AWS Cert - AI Practitioner Exam

 

AWS Certification Prep

AI Practitioner

AI

Process: Training Data (labeled or not; structured or not) => ML algorithm => Model 

Steps: 1) Get data, 2) ML (supervised, unsupervised, reinforced) 3)  Inferencing (batch or real-time)

SAGE MAKER

SageMaker = ML workflow mgmt that offers auto training, tuning, and deployment of ML and solves data prep and model training tasks. Uses are: predictive analytics, image recognition, voice AI with NLP, fraud detection, recommendation systems, and time-series forecasting.  Security via IAM. Does its own logging. CloudWatch can monitor metrics and set alarms. Integrates with S3 for storing data, Lambda for processing events, API Gateway for exposing models as APIs.

    Studio = web-based IDE for ML with tools for data prep, model building, training, and deployment.

    Studio notebooks = Jupyter notebooks with ML libraries and tools. Use Studio notebooks to write, run, and share code for data exploration, model training, and deployment.

     Training = trains ML models on various compute instances, including GPU accelerated instances using distributed service.

     Inference = deploys trained models as hosted services for real-time predictions or batch transformations.

     Ground Truth = data labeling service creates high-quality training datasets by sending the most difficult ones (hardest 30%) to crowd-sourced humans by outsourcing data labeling tasks.