AWS Certified Machine Learning

Description of the training

This training instills skills in the creation, training, evaluation and production of machine learning (ML) models on the AWS platform. It covers the AWS services and key concepts of machine learning, ranging from data preparation to model evaluation, optimisation and integration of solutions into a cloud environment.

Objectives of training

At the end of this training, participants will be able to:

• Mastering AWS services for Machine Learning: Learn how to use services like Amazon SageMaker, AWS Lambda, Rekognition, Polly, and Deep Learning AMIs.
• Prepare data for machine learning: Learn to clean, transform and organize data for training ML models with tools like AWS Glue and Amazon S3.
• Training and evaluating machine learning models: Choosing the right algorithms, training models on SageMaker, and evaluating their performance using appropriate metrics.
• Optimize machine learning models: Apply techniques to improve model accuracy by adjusting hyperparameters.
• Deploy ML models in production: Learn how to deploy models in real-time or batch mode via SageMaker.
• Monitor model performance in production: Use tools like SageMaker Model Monitor to track and adjust models in production.
• Apply best safety and compliance practices: Integrate security practices to protect data and ensure compliance with regulations using AWS services such as IAM and KMS.
• Prepare for the certification exam: Review key concepts and prepare for the AWS Certified Machine Learning – Specialty exam by doing simulations.

Who is this training for?

The training is aimed at a wide audience, including:

• Data Scientists: Professionals developing machine learning models and using AWS for training, evaluation and deployment.
• Engineers Machine Learning: Responsible for the implementation, optimization and integration of ML models into AWS production systems.
• Cloud solutions architects: Architectural designers integrating machine learning solutions to solve specific business problems with AWS.
• Software Developers: Those who integrate machine learning capabilities into their applications via AWS, using services like SageMaker.
• Advanced Data Analysts: Professionals analyzing massive data and wishing to deepen their machine learning skills on AWS.
• AI Consultants and Machine Learning: Consultants who help companies design and deploy machine learning solutions on AWS.

Prerequisites

No specific prerequisites are required. This training is available to anyone wishing to discover AWS, but a basic knowledge of computer science or information systems can be an asset.

Training programme

Introduction to machine learning on AWS and data preparation

• Introduction to AWS services for Machine Learning: Overview of AWS services dedicated to machine learning, such as Amazon SageMaker, AWS Lambda, Amazon Rekognition, Amazon Polly, Deep Learning AMIs.
• Preparation of data for machine learning with AWS Glue, Athena and S3. Practical workshop: Creating a data preparation pipeline with AWS Glue and storing data in S3.

Training, evaluation and optimization of machine learning models

• Model training with SageMaker. Practical workshop: Training a classification model with SageMaker.
• Evaluation and optimization of models with metrics such as precision, recall, etc. Practical workshop: Optimizing a classification model in SageMaker.

Deployment of models in production and performance monitoring

• Deployment of models with SageMaker Endpoints and SageMaker Batch Transform. Practical workshop: Deployment of a model in real time with SageMaker Endpoints.
• Model performance monitoring with SageMaker Model Monitor. Review review: Summary of key concepts and simulation of the review.

Training assets

• Pedagogical approach: An alternative between theory and practice for better assimilation of concepts.
• Qualified speakers: Specialist trainers with practical experience in the field of cloud.
• Educational tools and materials: Access to online resources, live demonstrations and real-life case studies.
• Accessibility: Training is open to all, without advanced technical prerequisites.

Pedagogical methods and tools used

• Live demonstrations with AWS cloud services.
• Practical workshops and real case studies in various sectors (industry, trade, health).
• Feedback: Sharing best practices and common mistakes in business.
• Simulations and tools: Using simulators and AWS for interactive workshops.

Evaluation

• MCQ at the end of training.
• Practical case studies.
• Continuous evaluation with personalized feedback.

Normative References

• AWS Well-Architected Framework.
• GDPR (General Data Protection Regulation).
• CCPA (California Consumer Privacy Act).
• HIPAA (Health Insurance Portability and Accountability Act).
• ISO 27001, SOC 2 (Service Organization Control).
• PCI-DSS (Payment Card Industry Data Security Standard).
• NIST Cybersecurity Framework.