AI & ML Cloud Deployment For Beginners

Explore cloud computing basics with the Azure platform, learn Microsoft Azure services, and apply artificial intelligence using automated machine learning in the Azure Machine Learning Designer.

Explore cloud computing fundamentals by examining compute, networking, storage, and analytics delivered over the internet. Learn about public, private, and hybrid cloud, and IaaS, PaaS, SaaS with the Azure platform.

Explore Azure AI services for creating, managing, and publishing machine learning models, covering anomaly detection, computer vision, natural language processing, and conversational experiences with automated machine learning and a workflow.

Explore automated machine learning in Microsoft Azure, creating an Azure machine learning workspace, defining a resource group, provisioning compute resources, preparing data, training models, and deploying as a service.

Create and configure a machine learning workspace on Azure, including resource groups, storage, and compute resources; launch Azure Machine Learning Studio to run notebooks, automated demos, and designer pipelines.

Create and configure compute resources on Azure Machine Learning, including compute instance, compute cluster, and inference cluster, for training, validation, and automated machine learning tasks.

Explore a bike rental dataset on Azure to prepare a regression model by inspecting data through dataset creation, schema details, and profiling, including distribution, missing values, and normalization.

Learn to train a regression model using automated machine learning on Azure, configuring an experiment with the bike rentals dataset, and evaluate models with metrics and explainability.

Deploy a trained model as a real-time service on Azure using a container service, then test the predictive service via a REST endpoint to obtain a score.

Created an automated machine learning workspace and compute resources, including a compute instance and a compute cluster. Explored dataset grained regression machine learning model and deployed it as a service.

Explore setting up a machine learning workspace, configuring compute resources, exploring data, building a regression based model, and deploying it as a service with machine learning designer and inference pipeline.

Explore the mobile price dataset in Azure machine learning designer by examining distributions, missing values, and data transformations to prepare a regression model with price as the target.

Select relevant columns, clean missing values, and normalize continuous features with min-max to prepare data for training a machine learning model.

Create and run a training pipeline on Microsoft Azure by performing a 70/30 random data split for a linear regression model, and begin building an inference pipeline.

Evaluate the model before placing it on pipeline, using an evaluation model that reports absolute error, squared error, relative error, relative absolute error, and proportion of determination on Microsoft Azure.

Create an inference pipeline for real-time predictions on unseen data, adapting the training pipeline, removing irrelevant features, and preparing the model to serve as a predictive service.

Publish a predictive service for client applications by applying and testing the service, then deploy a real-time inference endpoint and clean up resources to reduce compute costs.

Test a predictive service by querying the latest Azure deployment endpoint with the endpoint key to obtain a price prediction and validate the deployed regression model.

Explore cloud computing, artificial intelligence, and Microsoft Azure, with emphasis on Azure Machine Learning designer, automated machine learning, and regression use cases.

Explore deep learning on Microsoft Azure, covering deep networks, convolutional neural networks, transfer learning, and use cases on Azure Machine Learning Designer, with insights on segmentation using Microsoft Media.

Explore deep learning, a subset of machine learning inspired by the brain, where a deep neural network uses activations across connected layers with optimizers to produce regression or classification outputs.

Learn how optimizers like stochastic gradient descent and mini-batch gradient descent adjust weights and learning rate to optimize training, and how learning rate affects training speed and stability.

Train a deep neural network through forward propagation, loss calculation, backpropagation, and multiple epochs to improve model performance on Azure using notebooks and a machine learning workspace.

Implement a deep neural network with TensorFlow in Azure ML, using a GPU compute instance to train a three-class penguin classifier to about 98% accuracy and save the model.

Explore convolutional neural networks for image analysis, where convolutional layers extract features from images, activation functions process feature maps, pooling reduces dimensionality, and fully connected layers produce predictions.

Train a convolutional neural network on Azure, using TensorFlow, to classify circles, squares, and triangles from the Florida dataset. Use image data generators and dropout.

Explore transfer learning in CNNs for image classification, where pre-trained feature extractors drive a new fully connected classifier to map features to class probabilities.

Transfer learning with a TensorFlow cnn on Azure uses a pretrained base without the top layer, trained on 24x24 circle, square, and crime lab images.

delete the compute resource, the machine learning workspace, and the resource group after the experiment to avoid extra costs, including application insights, keyboard, and storage account.

Explore deep learning, a subset of machine learning, with neural networks, input and output layers, epochs, and optimizers, and learn convolutional networks for computer vision and transfer learning with Azure.

Explore computer vision as a key workload in Microsoft Azure, with introductions to cognitive services and practical coverage of image classification, the custom vision service, and object detection.

Explore computer vision, a core AI field that lets systems interpret images and videos and perform tasks such as image classification, object detection, semantic segmentation, and OCR, on Azure.

Learn to deploy computer vision on microsoft azure using computer vision and cognitive services, obtain api keys and endpoints, and manage resources, subscriptions, and pricing for image analysis.

Learn to create image classification models with Azure Custom Vision, upload and label images, train and evaluate, then publish with a prediction endpoint and key.

Explore image classification with Custom Vision in ai ml cloud deployment for beginners, building a three-class model for apples, bananas, and oranges through tagging and training, then publishing for deployment.

Build a custom vision project, tag images, train the model, and deploy with endpoint and prediction keys for object detection on Azure.

Builds on computer vision cognitive services to apply image classification, object detection, and custom vision use cases, with next topics on image analysis, optical character recognition, and form recognizer.

Explore computer vision concepts like image analysis and optical character recognition, and implement them on Microsoft Azure to analyze images and receipts with form recognizer subsets.

Explore how the Microsoft Azure computer vision service analyzes images to describe scenes, detect objects and faces, generate tags and metadata, and identify celebrities and local landmarks.

Learn to detect and analyze faces with the face service and cognitive services, including bounding boxes, facial landmarks, age and emotion, and find or verify similar faces.

Read text with computer vision using OCR and Read APIs to detect printed or handwritten text. Capture bounding boxes and digitize documents across languages, including medical records and historical documents.

Explore how the form recognizer service on Microsoft Azure analyzes receipts with OCR, extracting date, merchant, address, phone, and totals, and supports pre-built and custom models for automated data extraction.

Analyze images with computer vision to detect places, read text via OCR, and analyze receipts with form recognizer in Azure, then complete a final notebook project using a public dataset.

Explore natural language processing basics, analyze text with language services, and recognize and synthesize speech, then translate text and speech and build a language model using conversation language understanding (Lewis).

Explore natural language processing basics and its Azure use cases, including text and speech understanding, document classification, summarization, sentiment analysis, and semantic language modeling.

Explore four Azure NLP workloads—text analytics, speech, slicker text, and language understanding—covering language prediction, key phrase extraction, entity prediction, sentiment analysis, speech-to-text, text translation, and a custom language model.

Demonstrate a natural language processing workflow with text analytics, sentiment analysis, and entity extraction, showing how big entity search surfaces planets like Saturn and Jupiter.

Learn to analyze text with language services, applying natural language processing techniques—text analytics, sentiment analysis, entity recognition, and semantic modeling—using Microsoft Azure cognitive services.

Set up Azure cognitive services and Azure machine learning compute instance to run text analytics on reviews, then detect language, extract key phrases, and gauge sentiment with named entities.

Explore how speech recognition converts spoken input to text and speech synthesis turns text into audible speech using Azure speech service, with real-time and batch transcription and personalized voices.

Implement speech recognition and speech synthesis with Microsoft Azure cognitive services, handling microphone or audio file input, and demonstrate light on/off commands using the location instead of the endpoint.

Discover how speech recognition detects spoken input and speech synthesis generates spoken output through Microsoft speech cognitive services, using speech to text and text to speech APIs.

Explore translating text and speech using azure cognitive services, building on speech recognition and speech synthesis with speech-to-text and text-to-speech APIs.

Learn how text translation enables cross-language document, email, and web page translation, and how speech translation converts spoken language using Microsoft Azure across more than 60 languages.

Master Microsoft Azure cognitive services for text and speech translation, including profanity filtering and selective translations, with a single resource and endpoint for streamlined access.

Explore conversational language understanding with azure language service and Lewis, harness text and speech translation via cognitive services, including speech to text and text to speech, plus a lighting demo.

Implement conversational language understanding on Microsoft Azure by creating a language understanding or cognitive service, defining entities, intents, and utterances, training, testing, and publishing the model.

Discover how conversational AI integrates language services with Azure, review core concepts of utterances, entities and index in Lewis, and see how predefined domains and content support building conversational experiences.

Learn how conversational artificial intelligence enables automated support with chatbots that convert documentation into a knowledge base of question-and-answer pairs, using natural language processing to interpret varied questions across channels.

Learn to provision a language service, build and test a knowledge base with custom question answering, and deploy a chat bot on Azure using the Bot Framework SDK across channels.

Build a chatbot on Azure by creating a Q&A maker knowledge base, connecting a bot service, training and testing it, then publishing and embedding the web chat channel.

Delete the resource group and its resources to avoid extra billing and automatically remove the Azure Machine Learning workspace, prediction resources, Q&A resources, and related workspace components.

Build a conversational chatbot using existing epic documentation, create a knowledge base from question‑and‑answer pairs, and connect it to Microsoft Azure for a customized bot.