blueprint.yaml

slug: aws-sagemaker-timeseries-forecasting
type: Tutorial
category: AI and machine learning
expertiseLevel: "Intermediate"
tags: ["AWS", "AI", "ML", "Time series", "Insights"]
icon: assets/azure-sagemaker-tutorial.svg
gitrepo: https://github.com/particle-iot/blueprint-aws-sagemaker-tutorial
name: "AWS SageMaker tutorial"
shortDescription: "Time-Series Forecasting with AWS SageMaker Autopilot"
version: 1.0.0
models: []
language: ["Particle Wiring", "Python" ]
cloudServices:
  - name: Integrations
  - name: Secure variables
integrations:
  - name: Sagemaker
    icon: assets/azure-sagemaker.svg
supportedDevices:
  - name: Boron
  - name: Argon
  - name: M-SoM
  - name: Photon 2
  - name: Muon
  - name: Monitor One
  - name: Tracker One
  - name: T-SoM
  - name: E-SoM
  - name: P2
hardwareDependencies:
  - name: Supported device
  - name: Grove Temperature & Humidity Sensor (DHT11)
    link: https://wiki.seeedstudio.com/Grove-TemperatureAndHumidity_Sensor/
  - name: Particle Grove Shield
    link: https://docs.particle.io/reference/datasheets/accessories/gen3-accessories/
introduction: |
  This tutorial demonstrates how to set up time-series forecasting using AWS SageMaker with environmental data collected from Particle devices. Using the Grove Temperature & Humidity Sensor and AWS SageMaker's automated machine learning capabilities, you can predict future environmental trends based on temperature and humidity data. This is useful for applications such as weather prediction, industrial monitoring, and smart agriculture.

description: |
  This AWS SageMaker Tutorial provides a step-by-step guide for performing time-series forecasting using AWS SageMaker Autopilot. 
  The tutorial integrates with Particle devices, demonstrating how to collect and transmit environmental data from sensors, such 
  as temperature and humidity readings, to AWS SageMaker for advanced forecasting. Leveraging SageMaker Autopilot's automated 
  machine learning capabilities, users can predict future environmental trends, making it ideal for applications in weather 
  prediction, industrial monitoring, and smart agriculture.

  Key Topics Covered:
  - Configuring Particle devices to capture time-series data from temperature and humidity sensors.
  - Setting up secure data transmission to AWS SageMaker using cloud integrations and AWS Vault for managing credentials.
  - Using AWS SageMaker Autopilot to automatically train and deploy a time-series forecasting model.
  - Analyzing and visualizing forecasting results to gain insights from historical data.

  Prerequisites:
  - A Particle Feather-based development board (e.g., Argon, Boron).
  - A Grove Temperature & Humidity Sensor and a Particle Grove Shield for seamless sensor connectivity.
  - An AWS account with access to SageMaker services and permissions to create Autopilot jobs.

  This tutorial provides a comprehensive example of combining IoT hardware with cloud-based machine learning, enabling users to 
  easily collect data and generate actionable forecasts with minimal manual intervention.

additionalResources:
  - name: Device datasheets
    link: https://docs.particle.io/reference/datasheets