DP-100 Designing and implementing a data science solution on Azure
Certification: Microsoft Certified: Azure Data Scientist Associate
Dynamics Edge courses and labs are enhanced Instructor-Led Training (ILT) materials designed specifically for live, guided instruction and follow a structured curriculum.
Our materials are intentionally different from Microsoft Learn paths in both structure and flow to better prepare for actual work, answer questions, real-time engagement, and deeper learning. Microsoft Learn paths are self-paced study resources.
You will Learn:
Create and Manage Azure Machine Learning Workspaces
Prepare and Ingest Data for Machine Learning
Use Automated Machine Learning (AutoML)
Train and Track Models Using Scripts and MLflow
Perform Hyperparameter Tuning and Model Optimization
Deploy and Monitor Machine Learning Models
Apply Responsible AI Principles
Course outline
Module 1: Explore Azure Machine Learning Workspace Resources and Assets
Create an Azure Machine Learning workspace
Identify workspace resources and their purposes
Explore Azure Machine Learning assets (datasets, models, environments)
Train and manage models within the workspace
Understand asset tracking and versioning
Module 2: Explore Developer Tools for Workspace Interaction
Use Azure Machine Learning Studio for no-code interaction
Access and manage resources using the Python SDK
Automate tasks using the Azure CLI
Understand tool interoperability within the workspace
Choose the right tool based on use case
Module 3: Make Data Available in Azure Machine Learning
Understand the role of URIs in data access
Create and configure datastores for data storage
Create reusable data assets for experiments
Manage dataset versioning and access control
Organize data for scalable ML workflows
Module 4: Work with Compute Targets in Azure Machine Learning
Choose between compute instance, cluster, or attached compute
Create and use a development compute instance
Scale workloads using compute clusters
Optimize resource allocation for training jobs
Monitor compute usage and performance
Module 5: Work with Environments in Azure Machine Learning
Understand Azure ML environments and dependencies
Use curated environments for rapid experimentation
Create custom environments with Conda or Docker
Register and reuse environments across jobs
Manage environment versions for reproducibility
Module 6: Find the Best Classification Model with Automated Machine Learning
Preprocess data and configure featurization settings
Launch an AutoML classification experiment
Track progress and results through the UI and SDK
Evaluate models using built-in metrics
Select the best-performing model for deployment
Module 7: Track Model Training in Jupyter Notebooks with MLflow
Configure MLflow for local and remote tracking
Train models in notebooks using MLflow APIs
Log parameters, metrics, and artifacts
View experiment results and history
Use MLflow UI for model comparison
Module 8: Run a Training Script as a Command Job in Azure Machine Learning
Convert a Jupyter notebook to a Python script
Submit a script as a command job
Pass parameters and inputs to the training job
Monitor job execution and outputs
Use scripts for repeatable and scalable ML runs
Module 9: Track Model Training with MLflow in Jobs
Log metrics and artifacts using MLflow
View job metrics in Azure Machine Learning UI
Evaluate model performance across runs
Compare jobs and select optimal models
Integrate MLflow with sweep jobs and pipelines
Module 10: Perform Hyperparameter Tuning with Azure Machine Learning
Define hyperparameter search space
Choose a sampling strategy (random, grid, Bayesian)
Set early termination policies for efficiency
Launch sweep jobs to optimize model parameters
Analyze tuning results and select best model
Module 11: Run Pipelines in Azure Machine Learning
Create reusable pipeline components
Chain components together into a pipeline
Submit pipeline jobs for orchestration
Manage pipeline versions and dependencies
Monitor pipeline runs and outputs
Module 12: Register an MLflow Model in Azure Machine Learning
Log trained models with MLflow
Understand the MLflow model format and structure
Register a model in the Azure Machine Learning registry
Track model versions and metadata
Prepare models for deployment and monitoring
Module 13: Create and Explore the Responsible AI Dashboard
Understand the importance of Responsible AI
Create the Responsible AI dashboard for a trained model
Evaluate model fairness, explainability, and error analysis
Interpret dashboard insights for decision-making
Use the dashboard to improve model transparency
Module 14: Deploy a Model to a Managed Online Endpoint
Understand managed online endpoint architecture
Deploy an MLflow model to an online endpoint
Deploy custom models to endpoints
Test and validate endpoint functionality
Monitor endpoint status and usage
Module 15: Deploy a Model to a Batch Endpoint
Understand batch inference and use cases
Create and configure batch endpoints
Deploy MLflow or custom models to batch endpoints
Invoke batch jobs and analyze output
Troubleshoot batch deployment issues
Module 16: Introduction to Azure AI Foundry
Define what Azure AI Foundry is
Explore how Azure AI Foundry integrates with ML workflows
Understand AI Foundry’s capabilities in managing models
Identify when and why to use AI Foundry
Compare AI Foundry to other model management tools
Module 17: Explore and Deploy Models from the Model Catalog
Navigate the Azure AI Foundry model catalog
Explore available foundation and language models
Deploy catalog models to endpoints
Enhance model performance using tuning options
Monitor usage and outcomes of deployed models
Module 18: Get Started with Prompt Flow
Understand the development lifecycle for LLM apps
Explore core components and flow types
Connect to data sources using prompt flow connections
Configure runtimes and variants
Monitor and iterate on flow performance
Module 19: Build a RAG-Based Agent with Your Own Data
Understand Retrieval-Augmented Generation (RAG) principles
Prepare and index your data for searchability
Ground a language model using your own data
Build an agent using prompt flow in Azure AI Foundry
Evaluate grounded model outputs for accuracy
Module 20: Fine-Tune a Language Model
Identify when fine-tuning is appropriate
Prepare training data for fine-tuning chat models
Use Azure AI Studio to configure fine-tuning jobs
Deploy fine-tuned models to endpoints
Monitor fine-tuning metrics and results
Module 21: Evaluate the Performance of Generative AI Apps
Set benchmarks for model evaluation
Perform manual evaluations of model output
Use metrics to assess model quality and responsiveness
Compare app iterations to improve performance
Integrate evaluation into the development cycle
Module 22: Responsible Generative AI
Plan a responsible generative AI solution lifecycle
Identify and document potential harms (bias, toxicity, etc.)
Define harm measurement strategies
Apply mitigation techniques in model training and deployment
Implement governance and operations for responsible AI