Machine Learning

Official Documentation

Service Description

Using Machine Learning, forecasts for further data can be made on the basis of existing data sets. Existing data can be divided into training data and control data. Forecast models can be created from pre-built analysis algorithms or analysis algorithms written using the R programming language. These can be trained using the training data. The forecasting accuracy achieved with the models can then be verified using the control data. The models can, in turn, be provided as a Web service. All of this is possible in an integrated graphical development environment where dataflow charts of data and process components can be created and configured by drag & drop.

Getting Started

  1. Hands-On with Azure Machine Learning
    9/30/2016, Mva
  2. Cloud-Based Machine Learning for the Developer
    5/4/2015, Video, 1:13:33
  3. Intro and Overview Azure Machine Learning
    1/27/2017, Video, 0:26:04
  4. Azure ML Hands-on-Lab
    6/1/2016, Lab
  5. Introducing Azure Machine Learning
    6/26/2016, Whitepaper
  6. Azure Data Analytics for Developers
    11/21/2016, Mva
  7. Building Recommendation Systems in Azure
    9/17/2015, Mva
  8. Microsoft Azure Essentials: Azure Machine Learning
    3/11/2016, Ebook
  9. Data Science in the Cloud with Microsoft Azure Machine Learning and R
    3/17/2016, Ebook

Latest Content

Subscribe to News about Machine Learning


Azure Documentation

1. Overview
     1.1. What's Machine Learning?
     1.2. Team Data Science Process
          1.2.1. Overview
          1.2.2. Lifecycle
          1.2.3. Walkthroughs
     1.3. Machine Learning Studio
          1.3.1. What's the Studio?
          1.3.2. Studio capabilities
          1.3.3. Infographic: ML basics
     1.4. Frequently asked questions
     1.5. What's new?
2. Get Started
     2.1. Create your first experiment
     2.2. Example walkthrough
          2.2.1. Create a predictive solution
          2.2.2. 1: Create a workspace
          2.2.3. 2: Upload data
          2.2.4. 3: Create experiment
          2.2.5. 4: Train and evaluate
          2.2.6. 5: Deploy web service
          2.2.7. 6: Access web service
     2.3. Data Science for Beginners
          2.3.1. 1: Five questions
          2.3.2. 2: Is your data ready?
          2.3.3. 3: Ask the right question
          2.3.4. 4: Predict an answer
          2.3.5. 5: Copy other people's work
     2.4. R quick start
3. How To
     3.1. Set up tools and utilities
          3.1.1. Set up environments
          3.1.2. Set up virtual machines
          3.1.3. Customize Hadoop
          3.1.4. Set up a virtual machine
      DS VM overview
      How to use the DS VM
      Provision the DS VM
      Set up Azure VM
      Set up SQL VM
      Provision Linux VM
      Use Linux VM
          3.1.5. Manage a workspace
      Deploy using ARM
      Create in another region
     3.2. Analyze business needs
          3.2.1. Technical needs
          3.2.2. Identify your scenario
     3.3. Acquire and understand data
          3.3.1. Load data into storage
      Blob storage
      Use Storage Explorer
      Use AzCopy
      Use Python
      Use SSIS
      Move to a VM
      Move to SQL database
      Load into hive tables
      Load from on-prem SQL
      Load fromSQL partition tables
          3.3.2. Import training data
      From a local file
      From online sources
      From an experiment
      Use on-prem SQL
          3.3.3. Explore and visualize data
      Prepare data
      Explore data
           Use Pandas
           Use SQL VM
           Use Hive tables
      Sample data
           Use blob storage
           Use SQL Server
           Use Hive tables
      Process data
           Access with Python
           Process blob data
           Use Azure Data Lake
           Use SQL VM
           Use data pipeline
      Process data with Spark
           Explore data
           Score models
           Advanced data exploration
           Use Scala and Spark
     3.4. Develop models
          3.4.1. Engineer and select features
      Use Pandas
      Use SQL+Python
      Use Hive queries
      TDSP feature selection
          3.4.2. Create and train models
      Experiment lifecycle management
      Manage iterations
      Use PowerShell to create models
      Select algorithms
           Choose algorithms
           Algorithm cheat sheet
           Use linear regression
           Use text analytics
      Evaluate and interpret results
           Evaluate performance
           Optimize parameters
           Interpret results
      Use R and Python
           Execute R scripts
           Author custom R modules
           Execute Python scripts
     3.5. Operationalize models
          3.5.1. Overview
          3.5.2. Deploy models
      Deploy a web service
      How it works
      Prepare for deployment
      Use external data
      Deploy in multi-regions
      Use web service parameters
      Enable logging
          3.5.3. Manage web services
      Use Web Services portal
      Manage with APIs
      Create endpoints
          3.5.4. Retrain models
      Retrain programmatically
      Retrain a Classic web service
      Retrain with PowerShell
      Retrain an existing web service
          3.5.5. Consume models
      Use Excel
      Use Excel add-in
      Use web app template
      Use Batch Pool
     3.6. Examples
          3.6.1. Sample experiments
          3.6.2. Sample datasets
          3.6.3. Customer churn example
          3.6.4. End-to-end scenarios
      Use Hadoop clusters
      Use Hadoop with 1TB
      Use SQL Server
      Use SQL Data Warehouse
4. Reference
     4.1. Code samples
     4.2. PowerShell module (New)
     4.3. PowerShell module (Classic)
     4.4. Algorithm & Module reference
     4.5. REST API reference
     4.6. Web service error codes
5. Related
     5.1. Cortana Intelligence Gallery
          5.1.1. Overview
          5.1.2. Industries
          5.1.3. Solutions
          5.1.4. Experiments
          5.1.5. Jupyter Notebooks
          5.1.6. Competitions
          5.1.7. Competitions FAQ
          5.1.8. Tutorials
          5.1.9. Collections
          5.1.10. Custom Modules
     5.2. Cortana Intelligence Partner Solutions
          5.2.1. Cortana Intelligence publishing guide
          5.2.2. Cortana Intelligence solution evaluation tool
     5.3. Cortana Analytics
          5.3.1. APIs
      Anomaly detection
      Cognitive Services
      Predictive maintenance
           Technical guide
      Vehicle telemetry
6. Resources
     6.1. Azure Roadmap
     6.2. Net# Neural Networks Language
     6.3. Pricing
     6.4. Pricing calculator
     6.5. Service updates
     6.6. Blog
     6.7. MSDN forum
     6.8. Stack Overflow
     6.9. Videos
     6.10. Get help from live chat

Online Training Content

Date Title
1/16/2017 Cortana Intelligence Suite End-to-End
1/16/2017 Azure Developer Workshop (Storage, Cognitive, ML, Stream Analytics, Containers, and Docker)
11/21/2016 Azure Data Analytics for Developers
9/30/2016 Hands-On with Azure Machine Learning
7/4/2016 Design and Implement Big Data & Advanced Analytics Solutions
3/21/2016 Building Blocks: Big Data and Machine Learning
11/2/2015 Data Science and Machine Learning Essentials
9/17/2015 Building Recommendation Systems in Azure


Tool Description


Date Title Length
7/14/2017 Applying data science and machine learning to enhance the user experience 0:03:55
7/4/2017 Cloud Tech 10 - 3rd July 2017 - Azure Machine Learning, Jenkins, Petya detection and more 0:10:45
6/27/2017 IT Anomaly Insights: Operational Insights Over Telemetry Data 1:02:05
6/26/2017 Opportunity scoring - Machine Learning model on Sales Pipelines 0:45:49
5/20/2017 Sneak Peak: Network Watcher Connectivity and Machine Learning Integration 0:17:56
5/10/2017 Using StorSimple data with services in Azure (Media Services, HDInsights, AzureML, etc.) 0:25:31
5/4/2017 Project Fortis: Accelerating UN humanitarian aid planning and response with GraphQL 0:18:05
5/4/2017 How Brainshark is making salespeople better through artificial intelligence and Microsoft HoloLens 0:20:55
5/4/2017 How Brainshark is making salespeople better through artificial intelligence and Microsoft HoloLens 0:21:28
5/4/2017 Using Visual Studio for Machine Learning 0:20:57

Page 1 of 14