Generate ML predictions without writing any code
summary
A no‑code front end to Amazon SageMaker lets business analysts upload Excel or pull from Redshift/S3, automatically join and clean data, run a quick 2 to 15 minutes exploratory build or a full AutoML run (~100 algorithm/hyperparameter trials) that returns explainability and batch or single “what‑if” predictions. The demo produces 96% accuracy, picks XGBoost as the best model, exposes notebooks/artifacts for SageMaker Studio review, and can deploy the resulting model as a production endpoint; current limits: 1,000 columns and ~5 GB input.
AI-written summary · about
chapters (8)
- 0:00Start
- 0:41ML is key to innovation
- 1:45How ML drives value creation today
- 4:30How can you scale ML value creation?
- 7:18Challenges analysts face in building ML
- 16:02SageMaker Canvas use cases
- 50:11What questions can your data answer?
- 51:12Getting started
related talks
32mBuild quality ML models easily & quickly with Amazon SageMaker Autopilot
31mTrain and tune ML models to the highest accuracy using Amazon SageMaker
33mBuild, train, and deploy ML models using Amazon SageMaker
31mUse AutoML to create high-quality models with just a few clicks
44mPragmatic Machine Learning in the Cloud
45mDeploying ML Models to Production with Amazon SageMaker
29mGet started with ML in minutes with Amazon SageMaker JumpStart
27mScaling ML to the next level: Hosting thousands of models on SageMaker