
  17. Chapter 17 : Data Transformation using Recipes
    1. Recipe Overview 00:04:21
    2. Recipe Example 00:05:32
    3. Text Transformation 00:06:32
    4. Numeric Transformation - Quantile Binning 00:02:20
    5. Numeric Transformation – Normalization 00:03:21
    6. Cartesian product Transformation - Categorical and Text 00:02:00
    7. Summary 00:00:26
  18. Chapter 18 : Hyper Parameters, Model Optimization and Lifecycle
    1. Introduction 00:00:40
    2. Data Rearrangement, Maximum Model Size, Passes, Shuffle Type 00:07:41
    3. Regularization, Learning Rate 00:02:49
    4. Regularization Effect 00:03:06
    5. Improving Model Quality 00:07:13
    6. Model Maintenance 00:06:02
    7. AWS Machine Learning System Limits 00:02:18
    8. AWS Machine Learning Pricing 00:02:33
  19. Chapter 19 : Integration of AWS Machine Learning With Your Application
    1. Introduction 00:02:58
    2. Integration Scenarios 00:02:29
    3. Security using IAM 00:04:16
    4. Hands-on lab - List of Demos and Objective 00:02:32
    5. Lab: Enable Real Time End Point and Configure IAM Prediction User 00:08:35
    6. Lab: Invoking Prediction from AWS Command Line Interface 00:06:52
    7. Lab: Invoking Prediction from Python Client 00:04:16
    8. Lab: Python Client to Train, Evaluate Models and Integrate with AWS 00:13:38
  20. Chapter 20 : - Web Client with Cognito and AngularJS
    1. Lab: Invoking Prediction from Web Page AngularJS Client 00:08:20
    2. Demo Allowing Prediction Only For Registered Users 00:01:40
    3. Cognito Overview 00:01:40
    4. Lab: Cognito User Pool Configuration 00:07:22
    5. Lab: AngularJS Web Client - Invoke Prediction for authorized users 00:15:08
    6. Lab: Invoke Machine Learning Service From AWS EC2 Instance 00:06:27
    7. Summary 00:00:29
  21. Chapter 21 : Conclusion
    1. Conclusion 00:00:36