Rise of hybrid training and scoring...
Transcript of Rise of hybrid training and scoring...
Accelerating adoption of AI by developers
(consuming models)
Rise of hybrid training and scoring scenarios
Push scoring/inference to the edge, cloud, on-prem
Moving high-end developers into deep learning as
non-traditional path to DS / AI dev
Growth of diverse hardware arms race across all form factors
(CPU / GPU / FPGA / ASIC / device)
Demonstrating success of transfer learning techniques
while reducing dev complexity
Data prep
Model deployment & management
Model lineage & auditing
Explain-ability
D A T A S C I E N C E & A I
C H A L L E N G E SK E Y T R E N D S
Apps + insightsSocial
LOB
Graph
IoT
Image
CRM INGEST STORE PREP & TRAIN MODEL & SERVE
Data orchestration and monitoring
Data lake and storage
Hadoop/Spark/SQL and ML
.
IoT
Azure Machine Learning
T H E A I D E V E L O P M E N T L I F E C Y C L E
Spark
SQL Server
Virtual machines
GPUs
Container services
NotebooksIDEs
Azure Machine Learning Workbench
SQL Server
Machine Learning Server
O N - P R E M I S E S
A Z U R E M A C H I N E L E A R N I N G
E D G E C O M P U T I N G
Azure IoT Edge
Experimentation and
Model Management
A Z U R E M A C H I N E L E A R N I N G S E R V I C E S T R A I N & D E P LO Y O P T I O N S
A Z U R E
H Y P E R S C A L E ,
E N T E R P R I S E - G R A D E
I N F R A S T R U C T U R E
D E V E LO P E R TO O L S &
S E R V I C E S
O P E N P L AT F O R M F O R
D ATA S C I E N C E
Hardware
Storage management
Software
T H E M L & A I P L A T F O R M
AI Applications (1st & 3rd party) Cognitive Services Bot Framework
Spark AI Batch Training DS VM SQL Server ACS
BLOB Cosmos DB SQL DB/DW ADLS
CPUs FPGA GPUs IoT
Azure Machine Learning
Model deployment & management
Machine Learning toolkits
Experimentation management,
data prep, & collaboration
CNTK
Tensorflow
ML Server
Scikit-Learn
Other Libs.
PROSE
Docker
Cloud – Spark, SQL, other engines
ML Server – Spark, SQL, VMs
Edge
Plan for Hackathon
www.aka.ms/aalab
https://docs.microsoft.com/da-dk/azure/machine-learning/preview/quickstart-installation
Plan for Hackathon
Links for Azure Machine Learning Workbench
• https://azure.microsoft.com/en-us/blog/diving-deep-into-what-s-new-with-azure-machine-learning/
• https://docs.microsoft.com/da-dk/azure/machine-learning/preview/quickstart-installation
Begin building now with the tools and platforms you know
Build, deploy, and
manage at scale
Boost productivity with
agile development
Benefit from the fastest AI developer cloud
Build, deploy, and manage at scale
B u i l d an d dep l oy eve r y where – c l ou d , o n -premi ses , edge , an d i n - da t a
Dep l oy i n m i n u tes w i t h da t a d r i ven man agemen t an d re t r a i n i n g of a l l you r mode l s
P ro to t y pe l oca l l y t hen s ca l e u p an d ou t w i t h VMs , S pa r k c l u s te r s , an d GP U s
Rea l t i me , h i gh t h rou gh -pu t i n s i ght s eve r y where , i n c l u d i n g E xce l i n tegr a t i on
Deployment and management of models as HTTP services
Container-based hosting of real time and batch processing
Management and monitoring through Azure
(e.g., AppInsights)
First class support for SparkML, Python, CNTK, TF, TLC, R,
extensible to support others (Caffe, MXnet)
Service authoring in Python and .NET Core
Manage models
DOCKER
Single node deployment (cloud/on-prem)
Azure Container Service
Azure IoT Edge
Spark clusters
Deploy everywhere
Consume Azure Machine Learning models in Excel
Boost productivity with agile development
S pen d more t i me mode l i n g an d l e s s t i me prepp i n g w i t h bu i l t - i n i n te l l i gen t da t a wr an g l i n g
I n c rease co l l abor a t i on an d sha r i n g w i t h n otebook s an d G i t
Avo i d l o s i n g an y t h i n g w i t h ve r s i on con t ro l an d reprodu c i b i l i t y
Kn ow t he bes t pe r fo r mi n g mode l s w i t h met r i c s , l i n eage , r u n h i s to r y, a s se t man agemen t , an d more
Sample, understand, and prep data rapidly
Leverage PROSE SDK and more for intelligent, data prep by example
Extend/customize transform and featurize through Python
Generate Python and PySpark for execution at scale
Prep data
Experiment
Manage job for local and cloud experiments
Find support for Spark + Python + R (roadmap)
Execute jobs locally, on remote VMs (scale up),on Spark clusters (scale out), or SQL on-premises
Create with Git-backed experimentation tracking of code, config, parameters, and data
Discover and compare with detailed historical metadata
Begin building now with the tools and platforms you know
Choose be t ween v i su a l d r ag -an d-drop o r code - f i r s t au t hor i n g
U se you r f avo r i te I DE
Ca l l Azu re Mach i n e Lea r n i n g se r v i ces d i rec t l y i n VS Code*
B u i l d on an y f r amewor k o r l i b r a r y w i t h t he mos t popu l a r l an gu ages
Tr a i n qu i cke r an d eas i e r w i t h i n du s t r y -l ead i n g S par k an d GP Us
VISUAL DRAG -AND-DROP CODE-F IRST
Build as you like
Use your favorite IDE
Leverage all types of data
Use what you want
U S E T H E M O S T P O P U L A R I N N O VAT I O N S
U S E A N Y TO O L
U S E A N Y F R A M E W O R K O R L I B R A R Y
Integrated into your IDE
Visual Studio Code extension (more IDEs and notebooks on roadmap)
Begin building with your IDE - no extra tool needed
Integrated rich authoring for machine learning and deep learning
Call Azure Machine Learning services straight from your IDE or notebook
Plan for Hackathon
www.aka.ms/aalab
https://docs.microsoft.com/da-dk/azure/machine-learning/preview/quickstart-installation
Plan for Hackathon
Links for Azure Machine Learning Workbench
• https://azure.microsoft.com/en-us/blog/diving-deep-into-what-s-new-with-azure-machine-learning/
• https://docs.microsoft.com/da-dk/azure/machine-learning/preview/quickstart-installation
Try it for free
Lea r n more
St a r t n ow