Manufacturer : Udemy Manufacturer
Website : https://www.udemy.com/course/practical-aws-sagemaker-6-real-world-case-studies/
Author : Dr. Ryan Ahmed, Kirill Eremenko
Duration : 14h 43m Handout
Type : Video Tutorial
Language : English
Description : SageMaker is a fully managed service within AWS that allows data scientists and AI practitioners to train, test, and deploy AI/ML models quickly and efficiently.
In this course, students will learn how to create AI/ML models using AWS SageMaker.
The course covers many topics such as data engineering, AWS services and algorithms, and machine/deep learning basics in a practical way:
Data engineering: Data types, key python libraries (pandas, Numpy, scikit Learn, MatplotLib, and Seaborn), data distributions and feature engineering (imputation, binning, encoding, and normalization).
AWS services and algorithms: Amazon SageMaker, Linear Learner (Regression/Classification), Amazon S3 Storage services, gradient boosted trees (XGBoost), image classification, principal component analysis (PCA), SageMaker Studio and AutoML.
Machine and deep learning basics: Types of artificial neural networks (ANNs) such as feedforward ANNs, convolutional neural networks (CNNs), activation functions (sigmoid, RELU and hyperbolic tangent), machine learning training strategies (supervised/ unsupervised), gradient descent algorithm, learning rate, backpropagation, bias, variance, bias-variance trade-off, regularization (L1 and L2), overfitting, dropout, feature detectors, pooling, batch normalization, vanishing gradient problem, confusion matrix, precision, recall, F1-score, root mean squared error (RMSE), ensemble learning, decision trees, and random forest.
Sample files : present
Video format : MP4
Video : H264 AVC, 1920x1080, 16:9, 30fps, ~800kbps
Audio : AAC, 44.1kHz, 128kbps, stereo
Content
Introduction, Success Tips & Best Practices and Key Learning Outcomes
Course Introduction and Welcome Message
Updates on Udemy Reviews
Course Key Tips and Best Practices
Course Outline and Key Learning Outcomes
Get the Materials
BONUS: Learning Path
Section 2: Introduction to AI/ML, AWS and Cloud Computing
AWS Free Tier Account Setup and Overview
Introduction to AI, Machine Learning and Deep Learning
Introduction to AI, Machine Learning and Deep Learning - Part #2
Good Data Vs. Bad Data
Introduction to AWS and Cloud Computing
Key Machine Learning Components and AWS Management Console Tour
AWS Regions and Availability Zones
Amazon S3
Amazon EC2 and IAM
AWS SageMaker Overview
AWS SageMaker Walk-through
AWS SageMaker Studio Overview
AWS SageMaker Studio Walk-through
SageMaker Models Deployment
Section 3: Project #1 - Employee Salary Predictions Using AWS SageMaker Linear Learner
Project Overview
Simple Linear Regression Intuition
Least Sum of Squares
AWS SageMaker Linear Learner Overview
Coding Task #1A - Instantiate AWS SageMaker Notebook Instance (Method #1)
Coding Task #1B - Using AWS SageMaker Studio (Method #2)
Coding Task #2 - Import Key libraries and dataset
Coding Task #3 - Perform Exploratory Data Analysis
Coding Task #4 - Create Training and Testing Dataset
Coding Task #5 - Train a Linear Regression Model in SkLearn
Coding Task #6 - Evaluate Trained Model Performance
Coding Task #7 - Train a Linear Learner Model in AWS SageMaker
Coding Task #8 - Deploy Model & invoke endpoint in SageMaker
Section 4: Project #2 - Medical Insurance Premium Prediction
Project Overview and Introduction
Multiple Linear Regression Intuition
Regression Metrics and KPIs - RMSE, MSE, MAE, MAPE
Regression Metrics and KPIs - R2 and Adjusted R2
Coding Task #1 & #2 - Import Dataset and Key Libraries
Coding Task #3 - Perform Exploratory Data Analysis
Coding Task #4 - Perform Data Visualization
Coding Task #5 - Create Training and Testing Datasets
Coding Task #6 - Train a Machine Learning Model Locally
Coding Task #7 - Train a Linear Learner Model in AWS SageMaker
Coding Task #8 - Deploy Trained Model and Invoke Endpoint
Artificial Neural Networks for Regression Tasks
Activation Functions - Sigmoid, RELU and Tanh
Multilayer Perceptron Networks
How do Artificial Neural Networks Train?
Gradient Descent Algorithm
Backpropagation Algorithm
Coding Task #9 - Train Artificial Neural Networks for Regression Tasks
Section 5: Project #3 - Retail Sales Prediction Using AWS SageMaker XGBoost (Regression)
Introduction to Case Study
Basics: What is the difference between Bias & Variance
Basics: L1 & L2 Regularization - Part #1
Basics: L1 & L2 Regularization - Part #2
Introduction to XGBoost (Extreme Gradient Boosting) algorithm
What is Boosting?
Decision Trees and Ensemble Learning
Gradient Boosted Trees - Deep Dive - Part #1
Gradient Boosted Trees - Deep Dive - Part #2
AWS SageMaker XGBoost Algorithm
Project Introduction and Notebook Instance Instantiation
Coding Task #1 #2 #3 - Load Dataset/Libraries and Perform Data Exploration
Coding Task #4 - Merge and Manipulate DataFrame Using Pandas
Coding Task #5 - Explore Merged Datasets
Coding Task #6 #7 - Visualize Dataset
Coding Task #8 - Prepare the Data To Perform Training
Coding Task #9 - Train XGBoost Locally
Coding Task #10 - Train XGBoost Using SageMaker
Coding Task #11 - Deploy XGBoost endpoint and Make Predictions
Coding Task #12 - Perform Hyperparameters Tuning
Coding Task #13 - Retrain the Model Using best (optimized) Hyperparameters
Section 6: Project #4 - Predict Cardiovascular Disease Using PCA & XGBoost (Classification)
Introduction and Project Overview
Principal Component Analysis (PCA) Intuition
XGBoost for Classification Tasks (Review Lecture)
Confusion Matrix
Precision, Recall, and F1-Score
Area Under Curve (AUC) and Receiver Operating Characteristics (ROC) Metrics
Overfitting and Under fitting Models
Coding Task #1 - SageMaker Studio Notebook Setup
Coding Task #2 & #3 - Import Data/Libraries & Perform Exploratory data analysis
Coding Task #4 & #5 - Visualize Datasets & Prepare Training/Testing Data
Coding Task #6 - Train & Test XGboost and Perform Grid Search (Local Mode)
Coding Task #7 - Train a PCA Model in AWS SageMaker
Coding Task #8 - Deploy Trained PCA Model Endpoint & Envoke endpoint
Coding Task #9 - Train XGBoost (SageMaker Built-in) to do Classification Tasks
Coding Task #10 - Deploy Endpoint, Make Inference @ Test Model
Section 7: Project #5 - Deep Learning for Traffic Sign Classification Using AWS SageMaker
Project Overview and Introduction
What are Convolutional Neural Networks and How do they Learn? - Part #1
What are Convolutional Neural Networks and How do they Learn? - Part #2
How to Improve CNNs Performance?
Confusion Matrix
LeNet Network Architecture
Request AWS SageMaker Service Limit Increase
Coding Part #1 #2 - Import Images and Visualize Them
Coding #3 #4 - Upload Training/Testing Data to S3
Coding Task #5 - Build and Train CNNs
Coding Task #6 - Deploy Trained Model Using SageMaker
Section 8: Project #6 - SageMaker Studio DeepDive and AutoML
Introduction to Case Study
Download
Website : https://www.udemy.com/course/practical-aws-sagemaker-6-real-world-case-studies/
Author : Dr. Ryan Ahmed, Kirill Eremenko
Duration : 14h 43m Handout
Type : Video Tutorial
Language : English
Description : SageMaker is a fully managed service within AWS that allows data scientists and AI practitioners to train, test, and deploy AI/ML models quickly and efficiently.
In this course, students will learn how to create AI/ML models using AWS SageMaker.
The course covers many topics such as data engineering, AWS services and algorithms, and machine/deep learning basics in a practical way:
Data engineering: Data types, key python libraries (pandas, Numpy, scikit Learn, MatplotLib, and Seaborn), data distributions and feature engineering (imputation, binning, encoding, and normalization).
AWS services and algorithms: Amazon SageMaker, Linear Learner (Regression/Classification), Amazon S3 Storage services, gradient boosted trees (XGBoost), image classification, principal component analysis (PCA), SageMaker Studio and AutoML.
Machine and deep learning basics: Types of artificial neural networks (ANNs) such as feedforward ANNs, convolutional neural networks (CNNs), activation functions (sigmoid, RELU and hyperbolic tangent), machine learning training strategies (supervised/ unsupervised), gradient descent algorithm, learning rate, backpropagation, bias, variance, bias-variance trade-off, regularization (L1 and L2), overfitting, dropout, feature detectors, pooling, batch normalization, vanishing gradient problem, confusion matrix, precision, recall, F1-score, root mean squared error (RMSE), ensemble learning, decision trees, and random forest.
Sample files : present
Video format : MP4
Video : H264 AVC, 1920x1080, 16:9, 30fps, ~800kbps
Audio : AAC, 44.1kHz, 128kbps, stereo
Content
Introduction, Success Tips & Best Practices and Key Learning Outcomes
Course Introduction and Welcome Message
Updates on Udemy Reviews
Course Key Tips and Best Practices
Course Outline and Key Learning Outcomes
Get the Materials
BONUS: Learning Path
Section 2: Introduction to AI/ML, AWS and Cloud Computing
AWS Free Tier Account Setup and Overview
Introduction to AI, Machine Learning and Deep Learning
Introduction to AI, Machine Learning and Deep Learning - Part #2
Good Data Vs. Bad Data
Introduction to AWS and Cloud Computing
Key Machine Learning Components and AWS Management Console Tour
AWS Regions and Availability Zones
Amazon S3
Amazon EC2 and IAM
AWS SageMaker Overview
AWS SageMaker Walk-through
AWS SageMaker Studio Overview
AWS SageMaker Studio Walk-through
SageMaker Models Deployment
Section 3: Project #1 - Employee Salary Predictions Using AWS SageMaker Linear Learner
Project Overview
Simple Linear Regression Intuition
Least Sum of Squares
AWS SageMaker Linear Learner Overview
Coding Task #1A - Instantiate AWS SageMaker Notebook Instance (Method #1)
Coding Task #1B - Using AWS SageMaker Studio (Method #2)
Coding Task #2 - Import Key libraries and dataset
Coding Task #3 - Perform Exploratory Data Analysis
Coding Task #4 - Create Training and Testing Dataset
Coding Task #5 - Train a Linear Regression Model in SkLearn
Coding Task #6 - Evaluate Trained Model Performance
Coding Task #7 - Train a Linear Learner Model in AWS SageMaker
Coding Task #8 - Deploy Model & invoke endpoint in SageMaker
Section 4: Project #2 - Medical Insurance Premium Prediction
Project Overview and Introduction
Multiple Linear Regression Intuition
Regression Metrics and KPIs - RMSE, MSE, MAE, MAPE
Regression Metrics and KPIs - R2 and Adjusted R2
Coding Task #1 & #2 - Import Dataset and Key Libraries
Coding Task #3 - Perform Exploratory Data Analysis
Coding Task #4 - Perform Data Visualization
Coding Task #5 - Create Training and Testing Datasets
Coding Task #6 - Train a Machine Learning Model Locally
Coding Task #7 - Train a Linear Learner Model in AWS SageMaker
Coding Task #8 - Deploy Trained Model and Invoke Endpoint
Artificial Neural Networks for Regression Tasks
Activation Functions - Sigmoid, RELU and Tanh
Multilayer Perceptron Networks
How do Artificial Neural Networks Train?
Gradient Descent Algorithm
Backpropagation Algorithm
Coding Task #9 - Train Artificial Neural Networks for Regression Tasks
Section 5: Project #3 - Retail Sales Prediction Using AWS SageMaker XGBoost (Regression)
Introduction to Case Study
Basics: What is the difference between Bias & Variance
Basics: L1 & L2 Regularization - Part #1
Basics: L1 & L2 Regularization - Part #2
Introduction to XGBoost (Extreme Gradient Boosting) algorithm
What is Boosting?
Decision Trees and Ensemble Learning
Gradient Boosted Trees - Deep Dive - Part #1
Gradient Boosted Trees - Deep Dive - Part #2
AWS SageMaker XGBoost Algorithm
Project Introduction and Notebook Instance Instantiation
Coding Task #1 #2 #3 - Load Dataset/Libraries and Perform Data Exploration
Coding Task #4 - Merge and Manipulate DataFrame Using Pandas
Coding Task #5 - Explore Merged Datasets
Coding Task #6 #7 - Visualize Dataset
Coding Task #8 - Prepare the Data To Perform Training
Coding Task #9 - Train XGBoost Locally
Coding Task #10 - Train XGBoost Using SageMaker
Coding Task #11 - Deploy XGBoost endpoint and Make Predictions
Coding Task #12 - Perform Hyperparameters Tuning
Coding Task #13 - Retrain the Model Using best (optimized) Hyperparameters
Section 6: Project #4 - Predict Cardiovascular Disease Using PCA & XGBoost (Classification)
Introduction and Project Overview
Principal Component Analysis (PCA) Intuition
XGBoost for Classification Tasks (Review Lecture)
Confusion Matrix
Precision, Recall, and F1-Score
Area Under Curve (AUC) and Receiver Operating Characteristics (ROC) Metrics
Overfitting and Under fitting Models
Coding Task #1 - SageMaker Studio Notebook Setup
Coding Task #2 & #3 - Import Data/Libraries & Perform Exploratory data analysis
Coding Task #4 & #5 - Visualize Datasets & Prepare Training/Testing Data
Coding Task #6 - Train & Test XGboost and Perform Grid Search (Local Mode)
Coding Task #7 - Train a PCA Model in AWS SageMaker
Coding Task #8 - Deploy Trained PCA Model Endpoint & Envoke endpoint
Coding Task #9 - Train XGBoost (SageMaker Built-in) to do Classification Tasks
Coding Task #10 - Deploy Endpoint, Make Inference @ Test Model
Section 7: Project #5 - Deep Learning for Traffic Sign Classification Using AWS SageMaker
Project Overview and Introduction
What are Convolutional Neural Networks and How do they Learn? - Part #1
What are Convolutional Neural Networks and How do they Learn? - Part #2
How to Improve CNNs Performance?
Confusion Matrix
LeNet Network Architecture
Request AWS SageMaker Service Limit Increase
Coding Part #1 #2 - Import Images and Visualize Them
Coding #3 #4 - Upload Training/Testing Data to S3
Coding Task #5 - Build and Train CNNs
Coding Task #6 - Deploy Trained Model Using SageMaker
Section 8: Project #6 - SageMaker Studio DeepDive and AutoML
Introduction to Case Study
Download
Last edited: