Azure Machine Learning & ML Studio
Machine Learning with simplicity and power of cloud
Vinnie SainiData & AI Solution [email protected]
Azure ML & ML Studio
• Fully managed cloud service for building
Predictive Analytics solutions
• Reduces the intricacies of Machine Learning process
• Azure ML Studio is a powerful canvas for the
✓Composition of machine learning experiments
✓Subsequent operationalization
✓Consumption as machine learning web services
Azure ML Architecture
Data
Tables,
Hadoop (HDInsight),
Relational DB(Azure SQL)
Predictive Model
ML Studio
Operational Web API
Clients Interface
• Machine learning workspace
• Dataset format
• Upload data
• Prepare data
• Define features
Create an experiment
• Choose and apply Learning Algorithms
• Train and Evaluate
Train and Evaluate • Remove one model
• Convert the training experiment into a predictive experiment
• Deploy the predictive experiment as a web service
Deploy and Access web
service
Build a Data Science experiment in ML Studio
Get started with Machine Learning Studio• Initialize a Machine Learning workspace• The experiment has at least one dataset and one module• Import a number of data types into your experiment, depending on
what mechanism you use to import data :
• Upload a local file as a dataset• Use a module to import data from cloud data services• Import a dataset saved from another experiment
• Plain text (.txt)• Comma-separated values (CSV) with a header (.csv) or without (.nh.csv)• Tab-separated values (TSV) with a header (.tsv) or without (.nh.tsv)• Excel file• Azure table• Hive table• SQL database table• R object or workspace file (.RData) etc.
Metadata in ML Studio
• Explicitly specify or change the headings and data types for columns using the Edit Metadata.
Data types recognized by Machine Learning Studio:
• String
• Integer
• Double
• Boolean
• DateTime
• TimeSpan
Limits for data upload
• Modules in Machine Learning Studio support datasets of up to 10 GB of dense numerical data for common use cases.
• If a module takes more than one input, the 10 GB value is the total of all input sizes.
• For datasets that are larger than a couple GBs, upload data to Azure Storage or Azure SQL Database, or use Azure HDInsight rather than directly uploading from a local file.
ML Studio Demo
Deploy and Test the web service
Deploy the web service• Deploy as a Classic web
service
• Deploy as a New web service
Test the web service• When the web service is
accessed, the user's data enters through the Web service input module where it's passed to the Score Model module and scored.
• The model expects data in the same format as the original dataset. The results are returned to the user from the web service through the Web service output module.
Manage and Access the web service
Manage Web Services
Sign in to the Microsoft Azure Machine Learning Web Services portal
•Click Web services
•Click your web service
•Click the Dashboard
Access Web Services
•Request/Response
•Batch Execution
References
• Infographic of machine learning basics with links to algorithm examples
• How to choose algorithms for Microsoft Azure Machine Learning.
• A-Z list of Machine Learning Studio modules in Machine Learning Studio
• How to consume an Azure Machine Learning Web service
• Import training data into Machine Learning Studio
• Extend your experiment with R and Execute Python machine learning scripts in Azure Machine Learning Studio
• FAQs