0% found this document useful (0 votes)
73 views44 pages

Kubeflow: Simplifying ML Workflows

Kubeflow is an open source platform for developing, deploying, and running machine learning workflows on Kubernetes. It aims to make deploying machine learning workflows on Kubernetes simple, portable, and scalable. Kubeflow provides components like Jupyter notebooks, Pipelines for running workflows as portable containerized jobs, and services for deploying machine learning models, as well as tools for gathering data, preprocessing, training models, and evaluating results. The overall goal of Kubeflow is to change the development and deployment process for machine learning by building on Kubernetes.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
73 views44 pages

Kubeflow: Simplifying ML Workflows

Kubeflow is an open source platform for developing, deploying, and running machine learning workflows on Kubernetes. It aims to make deploying machine learning workflows on Kubernetes simple, portable, and scalable. Kubeflow provides components like Jupyter notebooks, Pipelines for running workflows as portable containerized jobs, and services for deploying machine learning models, as well as tools for gathering data, preprocessing, training models, and evaluating results. The overall goal of Kubeflow is to change the development and deployment process for machine learning by building on Kubernetes.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Kubeflow

Foro Internacional
de Ciencia de Datos
@ Lima - Perú
Sept 2020
Kubeflow
Acerca de mí ...

Juan Camilo Díaz Ortega


Big Data Architect at Globant
AWS Data & Analytics Certified
Azure Data Engineer Certified
Google Certified
Professional Architect
Professional Data Engineer
Why it is so painful to deploy
Machine Learning workflows?
Kubeflow Machine Learning Workflow

Gathering data

Data pre-processing

Researching the model that will be


best for the type of data

Training and testing the model

Evaluation
Kubeflow Machine Learning Workflow
Kubeflow Machine Learning Workflow
Kubeflow Machine Learning Workflow
Kubeflow Machine Learning Workflow
Kubeflow Definition

The Kubeflow project is dedicated to making deployments of machine learning


(ML) workflows on Kubernetes simple, portable and scalable.
https://www.kubeflow.org/docs/about/kubeflow/

Kubeflow is an open source Kubernetes-native platform for developing,


orchestrating, deploying, and running scalable and portable machine learning
workloads

Portable Machine Learning Stack


Kubeflow Kubeflow components
Kubeflow Changing the dev and deployment process
Kubeflow Components
Kubeflow ksonnet

https://ksonnet.io/
Kubeflow Central Dashboard

Kubeflow user interfaces (UIs)


Kubeflow Metadata

Help Kubeflow users understand


and manage their machine
learning workflows
Kubeflow Metadata
Kubeflow Jupyter Notebooks

Using multiple Jupyter notebooks


different namespaces
Kubeflow Jupyter Notebooks

Using Jupyter notebooks in


Kubeflow
Kubeflow Jupyter Notebooks
Kubeflow Jupyter Notebooks
Kubeflow Pipelines

Kubeflow Pipelines is a platform for


building and deploying portable and
scalable end-to-end ML workflows,
based on containers.

Code that performs one step in the


Pipeline. In other words a
containerized implementation of an
ML task.
Kubeflow Pipelines

A pipeline is a description of an
ML Workflow

It runs a containers which


provide portability, repeatability
and encapsulation, which is able
to decouples the execution
environment to code runtime.
Kubeflow Pipelines
Kubeflow Pipelines
Kubeflow Pipelines
Kubeflow Pipelines
Kubeflow Frameworks for Training

MPI Operator
Kubeflow Tools for Serving ML Models - KFServing
Kubeflow Tools for Serving ML Models - Seldon Core Serving
Kubeflow Tools for Serving ML Models - Bento
Kubeflow Katib - Hyperparameter Tuning
Kubeflow Katib - Hyperparameter Tuning
Kubeflow Katib - Hyperparameter Tuning
Kubeflow Katib - Hyperparameter Tuning
Google Cloud Demo
Why use Kubeflow ?
Kubeflow Data & AI Landscape

http://mattturck.com/wp-content/uploads/2019/07/2019_Matt_Turck_Big_Data_Landscape_Final_Fullsize.png
Kubeflow
Que impulsa la tendencia
Where I can Start ?
Kubeflow Cloud Computing - Cloud Providers

12 months of popular free services 12 months free services 12 months free services

+ + +

$200 credit to explore Azure for 30 days $300 free credit Short-term free trial offers start from the date
you activate a particular service
+ +
+
Always free 25+ services Always free products, which provides limited
access to many common Google Cloud Always free, free tier offers do not expire and
resources, free of charge. are available to all AWS customers
https://azure.microsoft.com/en-us/free/

https://cloud.google.com/free https://aws.amazon.com/free/

https://docs.microsoft.com/en-us/azure/ https://cloud.google.com/docs https://docs.aws.amazon.com/index.html


Kubeflow Kubeflow resources

https://www.kubeflow.org/

https://www.kubeflow.org/docs/

Getting started with Kubeflow


Kubeflow
¿Seguimos en contacto?

@jcamilodo https://www.linkedin.com/in/jcamilodo/
THANK YOU

You might also like