Overview of AWS Machine Learning Services

Whether you know Python, R or any other language which can help you to build machine learning models, AWS has a service for you. You just need to know which one.

AWS has probably the most comprehensive if not complete set of machine learning (ML) services available in the cloud. From pre-trained AI services that require no ML expertise, to fully managed platforms for building custom models, AWS has something for everyone.

I was curious to know what AWS has to offer for Generative AI and general machine learning, so I did some research and here’s a breakdown of the key services and when to use them.

List of AWS ML Services#

Following table summarizes the key AWS machine learning services and their primary use cases:

Service Name	Description
Amazon Rekognition	Image and video analysis (faces, objects, unsafe content)
Amazon Comprehend	Natural Language Processing (sentiment, entities, key phrases)
Amazon Polly	Text-to-Speech
Amazon Transcribe	Speech-to-Text
Amazon Translate	Language translation
Amazon Textract	Extract text, tables, and forms from scanned documents
Amazon Kendra	Enterprise search with natural language understanding
Amazon Lex	Build conversational interfaces (chatbots)
Amazon Forecast	Time series forecasting
Amazon Personalize	Real-time personalized recommendations
Amazon SageMaker	Build, train, and deploy custom ML models
Amazon Bedrock	Access and fine-tune foundation models for GenAI workloads
AWS Q Business	Build and deploy GenAI applications with no ML expertise required
AWS Q Developer	Build and deploy GenAI applications with ML expertise required

List of features of AWS SageMaker#

Feature Name	Description
SageMaker Studio	Integrated development environment for ML with Jupyter notebooks
SageMaker Autopilot	Automatically build, train, and tune models with minimal input
SageMaker Ground Truth	Labeling service for creating high-quality training datasets
SageMaker Training	Managed training of models at scale with distributed training capabilities
SageMaker Inference	Real-time and batch inference endpoints for deploying models
SageMaker Pipelines	CI/CD for ML workflows to automate model building and deployment
SageMaker Feature Store	Central repository for storing and managing features used in ML models
SageMaker Debugger	Real-time monitoring and debugging of training jobs
SageMaker Model Monitor	Continuous monitoring of deployed models for data drift and quality
SageMaker Neo	Optimize models for edge devices with automatic compilation
SageMaker JumpStart	Pre-built solutions and models for common use cases

All Services in Detail#

• Amazon Comprehend
  • Natural Language Processing (NLP) service that uses machine learning to find insights and relationships in text.
  • Key features include:
  • Sentiment analysis
  • Entity recognition
  • Key phrase extraction
  • Language detection
  • Topic modeling
  • Use cases:
  • Analyzing customer feedback
  • Automating content classification
  • Extracting insights from documents
• Amazon Rekognition