Build a prediction ML model using public data
summary
Shows how to build a no-code end-to-end ML pipeline on public Singapore government data using Amazon SageMaker Canvas, training on five years of transactions with added lat/long and distances to amenities. Canvas provides a model preview in minutes, a standard build taking ~2–4 hours (quick build limits to 50k rows), and exports a deployable SageMaker endpoint that integrates into a QuickSight dashboard or a Lex chatbot (examples return S$381k and S$3.1M). Free trial: 2 months, 10 models/month, 1M cells/model.
AI-written summary · about
chapters (10)
- 0:00Introduction and agenda
- 1:02Challenges in AI and ML adoption
- 3:51Scaling ML value creation
- 8:50Amazon SageMaker Canvas overview
- 10:15Demo: Predicting property prices
- 16:36Sharing models with data scientists
- 21:25Integration with QuickSight dashboards
- 26:11Chatbot integration with Amazon Lex
- 28:03Solution recap and deployment
- 30:11Key takeaways and resources
related talks
49mUse Amazon SageMaker to build high-quality ML models faster
52mGenerate ML predictions without writing any code
21mAWS On Air ft. Amazon SageMaker Canvas
33mBuild, train, and deploy ML models using Amazon SageMaker
29mAutomatically Build Models With Amazon Sagemaker
19mUsing Amazon SageMaker to develop and deploy ML models
29mGet started with ML in minutes with Amazon SageMaker JumpStart
31mBuild, train, and deploy ML models using Amazon SageMaker