AWS Certified AI PractitionerAWS Services for AI/ML and Generative AIMedium
A research team needs to train a custom deep learning model for a novel image classification task. They require a platform that provides integrated tools for data labeling, model training, hyperparameter tuning, and deployment, with full control over the underlying infrastructure and algorithms. Which AWS service offers this comprehensive, end-to-end machine learning lifecycle management?
- AAmazon Rekognition Custom Labels
- BAmazon Comprehend
- CAmazon SageMaker
- DAmazon Forecast
Show answer & explanationAnswer & explanation
Correct answer: C. Amazon SageMaker
Amazon SageMaker is a fully managed machine learning service that enables data scientists and developers to build, train, and deploy machine learning models quickly. It provides a complete end-to-end ML workflow, including data labeling (Ground Truth), notebooks, training jobs, hyperparameter tuning, and model deployment.
Why the other options are wrong
- A. Amazon Rekognition Custom Labels is for training custom object detection/classification models on images, but it's a higher-level service with less control compared to SageMaker's full capabilities.
- B. Amazon Comprehend is a pre-trained NLP service, not for building custom deep learning models.
- D. Amazon Forecast is a specialized service for time-series forecasting, not for general custom deep learning.
Amazon SageMaker
A fully managed service that provides every developer and data scientist with the ability to build, train, and deploy machine learning models quickly.
- Offers an integrated environment for the entire ML workflow.
- Includes tools for data labeling (Ground Truth), notebooks, training, tuning, and deployment.
- Supports various ML frameworks (TensorFlow, PyTorch, MXNet) and custom algorithms.
Memory trick: SageMaker handles the entire ML journey, making it wise for custom models.