az-cloudscale-adv-analytics

Azure Hands-on Lab for Cloud Scale Adv. Analytics

View on GitHub

Cloud Scale Advanced Analtyics

Experience end-to-end Advanced Analytics on cloud using Blob, ADF, Databricks, SQLDB and Azure Machine Leaning Studio.

In this hands on lab, you can understand how to apply following Azure services to your project

After the workshop you will be able to:

  1. Understand process and architecture for cloud scale andvanced analaytics project
  2. Create appropreate Azure services for data prep. & training environment
  3. Know how to wanggle data in a scale
  4. Expermiments on data and select the best model
  5. Deploy and interact with your score model

Architecture

overallarch

Scenario

Extract data from a web and load the data to Azure Blob storage. Mount the Azure Blob storage to Azure Databricks to prepare the data for Machine Learning. When save prepared data on Azure Blob, access the prepared data from Azure Machine Learning Studio. Conduct machine learning experiments and select the best model for prediction. When a model is selected, deploy the score model as a web service from the Azure Machine Learning Studio. Lastly, extract new data set from SQL Database and

Lab 00. Create hands on lab environment

  1. Create Hands-on Lab envrironment using a script

Lab 01. Ingest

  1. Create Azure Data Factory (v2)

  2. Create Data Pipeline

Lab 02. Create Azure Databricks

  1. Create Azure Databricks

  2. Create Azure Databricks cluster

  3. Import and Run Notebook

Labs for Data Scientist

Lab 03. Data Wrangling and Training

Import following url from Azure Databricks

https://github.com/xlegend1024/az-adb-aml-lab/raw/master/databricks/az-adb-aml-lab.dbc

Lab 04. Run Notebook within ADF Pipeline

Use Azure Data Factory for batch scoring

Lab 05. Operationalize score model using Azure Machine Leaerning

Import following url from Azure Databricks

https://github.com/xlegend1024/az-cloudscale-adv-analytics/blob/master/AzureDatabricks/07_MLlib_Classification_Deployment.ipynb

Lab 06. Azure AutoML

Import following url from Azure Databricks

https://github.com/xlegend1024/az-adb-aml-lab/raw/master/databricks/az-adb-aml-lab.dbc

Labs for Citizenship Data Scientist

Start Lab > 01. Ingest Data

Sources and references