Data Science in the cloud with Microsoft Azure
Download as pptx, pdf2 likes889 views
The document outlines the significance of data science and its integration with Microsoft Azure, detailing the platform's services that support data science, such as scalable storage and machine learning tools. It provides a worked example involving NYC taxi data, showcasing various predictive analytics techniques. The conclusion emphasizes Azure's flexibility, low management overhead, and integration of both Microsoft and open-source tools for effective data science solutions.
Data Science in the cloud with Microsoft Azure
- 1. Data Science in the cloud with Microsoft Azure MARTIN THORNALLEY DATA SOLUTION ARCHITECT, MICROSOFT
- 2. Introduction
- 3. Data Science Definition “Data science is an interdisciplinary field about processes and systems to extract knowledge or insights from data in various forms, either structured or unstructured, which is a continuation of some of the data analysis fields such as statistics, machine learning, data mining, and predictive analytics” https://en.wikipedia.org/wiki/Data_science
- 5. The Cloud Why does the Cloud matter for Data Science? High capacity and cost effective data storage Flexible, elastic compute capacity Ready to use technologies Choice of Infrastructure or Platform Enables Agile & DevOps Operational reliability and security Pay as you go
- 6. Microsoft Azure Cloud Platform Wide range of services covering Compute, Web & Mobile, Data & Storage, Analytics, Internet of Things & Intelligence plus many more, see http://azureplatform.azurewebsites.net/en-us/ Easy to get started, free to try for 30 days but limited spend, also MSDN licence free credits, see https://azure.microsoft.com/en- gb/free/ Comprehensive documentation and examples Global presence with many recognisable brands fully committed Huge investment and growing rapidly
- 9. NYC taxis 2013 NYC taxi trips and fares – open but non-trivial dataset 24 CSV files - 12 trip, 12 fare, 1 for each month ~20GB compressed, ~50GB uncompressed, 170+ million records medallion – vehicle identifier hack license – driver identifier passenger count pickup & dropoff – datetime, longitude, latitude trip – time and distance fare - payment type, fare amount, surcharge, mta tax, tip amount, tolls amount, total amount http://www.andresmh.com/nyctaxitrips/
- 10. Predictions Predict whether a specific journey will result in a tip – binary classification Predict what class of tip will be for a specific journey – multiclass classification Predict how much a tip will be for a specific journey – regression
- 12. Data Science Virtual Machine Create Linux and Windows virtual machines in minutes Wide range of configurations - CPU cores, memory, disks, network speeds Scale to what you need Pay only for what you use Enhance security and compliance Preloaded with full set of tools and utilities from Azure MarketPlace e.g. SQL Server 2016 Developer edition, Azure SDK, Python, R, Jupyter, etc.
- 13. Storage Accounts Massively scalable cloud storage for your applications Security-enhanced, durable, and highly available across the globe Industry-leading performance with exabytes of capacity Pay only for what you use Open, multi-platform support
- 14. HDInsight A managed Apache Hadoop, Spark, R, HBase, and Storm cloud service made easy Scale to petabytes on demand Crunch all data—structured, semi-structured, unstructured Skip buying and maintaining hardware Spin up Apache Hadoop, Spark, and R clusters in the cloud Use Excel or your favourite BI tool to visualize Hadoop data Connect on-premises Hadoop clusters with the cloud
- 15. Azure Machine Learning A fully managed cloud service that enables you to easily build, deploy, and share predictive analytics solutions. Powerful cloud based analytics, now part of Cortana Intelligence Suite Azure Machine Learning Studio includes hundreds of built-in packages and support for custom code Share your solution with the world in the Gallery or on the Azure Marketplace
- 16. The Process
- 17. Preparation & Exploration Copy data using Azcopy and decompress Inspect files and load in to RStudio Create external Hive tables and load Query over full dataset for further exploration Remove erroneous data e.g. passenger numbers, lat/long Engineer features using Hive Distance from start to finish using Haversine calculation Binary indicator for tips Tip level based on ranges for multiclass classification Downsample dataset and save as internal table for Machine Learning
- 18. Machine Learning & Deployment Import Data using Hive Query Build Training Experiments Evaluate model performance Create Predictive Experiments Publish Web Service Test Web Service Call from Excel
- 19. Next Steps To build a fully fledged enterprise solution with regular data ingestion and model execution consider the following: Data Catalog Data Factory Event Hubs & Stream Analytics Power BI Cognitive Services
- 20. Conclusion
- 21. Summary Microsoft Azure provides a wide range of technologies for Data Science activities Platform services reduce the management overhead No capacity limitations and flexible provisioning – pay as you go Choice of Open Source and Microsoft – use the best tool for the task The tools are well integrated Azure Machine Learning makes it trivial to deploy your models It’s quick and easy to get started
- 22. Getting Started Sign up for free https://azure.microsoft.com/en-gb/free/ Create a Data Science VM https://azure.microsoft.com/en-us/marketplace/partners/microsoft- ads/standard-data-science-vm/ Visit Cortana Intelligence Gallery https://gallery.cortanaintelligence.com/
- 23. Q&A
- 24. Thank You Martin Thornalley Data Solution Architect, Microsoft @mthornal [email protected] https://www.linkedin.com/in/martinthornalley