Deployment of Machine Learning Models in Production - eLearning

Course timeline

1. Introduction

   Lesson 01

   • What is Model Deployment?
   • Types of Model Deployment
   • How to Choose the Model Deployment Type?

2. AWS SageMaker

   Lesson 02

   • AWS SageMaker Equivalent on GCP and Azure
   • Sign into Your AWS Account
   • Setting up AWS SageMaker Studio
   • Opening Jupyter on SageMaker Studio

3. Model Training

   Lesson 03

   • Cloning the Lesson Repository
   • Downloading Data-Part
   • Exploratory Data Analysis and Feature Engineering
   • Base Model Training Code
   • Test Model Locally
   • SageMaker Training Job
   • Hyperparameter Tuning
   • Analyze Results

4. SageMaker Real-time Inference

   Lesson 04

   • Architecture of SageMaker Real-time Inference
   • Create the Inference Script
   • Real-time Endpoint Deployment
   • Invoke the Model
   • Cleanup
   • Introduction to Multi-model Endpoint
   • Deploying Multi-model Endpoint
   • Invoke the Multi-model Endpoint
   • Introduction to Serverless
   • Deploying as a Serverless Inference

5. SageMaker Batch Transform

   Lesson 05

   • Architecture of SageMaker Batch Transform
   • Create the Inference Script for Batch Transform
   • Trigger a Batch Transform Job
   • Analyze Results

6. MLOps on SageMaker

   Lesson 06

   • MLOps: Machine Learning Operations
   • Implement MLOps on AWS Cloud Using SageMaker
   • Create an MLOps Project with a SageMaker Template
   • SageMaker Project Template Code
   • Debug Application Errors with Jupyter Notebook
   • Push Code Changes to Trigger CI/CD
   • Test the Endpoint
   • Cleanup