SlideShare a Scribd company logo
1 of 24
Making your Machine
Learning Model
useable by others
Hello!
I am John Robert
I put machine learning in production at Adtriba.
Adtriba uses machine learning to improve marketing, optimize
budget allocation and increase revenue
@trojrobert
2
Usability
Deploy machine learning models to production
Make a product or service with machine learning
model
3
Examples of usable models
4
From ‘Attention Is All You Need’ by Vaswani et al.
Examples of usable models
Tesla Autopilot
5
Machine Learning Life Cycle
6
Source - ttps://woolpert.com/media/blogs/geospatial/what-does-a-machine-learning-project-look-like/
Usable Model can
be compared with
serving prepared
food.
7
Whole roasted scup with piperade, grilled cauliflower and bronze fennel -
https://www.tripadvisor.com/LocationPhotoDirectLink-g41790-d11620864-
i325210489-Feather_Wedge-Rockport_Cape_Ann_Massachusetts.html
8
Source and prepare high quality data Train model Deploy model
Source and prepare high quality ingredients Cook a meal Serve the meal
AI system
“Plates” used in serving/deploying ML models
9
“Table” for serving/ deploying ML Models
10
Source - https://allcode.com/cloud-providers/
AI Systems = Model(code) + Data
DevOps
MLOps
Normal system = Code
11
Source - https://towardsdatascience.com/challenges-deploying-machine-learning-models-to-production-ded3f9009cb3
Process of serving/deploying ML Models
12
Serving/ deploying traditional systems(DevOps)
13
Tools for serving /deploying ML Models
14
create and share beautiful, custom web apps for machine learning
and data science
is continuous integration for machine learning. Bring DevOps
practices to your projects for automatic, reproducible, and fast
machine learning.
Allow you to manage and deploy models from a variety of ML
libraries to a variety of model serving and inference platforms.
end-to-end enterprise AI platform that automates and accelerates
every step of your path from data to value.
end-to-end machine learning platform to build and deploy AI models
at scale
making deployments of machine learning (ML) workflows on
Kubernetes simple, portable and scalable.
Tools for serving /deploying ML Models
15
Example of served/deployed ML Model
16
Roadblock of deploying ML model (development)
Data Leakage
happens when your
training data contains
information about the
target, but similar
data will not be
available when the
model is used for
prediction.
17
Roadblocks when serving/deploying ML Model
Data Drift
is the change in input
data distribution of
the model.
Prediction Drift
is the change in
output of the model.
For instance change
in target data
Covariate Drift
is the change in
pattern learnt by the
model. Change in the
relationship between
the features and the
target
18
Model Degradation
Roadblocks when serving/deploying ML Model
19
Source - MIT 6.S191: Evidential Deep Learning and Uncertainty
Roadblocks when serving/deploying ML Model
Out Of Distribution
this occur when you try to
predict a class that the
model was not trained on
20
What will the model predict when the input image is a
car?
Roadblocks when serving/deploying ML Model
Infrastructure
21
Source - https://severalnines.com/database-
blog/scaling-postgresql-large-amounts-data
Source - https://akfpartners.com/growth-blog/what-is-latency
When your model is too big for the plate
Quantization
Quantization is
reducing the bitwidths
of weights in a model.
For instance convert
values from float to
integer.
Pruning
reducing the size of a
machine learning
model. For instance
removing some
layers.
22
Model compression
Preserve your served/ deployed ML model
Retraining
Since the world is
changing, data
generated is also
changing. Models
need to adapt to the
changes.
Monitoring
Track and analyse
data and model
performance
23
Data and Model Health (DataRobot)
John Robert: Making your machine learning model usable by others

More Related Content

What's hot

Using MLOps to Bring ML to Production/The Promise of MLOps
Using MLOps to Bring ML to Production/The Promise of MLOpsUsing MLOps to Bring ML to Production/The Promise of MLOps
Using MLOps to Bring ML to Production/The Promise of MLOps
Weaveworks
 

What's hot (20)

Kyrylo Perevozchykov "Continuous delivery for Machine Learning, the future of...
Kyrylo Perevozchykov "Continuous delivery for Machine Learning, the future of...Kyrylo Perevozchykov "Continuous delivery for Machine Learning, the future of...
Kyrylo Perevozchykov "Continuous delivery for Machine Learning, the future of...
 
Apache Liminal (Incubating)—Orchestrate the Machine Learning Pipeline
Apache Liminal (Incubating)—Orchestrate the Machine Learning PipelineApache Liminal (Incubating)—Orchestrate the Machine Learning Pipeline
Apache Liminal (Incubating)—Orchestrate the Machine Learning Pipeline
 
Getting Started with Azure AutoML
Getting Started with Azure AutoMLGetting Started with Azure AutoML
Getting Started with Azure AutoML
 
What is MLOps
What is MLOpsWhat is MLOps
What is MLOps
 
Productionising Machine Learning Models
Productionising Machine Learning ModelsProductionising Machine Learning Models
Productionising Machine Learning Models
 
Olivier Blais: Want to adopt AI in your business: good luck!
Olivier Blais: Want to adopt AI in your business: good luck!Olivier Blais: Want to adopt AI in your business: good luck!
Olivier Blais: Want to adopt AI in your business: good luck!
 
Using MLOps to Bring ML to Production/The Promise of MLOps
Using MLOps to Bring ML to Production/The Promise of MLOpsUsing MLOps to Bring ML to Production/The Promise of MLOps
Using MLOps to Bring ML to Production/The Promise of MLOps
 
Tableau and PowerPivot jenstirrup
Tableau and PowerPivot jenstirrupTableau and PowerPivot jenstirrup
Tableau and PowerPivot jenstirrup
 
Machine learning life cycle
Machine learning life cycleMachine learning life cycle
Machine learning life cycle
 
201906 01 Introduction to ML.NET 1.0
201906 01 Introduction to ML.NET 1.0201906 01 Introduction to ML.NET 1.0
201906 01 Introduction to ML.NET 1.0
 
Pm.ais ummit 180917 final
Pm.ais ummit 180917 finalPm.ais ummit 180917 final
Pm.ais ummit 180917 final
 
An Exploratory Study on Machine Learning Model Stores
An Exploratory Study on Machine Learning Model StoresAn Exploratory Study on Machine Learning Model Stores
An Exploratory Study on Machine Learning Model Stores
 
MLSEV Virtual. ML Platformization and AutoML in the Enterprise
MLSEV Virtual. ML Platformization and AutoML in the EnterpriseMLSEV Virtual. ML Platformization and AutoML in the Enterprise
MLSEV Virtual. ML Platformization and AutoML in the Enterprise
 
Norman Sasono - Incorporating AI/ML into Your Application Architecture
Norman Sasono - Incorporating AI/ML into Your Application ArchitectureNorman Sasono - Incorporating AI/ML into Your Application Architecture
Norman Sasono - Incorporating AI/ML into Your Application Architecture
 
Human-Centered AI: Scalable, Interactive Tools for Interpretation and Attribu...
Human-Centered AI: Scalable, Interactive Tools for Interpretation and Attribu...Human-Centered AI: Scalable, Interactive Tools for Interpretation and Attribu...
Human-Centered AI: Scalable, Interactive Tools for Interpretation and Attribu...
 
2021 02 23 MVP Fusion Getting Started with Machine Learning.Net and AutoML
2021 02 23 MVP Fusion Getting Started with Machine Learning.Net and AutoML2021 02 23 MVP Fusion Getting Started with Machine Learning.Net and AutoML
2021 02 23 MVP Fusion Getting Started with Machine Learning.Net and AutoML
 
201906 02 Introduction to AutoML with ML.NET 1.0
201906 02 Introduction to AutoML with ML.NET 1.0201906 02 Introduction to AutoML with ML.NET 1.0
201906 02 Introduction to AutoML with ML.NET 1.0
 
Machine Learning API'S By Mushahid Ali
Machine Learning API'S By Mushahid AliMachine Learning API'S By Mushahid Ali
Machine Learning API'S By Mushahid Ali
 
Productionizing Machine Learning in Our Health and Wellness Marketplace
Productionizing Machine Learning in Our Health and Wellness MarketplaceProductionizing Machine Learning in Our Health and Wellness Marketplace
Productionizing Machine Learning in Our Health and Wellness Marketplace
 
Data Science Salon: Kaggle 1st Place in 30 minutes: Putting AutoML to Work wi...
Data Science Salon: Kaggle 1st Place in 30 minutes: Putting AutoML to Work wi...Data Science Salon: Kaggle 1st Place in 30 minutes: Putting AutoML to Work wi...
Data Science Salon: Kaggle 1st Place in 30 minutes: Putting AutoML to Work wi...
 

Similar to John Robert: Making your machine learning model usable by others

Cloud based Machine Learning Platforms, a review - Sagar Khashu
Cloud based Machine Learning Platforms, a review - Sagar KhashuCloud based Machine Learning Platforms, a review - Sagar Khashu
Cloud based Machine Learning Platforms, a review - Sagar Khashu
Sagar Khashu
 
Machine Learning Platformization & AutoML: Adopting ML at Scale in the Enterp...
Machine Learning Platformization & AutoML: Adopting ML at Scale in the Enterp...Machine Learning Platformization & AutoML: Adopting ML at Scale in the Enterp...
Machine Learning Platformization & AutoML: Adopting ML at Scale in the Enterp...
Ed Fernandez
 
Model Monitoring at Scale with Apache Spark and Verta
Model Monitoring at Scale with Apache Spark and VertaModel Monitoring at Scale with Apache Spark and Verta
Model Monitoring at Scale with Apache Spark and Verta
Databricks
 

Similar to John Robert: Making your machine learning model usable by others (20)

Introducing MLOps.pdf
Introducing MLOps.pdfIntroducing MLOps.pdf
Introducing MLOps.pdf
 
Cloud based Machine Learning Platforms, a review - Sagar Khashu
Cloud based Machine Learning Platforms, a review - Sagar KhashuCloud based Machine Learning Platforms, a review - Sagar Khashu
Cloud based Machine Learning Platforms, a review - Sagar Khashu
 
Machine Learning Platformization & AutoML: Adopting ML at Scale in the Enterp...
Machine Learning Platformization & AutoML: Adopting ML at Scale in the Enterp...Machine Learning Platformization & AutoML: Adopting ML at Scale in the Enterp...
Machine Learning Platformization & AutoML: Adopting ML at Scale in the Enterp...
 
artificggggggggggggggialintelligence.pdf
artificggggggggggggggialintelligence.pdfartificggggggggggggggialintelligence.pdf
artificggggggggggggggialintelligence.pdf
 
DutchMLSchool. ML Automation
DutchMLSchool. ML AutomationDutchMLSchool. ML Automation
DutchMLSchool. ML Automation
 
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...
 
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)
 
Human-Centered Interpretable Machine Learning
Human-Centered Interpretable  Machine LearningHuman-Centered Interpretable  Machine Learning
Human-Centered Interpretable Machine Learning
 
Notes on Deploying Machine-learning Models at Scale
Notes on Deploying Machine-learning Models at ScaleNotes on Deploying Machine-learning Models at Scale
Notes on Deploying Machine-learning Models at Scale
 
A case study in using ibm watson studio machine learning services ibm devel...
A case study in using ibm watson studio machine learning services   ibm devel...A case study in using ibm watson studio machine learning services   ibm devel...
A case study in using ibm watson studio machine learning services ibm devel...
 
2020 11 19 MVP Days Israel 2020 - Introduction to Machine Learning.Net and Au...
2020 11 19 MVP Days Israel 2020 - Introduction to Machine Learning.Net and Au...2020 11 19 MVP Days Israel 2020 - Introduction to Machine Learning.Net and Au...
2020 11 19 MVP Days Israel 2020 - Introduction to Machine Learning.Net and Au...
 
DutchMLSchool 2022 - Automation
DutchMLSchool 2022 - AutomationDutchMLSchool 2022 - Automation
DutchMLSchool 2022 - Automation
 
2021 06 19 ms student ambassadors nigeria ml net 01 slide-share
2021 06 19 ms student ambassadors nigeria ml net 01   slide-share2021 06 19 ms student ambassadors nigeria ml net 01   slide-share
2021 06 19 ms student ambassadors nigeria ml net 01 slide-share
 
2024-02-24_Session 1 - PMLE_UPDATED.pptx
2024-02-24_Session 1 - PMLE_UPDATED.pptx2024-02-24_Session 1 - PMLE_UPDATED.pptx
2024-02-24_Session 1 - PMLE_UPDATED.pptx
 
Why do the majority of Data Science projects never make it to production?
Why do the majority of Data Science projects never make it to production?Why do the majority of Data Science projects never make it to production?
Why do the majority of Data Science projects never make it to production?
 
Model Monitoring at Scale with Apache Spark and Verta
Model Monitoring at Scale with Apache Spark and VertaModel Monitoring at Scale with Apache Spark and Verta
Model Monitoring at Scale with Apache Spark and Verta
 
[DSC Adria 23] Antoni Ivanov Practical Kimball Data Patterns.pptx
[DSC Adria 23] Antoni Ivanov Practical Kimball Data Patterns.pptx[DSC Adria 23] Antoni Ivanov Practical Kimball Data Patterns.pptx
[DSC Adria 23] Antoni Ivanov Practical Kimball Data Patterns.pptx
 
Simulators as Drivers of Cutting Edge Research
Simulators as Drivers of Cutting Edge ResearchSimulators as Drivers of Cutting Edge Research
Simulators as Drivers of Cutting Edge Research
 
“Houston, we have a model...” Introduction to MLOps
“Houston, we have a model...” Introduction to MLOps“Houston, we have a model...” Introduction to MLOps
“Houston, we have a model...” Introduction to MLOps
 
Scaling ml @ careem (oreilly ai conf)
Scaling ml @ careem (oreilly ai conf)Scaling ml @ careem (oreilly ai conf)
Scaling ml @ careem (oreilly ai conf)
 

More from Edunomica

More from Edunomica (20)

Daniel Samaan: ChatGPT and the Future of Work
Daniel Samaan: ChatGPT and the Future of WorkDaniel Samaan: ChatGPT and the Future of Work
Daniel Samaan: ChatGPT and the Future of Work
 
Fanni Kadocsa: Unlocking the power of capability building: Maximizing the imp...
Fanni Kadocsa: Unlocking the power of capability building: Maximizing the imp...Fanni Kadocsa: Unlocking the power of capability building: Maximizing the imp...
Fanni Kadocsa: Unlocking the power of capability building: Maximizing the imp...
 
Zack Johnson: Session title: People Analytics: the epicenter of management an...
Zack Johnson: Session title: People Analytics: the epicenter of management an...Zack Johnson: Session title: People Analytics: the epicenter of management an...
Zack Johnson: Session title: People Analytics: the epicenter of management an...
 
Anita Zbieg: How to make data actionable? Lessons from the teams on how to tu...
Anita Zbieg: How to make data actionable? Lessons from the teams on how to tu...Anita Zbieg: How to make data actionable? Lessons from the teams on how to tu...
Anita Zbieg: How to make data actionable? Lessons from the teams on how to tu...
 
Aizhan Tursunbayeva: The ethics of people analytics: risks, opportunities and...
Aizhan Tursunbayeva: The ethics of people analytics: risks, opportunities and...Aizhan Tursunbayeva: The ethics of people analytics: risks, opportunities and...
Aizhan Tursunbayeva: The ethics of people analytics: risks, opportunities and...
 
Kevin Martin: The New Corporate Currency
Kevin Martin: The New Corporate CurrencyKevin Martin: The New Corporate Currency
Kevin Martin: The New Corporate Currency
 
Catherine Coppinger: Why Anchor Days Are Sinking Productivity & What to Do Ab...
Catherine Coppinger: Why Anchor Days Are Sinking Productivity & What to Do Ab...Catherine Coppinger: Why Anchor Days Are Sinking Productivity & What to Do Ab...
Catherine Coppinger: Why Anchor Days Are Sinking Productivity & What to Do Ab...
 
Kevin Martin: Empowering Your Board with the People Analytics That Matter
Kevin Martin: Empowering Your Board with the People Analytics That MatterKevin Martin: Empowering Your Board with the People Analytics That Matter
Kevin Martin: Empowering Your Board with the People Analytics That Matter
 
Aizhan Tursunbayeva: Navigating Opportunities and Risks: A Responsible Approa...
Aizhan Tursunbayeva: Navigating Opportunities and Risks: A Responsible Approa...Aizhan Tursunbayeva: Navigating Opportunities and Risks: A Responsible Approa...
Aizhan Tursunbayeva: Navigating Opportunities and Risks: A Responsible Approa...
 
Alan Susi: Organizational Health: A People Team’s path to Minimum Viable ‘Wo...
Alan Susi: Organizational Health:  A People Team’s path to Minimum Viable ‘Wo...Alan Susi: Organizational Health:  A People Team’s path to Minimum Viable ‘Wo...
Alan Susi: Organizational Health: A People Team’s path to Minimum Viable ‘Wo...
 
Cole Napper: Are you ready for generative AI in people analytics?
Cole Napper: Are you ready for generative AI in people analytics?Cole Napper: Are you ready for generative AI in people analytics?
Cole Napper: Are you ready for generative AI in people analytics?
 
Fahim Karim: Attrition Prevention
Fahim Karim: Attrition PreventionFahim Karim: Attrition Prevention
Fahim Karim: Attrition Prevention
 
Taras Filatov: Building your own metaverse & NFT app
Taras Filatov: Building your own metaverse & NFT appTaras Filatov: Building your own metaverse & NFT app
Taras Filatov: Building your own metaverse & NFT app
 
Alex Poon: Should you gamify community contributions?
Alex Poon: Should you gamify community contributions?Alex Poon: Should you gamify community contributions?
Alex Poon: Should you gamify community contributions?
 
Julio Holon: Decentralised colaboration
Julio Holon: Decentralised colaborationJulio Holon: Decentralised colaboration
Julio Holon: Decentralised colaboration
 
Startup Presentation: Gaianet
Startup Presentation: GaianetStartup Presentation: Gaianet
Startup Presentation: Gaianet
 
Shawn Grubb: Minnows v. whales: Quadratic Governance to the rescue
Shawn Grubb: Minnows v. whales: Quadratic Governance to the rescueShawn Grubb: Minnows v. whales: Quadratic Governance to the rescue
Shawn Grubb: Minnows v. whales: Quadratic Governance to the rescue
 
Joachim Stroh: Hypha DAO, the 3rd generation of DAOs
Joachim Stroh: Hypha DAO, the 3rd generation of DAOsJoachim Stroh: Hypha DAO, the 3rd generation of DAOs
Joachim Stroh: Hypha DAO, the 3rd generation of DAOs
 
Vikram Aditya: Biggest Opportunity Areas in the DAOverse
Vikram Aditya: Biggest Opportunity Areas in the DAOverseVikram Aditya: Biggest Opportunity Areas in the DAOverse
Vikram Aditya: Biggest Opportunity Areas in the DAOverse
 
Tamara Helenius: The Commons are Coming
Tamara Helenius: The Commons are ComingTamara Helenius: The Commons are Coming
Tamara Helenius: The Commons are Coming
 

Recently uploaded

Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
amitlee9823
 
Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...
Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...
Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...
amitlee9823
 
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts ServiceCall Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 
Call Girls Indiranagar Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
Call Girls Indiranagar Just Call 👗 7737669865 👗 Top Class Call Girl Service B...Call Girls Indiranagar Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
Call Girls Indiranagar Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
amitlee9823
 
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
amitlee9823
 
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
ZurliaSoop
 
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
amitlee9823
 

Recently uploaded (20)

Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...
 
Halmar dropshipping via API with DroFx
Halmar  dropshipping  via API with DroFxHalmar  dropshipping  via API with DroFx
Halmar dropshipping via API with DroFx
 
Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...
Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...
Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...
 
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts ServiceCall Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
Call Girls In Shalimar Bagh ( Delhi) 9953330565 Escorts Service
 
Discover Why Less is More in B2B Research
Discover Why Less is More in B2B ResearchDiscover Why Less is More in B2B Research
Discover Why Less is More in B2B Research
 
Sampling (random) method and Non random.ppt
Sampling (random) method and Non random.pptSampling (random) method and Non random.ppt
Sampling (random) method and Non random.ppt
 
CebaBaby dropshipping via API with DroFX.pptx
CebaBaby dropshipping via API with DroFX.pptxCebaBaby dropshipping via API with DroFX.pptx
CebaBaby dropshipping via API with DroFX.pptx
 
ALSO dropshipping via API with DroFx.pptx
ALSO dropshipping via API with DroFx.pptxALSO dropshipping via API with DroFx.pptx
ALSO dropshipping via API with DroFx.pptx
 
Call Girls Indiranagar Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
Call Girls Indiranagar Just Call 👗 7737669865 👗 Top Class Call Girl Service B...Call Girls Indiranagar Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
Call Girls Indiranagar Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
 
Ravak dropshipping via API with DroFx.pptx
Ravak dropshipping via API with DroFx.pptxRavak dropshipping via API with DroFx.pptx
Ravak dropshipping via API with DroFx.pptx
 
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort ServiceBDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service
 
Capstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics ProgramCapstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics Program
 
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Marol Naka Call On 9920725232 With Body to body massage...
 
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
 
VidaXL dropshipping via API with DroFx.pptx
VidaXL dropshipping via API with DroFx.pptxVidaXL dropshipping via API with DroFx.pptx
VidaXL dropshipping via API with DroFx.pptx
 
Call Girls in Sarai Kale Khan Delhi 💯 Call Us 🔝9205541914 🔝( Delhi) Escorts S...
Call Girls in Sarai Kale Khan Delhi 💯 Call Us 🔝9205541914 🔝( Delhi) Escorts S...Call Girls in Sarai Kale Khan Delhi 💯 Call Us 🔝9205541914 🔝( Delhi) Escorts S...
Call Girls in Sarai Kale Khan Delhi 💯 Call Us 🔝9205541914 🔝( Delhi) Escorts S...
 
ELKO dropshipping via API with DroFx.pptx
ELKO dropshipping via API with DroFx.pptxELKO dropshipping via API with DroFx.pptx
ELKO dropshipping via API with DroFx.pptx
 
Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% SecureCall me @ 9892124323  Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure
 
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
 
BabyOno dropshipping via API with DroFx.pptx
BabyOno dropshipping via API with DroFx.pptxBabyOno dropshipping via API with DroFx.pptx
BabyOno dropshipping via API with DroFx.pptx
 

John Robert: Making your machine learning model usable by others

  • 1. Making your Machine Learning Model useable by others
  • 2. Hello! I am John Robert I put machine learning in production at Adtriba. Adtriba uses machine learning to improve marketing, optimize budget allocation and increase revenue @trojrobert 2
  • 3. Usability Deploy machine learning models to production Make a product or service with machine learning model 3
  • 4. Examples of usable models 4 From ‘Attention Is All You Need’ by Vaswani et al.
  • 5. Examples of usable models Tesla Autopilot 5
  • 6. Machine Learning Life Cycle 6 Source - ttps://woolpert.com/media/blogs/geospatial/what-does-a-machine-learning-project-look-like/
  • 7. Usable Model can be compared with serving prepared food. 7 Whole roasted scup with piperade, grilled cauliflower and bronze fennel - https://www.tripadvisor.com/LocationPhotoDirectLink-g41790-d11620864- i325210489-Feather_Wedge-Rockport_Cape_Ann_Massachusetts.html
  • 8. 8 Source and prepare high quality data Train model Deploy model Source and prepare high quality ingredients Cook a meal Serve the meal AI system
  • 9. “Plates” used in serving/deploying ML models 9
  • 10. “Table” for serving/ deploying ML Models 10 Source - https://allcode.com/cloud-providers/
  • 11. AI Systems = Model(code) + Data DevOps MLOps Normal system = Code 11 Source - https://towardsdatascience.com/challenges-deploying-machine-learning-models-to-production-ded3f9009cb3
  • 13. Serving/ deploying traditional systems(DevOps) 13
  • 14. Tools for serving /deploying ML Models 14 create and share beautiful, custom web apps for machine learning and data science is continuous integration for machine learning. Bring DevOps practices to your projects for automatic, reproducible, and fast machine learning. Allow you to manage and deploy models from a variety of ML libraries to a variety of model serving and inference platforms. end-to-end enterprise AI platform that automates and accelerates every step of your path from data to value. end-to-end machine learning platform to build and deploy AI models at scale making deployments of machine learning (ML) workflows on Kubernetes simple, portable and scalable.
  • 15. Tools for serving /deploying ML Models 15
  • 17. Roadblock of deploying ML model (development) Data Leakage happens when your training data contains information about the target, but similar data will not be available when the model is used for prediction. 17
  • 18. Roadblocks when serving/deploying ML Model Data Drift is the change in input data distribution of the model. Prediction Drift is the change in output of the model. For instance change in target data Covariate Drift is the change in pattern learnt by the model. Change in the relationship between the features and the target 18 Model Degradation
  • 19. Roadblocks when serving/deploying ML Model 19 Source - MIT 6.S191: Evidential Deep Learning and Uncertainty
  • 20. Roadblocks when serving/deploying ML Model Out Of Distribution this occur when you try to predict a class that the model was not trained on 20 What will the model predict when the input image is a car?
  • 21. Roadblocks when serving/deploying ML Model Infrastructure 21 Source - https://severalnines.com/database- blog/scaling-postgresql-large-amounts-data Source - https://akfpartners.com/growth-blog/what-is-latency
  • 22. When your model is too big for the plate Quantization Quantization is reducing the bitwidths of weights in a model. For instance convert values from float to integer. Pruning reducing the size of a machine learning model. For instance removing some layers. 22 Model compression
  • 23. Preserve your served/ deployed ML model Retraining Since the world is changing, data generated is also changing. Models need to adapt to the changes. Monitoring Track and analyse data and model performance 23 Data and Model Health (DataRobot)