AI-900 Course Azure AI Fundamentals with Virtual Simulations

Explore Azure anomaly detector's time series based anomaly detection, featuring univariate and multivariate methods, streaming and batch processing, change point detection, and REST API access, and up to 300 signals.

Understand anomaly detection and exam concepts, noting that Azure anomaly detector resource is retired; no resources after Sept 20, 2023, with retirement by Oct 1, 2026, plus a hands-on simulation.

Explore univariate anomaly detection with Azure by provisioning an anomaly detector, configuring endpoint and key, and streaming CSV data to detect and visualize anomalies in real time.

Train a multivariate anomaly detector on sensor data from optical, pump pressure, rotational, and vibration readings using Azure's anomaly detector engine to identify simultaneous abnormalities.

Discover how Azure AI Vision analyzes images and text to detect faces, read text via OCR, and apply spatial analysis, enabling image tagging, tracking, and comprehensive visual understanding.

Explore natural language processing (NLP) with Microsoft’s spark-based pipelines, covering entity recognition, text classification, sentiment scoring, summarization, translation, information extraction, tagging, and content-based retrieval.

Knowledge mining ingests structured and unstructured data—PDFs, images, videos, and audio—to enrich, explore, translate, and categorize insights with Azure Vision and language capabilities.

Learn to redo course simulations quickly by going to the assignment summary, returning to the assignment, and opening the instructions to access the simulation link.

Discover responsible AI concepts like fairness, reliability and security, privacy, inclusiveness, transparency, and accountability, aligned with official Microsoft guidance. Follow concise slides focusing on terms and definitions for exam readiness.

Understand fairness and inclusiveness in AI, ensuring equal treatment across scenarios, with Azure machine learning's fairness assessment in the responsible AI dashboard.

Enhance reliability and safety in Azure's responsible AI dashboard by ensuring systems operate as designed, resist manipulation, and using error analysis to study failures across cohorts.

Explains how Azure AI safeguards privacy and security through compliance policies, access controls, encryption, vulnerability scanning, and audit logging, and introduces smart noise and counterfeit for compliance and cyber-attack testing.

Understand transparency in AI systems by exploring global, local, and model explanations, interpretable outputs, and how stakeholders use the Azure machine learning dashboard to improve fairness and accountability.

Explore how organizations build accountability into AI systems with standards and human oversight, using MLOps to log, govern, monitor, and alert on model changes and data drift in ML lifecycle.

Create an Azure machine learning workspace in the portal by provisioning a resource group, storage, key vault, application insights, and a container registry for ML data and deployment.

Learn how regression models the relationship between features and a dependent variable to predict outcomes like sales using data and Azure learning.

Learn how to build a regression machine learning pipeline in azure machine learning studio designer to clean data, select columns, and predict price from automotive data.

Evaluate a regression pipeline in Azure Machine Learning by reviewing score data and evaluation metrics—mean absolute error, root mean squared error, and coefficient of determination for predicting price.

Identify how classification machine learning predicts an object's category using binary or multi-class labels, based on features such as points, surfaces, colors, and other variables.

Create an Azure machine learning workspace, prepare the data, and build a 70/30 split classification pipeline using a two-class logistic regression to predict income.

Learn to label a confusion matrix by distinguishing actual versus predicted, identify true positives and true negatives first, and then classify false positives and false negatives through practical examples.

Explore unsupervised clustering with k-means in Azure Machine Learning to partition objects into distinct clusters. Build a pipeline that normalizes data, trains the cluster model, and outputs a csv.

Explore datasets with labels in Azure Machine Learning, learn how to identify, rename, and select label columns in tabular data, and apply labeling within a 70/30 split pipeline.

Learn how train models drive machine learning by splitting data into training and test sets. Use Azure SDK for Python, Designer, or CLI to build, validate, and score results.

Explore automated machine learning in Azure by launching Studio, running multiple models on your data, and reviewing results to streamline model development.

Use Azure machine learning designer, an interface that lets you drag and drop components to build AI pipelines and run data through models for predictions, with assets, data, and models.

Clean up your Azure environment by deleting unused resource groups in portal.azure.com to avoid wasting Azure credits and prepare a clean workspace for the next lessons.

Create an Azure computer vision resource in the portal, configure a resource group and unique name, choose pricing, and retrieve the endpoint and key for API access.

Explore image classification and object detection in Azure Vision Studio, examining image analysis, product recognition on shelves, OCR integration, and JSON outputs for API-driven app displays.

Explore optical character recognition in Vision Studio with real time processing, sample images, and json output, including handwritten notes and nutritional facts, with the ability to upload your own image.

Explore facial recognition and facial analysis solutions with Microsoft’s Vision API in Azure AI, using Vision Studio to test face mask detection and review the Json data returned.

Discover spatial analysis in Vision Studio on portal.azure.com, using video summary and frame locator to analyze frames and detect events like forklift movements and spills.

Download, install, and create a free Postman account to interact with Azure AI services, then send API requests to Azure's AI APIs.

Configure the Azure computer vision resource, build a Postman-based request using the image analysis API with your endpoint and key, and extract coordinates for people and objects in images.

Learn to build and train custom vision service image classification models using Azure, including creating a project, uploading labeled images of trees and flowers, and evaluating predictions.

Learn to use the Azure face service API with Postman to detect faces and retrieve attributes such as head pose and glasses, including obtaining endpoint, keys, and request details.

Learn to use the Microsoft Form Recognizer for optical character recognition, obtaining endpoint and key, and calling read analyze to extract handwriting from an image.

Discover azure language services: named entity recognition, pii/phi detection, language detection, sentiment analysis, and summarization. Explore key phrase extraction, entity linking, health text analytics, and conversational language understanding for chatbots.

Create an Azure AI language service resource in the portal, using default features and a free tier, then locate the keys and endpoint.

Explore key phrase extraction in Azure language studio by running samples from a configured language service, viewing extracted phrases, and understanding the JSON output for developers.

Explore key named entity recognition in language studio by extracting name entities from samples (legal NDA, email, banking), viewing json outputs and confidence scores, and formatting results for developers.

Explore sentiment analysis in Language Studio by classifying text, analyzing a product review, and exporting results as JSON for graphical visualization.

Explore azure ai language studio's understand questions and conversational language feature, learn to answer questions by analyzing data and returning short and long answers from built-in samples.

Explore real-time speech to text and speech synthesis demos using Speech Studio, record audio with Audacity, save as wav, and verify text output with Azure speech services.

Explore speech translation in Azure AI via Speech Studio, translating an English audio clip to Spanish, and viewing the original text and JSON data.

Explore Azure language service access, keys and endpoints, and interact via rest api, Postman, or Docker, then craft json requests for language detection and testing.

Explore the Azure speech service, create a speech resource, and learn to perform speech to text using REST API endpoints and keys, Speech Studio, and Postman.

Learn how to set up the Azure translator service, create a translator resource, obtain keys and endpoint, and perform a translate post via the REST API.

Configure Azure AI language to support questions and answers by creating and linking a cognitive search service, building a knowledge base, deploying it, and testing with Postman.