Use Amazon SageMaker Studio Lab starter assets - Amazon SageMaker

Use Amazon SageMaker Studio Lab starter assets

Amazon SageMaker Studio Lab supports the following assets to help machine learning (ML) practitioners get started. This guide shows you how to clone notebooks for your project.

Getting started notebook

Studio Lab comes with a starter notebook that gives general information and guides you through key workflows. When you launch your project runtime for the first time, this notebook automatically opens.

Dive into Deep Learning

Dive into Deep Learning (D2L) is an interactive, open-source book that teaches the ideas, mathematical theory, and code that power machine learning. With over 150 Jupyter notebooks, D2L provides a comprehensive overview of deep learning principles. For more information about D2L, see the D2L website.

The following procedure shows how to clone the D2L Jupyter notebooks to your instance.

  1. Start and open the Studio Lab project runtime environment by following Start your project runtime.

  2. Once Studio Lab is open, choose the Git tab ( Black square icon representing a placeholder or empty image. ) on the left sidebar.

  3. Choose Clone a Repository. Under Git repository URL (.git) paste the MLU git repository D2L by following the steps below. If you do not see the Clone a Repository option because you are currently in a Git repository, return to the user directory to clone a new repository. You return to the user directory by choosing the Folder tab ( Black square icon representing a placeholder or empty image. ) on the left sidebar. In the Folder tab beneath the file search bar choose the folder icon to the left of the currently open repository. Once you are in the user directory, choose the Git tab on the left sidebar and choose Clone a Repository.

  4. Navigate to the Studio Lab project overview page. The URL takes the following format.

    https://studiolab.sagemaker.aws/users/<YOUR_USER_NAME>
  5. Under New to machine learning?, choose Dive into Deep Learning.

  6. From the new Dive into Deep Learning browser tab, choose GitHub to open a new page with the example notebooks.

  7. Choose Code and copy the GitHub repository's URL in the HTTPS tab.

  8. Return to the Studio Lab open project browser tab, paste the D2L repository URL, and clone the repository.

AWS Machine Learning University

The AWS Machine Learning University (MLU) provides access to the machine learning courses used to train Amazon’s own developers. With AWS MLU, any developer can learn how to use machine learning with the learn-at-your-own-pace MLU Accelerator learning series. The MLU Accelerator series is designed to help developers begin their ML journey. It offers three-day foundational courses on these three subjects: Natural Language Processing, Tabular Data, and Computer Vision. For more information, see Machine Learning University.

The following procedure shows how to clone the AWS MLU Jupyter notebooks to your instance.

  1. Start and open the Studio Lab project runtime environment by following Start your project runtime.

  2. Once Studio Lab is open, choose the Git tab ( Black square icon representing a placeholder or empty image. ) on the left sidebar.

  3. Choose Clone a Repository. Under Git repository URL (.git) paste the MLU git repository URL by following the steps below. If you do not see the Clone a Repository option because you are currently in a Git repository, return to the user directory to clone a new repository. You return to the user directory by choosing the Folder tab ( Black square icon representing a placeholder or empty image. ) on the left sidebar. In the Folder tab beneath the file search bar choose the folder icon to the left of the currently open repository. Once you are in the user directory, choose the Git tab on the left sidebar and choose Clone a Repository.

  4. Navigate to the Studio Lab project overview page. The URL takes the following format.

    https://studiolab.sagemaker.aws/users/<YOUR_USER_NAME>
  5. Under New to machine learning?, choose AWS Machine Learning University.

  6. From the new AWS Machine Learning University browser tab, find a course that interests you by reading the Course Summary for each course.

  7. Choose the corresponding GitHub repository of interest under Course Content, to open a new page with the example notebooks.

  8. Choose Code and copy the GitHub repository's URL in the HTTPS tab.

  9. Return to the Studio Lab open project browser tab, paste the D2L repository URL, and choose Clone to clone the repository.

 Roboflow

Roboflow gives you the tools to train, fine-tune, and label objects for computer vision applications. For more information, see https://roboflow.com/.

The following procedure shows how to clone the Roboflow Jupyter notebooks to your instance.

  1. Navigate to the Studio Lab project overview page. The URL takes the following format.

    https://studiolab.sagemaker.aws/users/<YOUR_USER_NAME>
  2. Under Resources and community, find Try Computer Vision.

  3. Under Try Computer Vision choose a Roboflow model. For more information, see https://roboflow.com/.

  4. Follow the tutorial under the Notebook preview.