Datasaur Dinamic with Amazon SageMaker Autopilot

Amazon SageMaker Autopilot

Introduction

Datasaur Dinamic with Amazon SageMaker integration enables you to create an end-to-end model building process using the combination of Datasaur and AWS. This will directly connect your labeled data to your Amazon Sagemaker environment and start the training.

Quick Start Guide

Here's a step-by-step guide to achieving optimal results with our Datasaur Dinamic:

Setup Datasaur Dinamic Extension

Please make sure you have access to our Datasaur Dinamic feature to enable the Datasaur Dinamic extension in the extension settings.

1. Create a custom project for row labeling and add your pre-labeled data or unlabeled data to this project.

2. We recommend labeling the data and reviewing it to check data quality.

3. Click Manage extensions (gear icon) from the extension panel on the right.

4. The Manage extensions dialog will appear and you can enable the Datasaur Dinamic extension.

Manage Extension Pop Up

1. Datasaur Dinamic is now enabled and ready for you to use.

   

Setup Role in AWS Identity and Access Management (IAM)

Please make sure you have an AWS S3 Bucket available to run your model training and store it. Write the AWS S3 Bucket Name in Datasaur Dinamic.

Before we start to create Role in the IAM page, please create these policies first for your permissions when creating a new Role.

1. Go to your IAM page, navigate to the Policies section, and then create a new policy.

   
2. You will need to create 2 policies, sagemaker-limited-access and sagemaker-s3-specific-permissions. You can download the examples below and edit the variables according to your bucket information.

1KB

sagemaker-s3-specific-permissions.txt

Open

sagemaker-s3-specific-permissions

3KB

sagemaker-limited-access.txt

Open

sagemaker-limited-access

Now you have 2 policies for role permission access. We can continue to create a new Role

1. Go to the IAM page and navigate to the Roles section

2. Click on Create roles then select AWS account as the role type and insert the Account ID and External ID. These values are automatically generated within the Datasaur Dinamic and can be copied and pasted into the AWS platform.

   

   Select trusted entity
3. After finishing the first step, you can see the Add permissions step and search the policies you created before. Add them to the permissions policies.

   

   Add permissions
4. Now in IAM Roles step 3 (Name, review, and create), you can start to specify the name of the Role and detail.

5. After you are done creating the Role, go to the Trust Relationships tab to edit it and add these lines

   {
       "Effect": "Allow",
       "Principal": {
           "Service": "sagemaker.amazonaws.com"
       },
       "Action": "sts:AssumeRole"
   }

Edit Trust Relationships

1. You should able to get the ARN of that role in the Role Summary and copy it to Datasaur Dinamic.

   

   Role Summary

Start Training

1. At this point, you should have AWS S3 Bucket Name and Role ARN inputs filled in the Datasaur Dinamic extension.

Datasaur Dinamic with Amazon Sagemaker

1. Now, click the Train button to start your training using Amazon SageMaker Autopilot.

Datasaur Dinamic training progress

1. Voila! You will get your model name, and you can search for it in your Amazon SageMaker Model list.

Datasaur Dinamic Model Name