Amazon SageMaker by Miguel Quintanilla

Amazon SageMaker is a fully managed service introduced by Amazon Web Services (AWS) in November 2017. Its purpose is to make it easier for developers and data scientists to build, train, and deploy machine learning (ML) models at scale. Before SageMaker, creating and deploying ML models required significant expertise and manual effort in setting up infrastructure, managing data, and optimizing models.

Create a New SageMaker Project:

  1. Access SageMaker Studio: Once logged in, from the AWS Management Console, search for SageMaker in the search bar and click on SageMaker Studio.
  2. Set Up SageMaker Studio: If it’s your first time using SageMaker Studio, you’ll need to create a SageMaker Domain by selecting a user profile, instance type, and permissions.
  3. After the setup, click on Launch Studio to open the SageMaker Studio interface.

Create a New Notebook:

  1. Start a Notebook: In the SageMaker Studio dashboard, click on File > New > Notebook. Choose the environment (like a Python 3 kernel) for the notebook.
  2. Configure Notebook Settings: Specify instance type, networking, and storage if needed.
  3. Initialize the Notebook: Click Create Notebook to start a new SageMaker notebook where you can write and execute machine learning code.

Train a Model:

  1. Import Data: Use Amazon S3 to store and access the dataset you want to use for training.
  2. Select an Algorithm: You can either use a built-in algorithm or import your own custom algorithm.
  3. Start Training: Define the training job by specifying the algorithm, dataset, instance type, and hyperparameters. Run the training job directly from your notebook.

Amazon SageMaker Console:
https://aws.amazon.com/sagemaker/

Amazon SageMaker Studio Overview:
https://docs.aws.amazon.com/sagemaker/latest/dg/studio.html

Creating a SageMaker Domain (First-time setup for SageMaker Studio):
https://docs.aws.amazon.com/sagemaker/latest/dg/onboard-quick-start.html

Getting Started with Amazon SageMaker Notebooks:
https://docs.aws.amazon.com/sagemaker/latest/dg/notebooks.html

Training a Model in SageMaker:
https://docs.aws.amazon.com/sagemaker/latest/dg/how-it-works-training.html