Sagemaker Bring Your Own Model Python

The code here will be used to showcase the Bring Your Own Model capabilities of Amazon Sagemaker.

Instance Setup

In this Step, we will configure the Sagemaker Instance that we are going to use throughout this session.

Lifecycle Management

In this section we will configure the Lifecycle Policy for our Sagemaker instance

1. Go to create a new Lifecycle Policy

#!/bin/bash

set -e
pip install --upgrade pip
git config --system user.name "<YOUR NAME>"
git config --system user.email "<YOUR@EMAIL>"

Add Repository

1. URL https://github.com/purnesh/sagemaker-byom-python
2. Branch Name master
3. Create New Secret
4. Add Repository

Role Creation

1. Create a new Role
2. Attach EC2ContainerRegistryFullAccess policy to the newly created role.

Verification and Debugging

1. Go to Cloudwatch
2. Click on Log Groups in the left hand menu
3. Go to /aws/sagemaker/NotebookInstances
4. Click on the appropriate option to view logs