Lesson List

Welcome to the Nanodegree program

Introduction to Software Engineering

In this lesson, you’ll write production-level code and practice object-oriented programming, which you can integrate into machine learning projects.

Software Engineering Practices Pt I

Software Engineering Practices Pt II

OOP

Portfolio Exercise: Upload a Package to PyPi

Part 15-Module 01-Lesson 06_Web Development

Portfolio Exercise: Deploy a Data Dashboard

Introduction to Data Engineering

ETL Pipelines

Introduction to NLP

Learn Natural Language Processing one of the fields with the most real applications of Deep Learning

Machine Learning Pipelines

Disaster Response Pipeline

Project1: Disaster Response Pipeline

Part 17-Module 02-Lesson 01_Concepts in Experiment Design

Part 17-Module 02-Lesson 02_Statistical Considerations in Testing

Statistical Considerations in Testing

Part 17-Module 02-Lesson 03_AB Testing Case Study

A/B Testing Case Study

Part 17-Module 02-Lesson 04_Portfolio Exercise Starbucks

Part 17-Module 03-Lesson 01_Introduction to Recommendation Engines

Part 17-Module 03-Lesson 02_Matrix Factorization for Recommendations

Part 17-Module 04-Lesson 01_Recommendation Engines

Part 17-Module 05-Lesson 01_Upcoming Lesson

Sentiment Prediction RNN

Convolutional Neural Networks

Transfer Learning

Weight Initialization

Autoencoders

Job Search

Find your dream job with continuous learning and constant effort

Refine Your Entry-Level Resume

Craft Your Cover Letter

Optimize Your GitHub Profile

Develop Your Personal Brand

Go to course home
01. Updating a Model

In this lesson we are going to take a look at updating an existing endpoint so that it conforms to a different endpoint configuration. There are many reasons for wanting to do this, the two that we will look at are, performing an A/B test and updating a model which is no longer performing as well.

To start, we will look at performing an A/B test between two different models. Then, once we’ve decided on a model to use, updating the existing endpoint so that it only sends data to a single model.

For the second example, it may be the case that once we’ve built a model and begun using it, the assumptions on which our model is built begin to change.

For instance, in the sentiment analysis examples that we’ve looked at our models are based on a vocabulary consisting of the 5000 most frequently appearing words in the training set. But what happens if, over time, the usage of words changes? Then our model may not be as accurate.

When this happens we may need to modify our model, often this means re-training it. When we do, we’d like to update the existing endpoint without having to shut it down. Fortunately, SageMaker allows us to do this in a straightforward way.

Previous
Next