Azure ML Studio

Azure ML
Studio
A NO-CODE SOLUTION TO
MACHINE LEARNING
10TH APRIL 2020
VIKAS SINHA

Contents
What is Azure Machine Learning
Introduction to Azure ML Studio
Practical Experiment (Creating a simple ML model, train and deploy)
Certifications
Q and A

Objective and Usages
The aim of this presentation is to introduce you to the Azure ML Designer / Azure ML Studio to create ML models and
NOT TO discuss the concepts of ML and their models in detail.
Enable anyone who is interested in Machine Learning or Data Analysis.
Enable anyone who wants to quick start with Machine Learning with absolute ZERO coding experience.
In hardly any no time, you will be able to deploy your own Machine Learning model as a REST Service.
You may create your app using any language like Java , C# , R Shiny or Python etc. and consume your deployed model to
make your application or devices SMARTER.
You have lots of options to perform , right from running classic Machine Learning algorithms to creating and testing a
recommender system , to perform Sentiment Analysis and other textual analysis.
In today’s session I will walk you through a very simple Regression experiment and deploy it as a service.

Azure Machine Learning
Azure Machine Learning is a platform for operating machine learning workloads in the cloud.
Azure Machine Learning is a cloud-based environment which you can use to build, train, deploy, automate, manage, and track ML models.
Just to revise, Machine Learning is a field of Data Science which enables programs to learn from itself and perform predictions, trend
analysis, cluster analysis, classification , textual analysis etc. The field entirely depends upon the Mathematical concepts of
“Probability” and is not a rule based predictor.
Azure Machine Learning can be used for any kind of machine learning, from classical ML to Deep learning, supervised, and unsupervised
learning. Whether you prefer to write Python or R code or zero-code/low-code options such as the designer, you can build, train, and
track highly accurate machine learning and deep-learning models in an Azure Machine Learning Workspace.
Azure Machine Learning provides the widest range of cutting-edge machine learning tools such as -
a) Use Python notebooks and SDKs to train & deploy ML models
b) Use R Markdown and SDKs to train & deploy ML models
c) Use automated machine learning to train & deploy ML models (AutoML)
d) Use the designer's drag & drop capabilities to train & deploy (D-n-D Designer)
e) Use the machine learning CLI to train and deploy a model (ML-CLI)

Machine Learning Studio (Designer)
Machine Learning Studio is a classic tool for training , deploying , tracking and managing your ML models.
It is a complete No-Code solution.
It allows users to use drag and drop modules onto the experiment board.
You can user your existing Python or R script and contain it in a “Python Script” or “R Script” module and use it in your
experiment.
Easy to user interface for any novice.
Quick deployment of ML model as a Web Service.
Save time and resources
Leverage data science best practices
Provide agile problem-solving

Key Terms in any Azure ML
Activity
Workspace
◦ Experiments
◦ Run
◦ Run configuration
◦ Snapshot
◦ Git tracking
◦ Logging
◦ ML pipelines
◦ Models
◦ Environments
◦ Training script
◦ Estimators
◦ Endpoints
◦ Web service
◦ IoT modules
◦ Dataset & datastores
◦ Compute targets

Creating a ML Workspace
Log into portal.azure.com
Click on “Create a Resource”
Search for “Machine Learning”
Click “Create” and fill in all the details in the subsequent screen.
Kindly select “Enterprise” under “Workspace Edition” in order to use designer tool.
Click “Review + Create” and finally “Create”.

Creating a ML Workspace
You will be able to see an “activity” going on to deploy the resources under your “Notification” tab.
Once the deployment is done, you can click on “Go to Resource”
You may be navigated to “ml.azure.com”

Creating a ML Pipeline
Click on “Designer” tab to create a new ML Pipeline.
A ML pipeline consists of set of modules required to build , train , validate and deploy the ML model.
There are multiple modules available which are pre-built in Azure ML Designer such as ,

Creating a ML Pipeline
Each of these Pipeline steps has pre-built Modules in them.
For example,

Train , Test , Score and Evaluate a Model
Now we are ready to build test and deploy our model.
General steps to do so are –
1) Import Dataset
2) Exploratory Data Analysis including Visualization
3) Perform Data Manipulations like, cleaning data, removing or imputing missing values, synthetic oversampling (SMOTE),
feature selection , feature engineering etc.
4) Split the data
5) Train the data
6) Score the model and
7) Evaluate the model
8) Compare the models
9) Create the compute target
10) Submit the experiment

If is the first run, it may take up to 20 minutes for your pipeline to finish running. The default compute settings have a
minimum node size of 0, which means that the designer must allocate resources after being idle. Repeated pipeline runs will
take less time since the compute resources are already allocated. Additionally, the designer uses cached results for each
module to further improve efficiency.

Deploy the Model
Create a real-time inference pipeline.
To deploy your pipeline, you must first convert the training pipeline into a real-time inference pipeline. This process removes training modules and
adds web service inputs and outputs to handle requests.

Deploy the Model
Create an inferencing cluster.
We can select from any existing Azure Kubernetes Service (AKS) clusters or create a new one to deploy your model to.
Deploy the real-time endpoint.

Deploy the Model
Once successfully deployed, you will be able to see it under “Endpoints” palette on Left hand navigation bar.

Deploy the Model
If your service is successfully deployed and is up and running then the deployment state will be shown as “Healthy”. Below are
some more details-

Test the Service (via Endpoint)
Provide the test data and click on “Test” button to test your service endpoint.

Consume the Service (via Endpoint)
You may also want to consume the service via its endpoint and key authentication in your C# code or Python Code or R code.

Clean up the resources
If you don't plan to use anything that you created, delete the entire resource group so you don't incur any charges.
Navigate to “portal.azure.com” and click on “Resource Groups”.
Select the resource group which you want to delete.
Click on Overview and then “Delete Resource Group”.

Trainings and Certifications
https://docs.microsoft.com/en-us/learn/certifications/azure-data-scientist
https://www.udemy.com/course/machine-learning-using-azureml/
Week of AI Webinar Links:
http://aka.ms/weekofai-ondemand
http://aka.ms/weekofai2-ondemand
Some other resource:
https://www.gods-online.com/
https://www.kaggle.com/

Azure ML Studio

More Related Content

What's hot

Similar to Azure ML Studio

Recently uploaded

In this document

Azure ML Studio