From the course: Leveraging Cloud-Based Machine Learning on AWS: Real-World Applications

Create a SageMaker notebook - Amazon Web Services (AWS) Tutorial

From the course: Leveraging Cloud-Based Machine Learning on AWS: Real-World Applications

Start my 1-month free trial

Create a SageMaker notebook

“

- [Instructor] So on leveraging SageMaker, the first thing we're going to do is enter the AWS Management Console. This is really the jumping off point for pretty much everything you're going to do within AWS. So from here, we'll go ahead and find SageMaker under Machine Learning and there it is and I'll go ahead and click on that. Now, note it gives you different ways to do things, such a Ground Truth as we discussed, the ability to build and leverage notebooks, the ability to train the models and the ability to, in essence, create models from the training jobs or import external models. Here we can see Recent activity. And if I had any, it would appear there as well as Learning Content and Feature Spotlight. So the first thing we're going to do is create a Notebook instance and we can go to Notebook Instances here. And up here press Create an Instance. I can search for existing instances if some are there and typically you're going to have many different notebooks that you've created. I'll go ahead and create a new one. I have to give it an instance name and this time I'll call this Demo-1 and -Lynda. And notice the restrictions here, the maximum of 63 alphanumeric characters. Can include hyphens but not spaces. It must be unique within your account and AWS Region. And now we have to pick the Notebook instance type and having done this a few times, I'm going to go ahead and pick extra large 'cause some reason, if you use a medium instance, it runs out of space and Elastic Inference, I can choose extra large. I'll go ahead and do that. And then there's also Additional configuration data such as lifecycle configuration which is optional. I have, you can create those things and the volume size. And gigabytes, it defaults to five and I think that's going to be fine. So Permissions and encryption. I've an Identity and Access Management role. So this basically provides me with the ability to create a new role or use an existing role and I've created a role before and I'm going to go ahead and use that. And for you to access optional, I'm going to give users root access to the notebook and since this is just a demo, that shouldn't be a problem. An encryption key, optional, I can actually enter an encryption key here if I'm dealing with very secure data but this being a demo, I'm going to have no custom encryption. We have the Virtual Private Cloud and your notebook instance will be provided with SageMaker to ultimately provide access to a VPC, Virtual Private Cloud and that's an optional thing, so I'm going to ahead and not specify it. I can leverage a Git repository. I'm not going to need one for this demo but I can do that and I can deal with tagging. Keys and value pair 'til we basically filled the systems. So from there, hopefully we've set all the parameters correctly. I'm going to go ahead and Create notebook. And you notice that I get back to the Console. Success, your notebook instance is being created. Open the notebook instance when the status is inService and open the template notebook to get started. And notice it's going to be Pending and this make take five minutes, 10 minutes sometimes before it goes into the AWS infrastructure and allocates everything that's needed.

Contents