“Houston, we have a model...” Introduction to MLOps

Feb 25, 20211 like338 views

The document introduces MLOps (Machine Learning Operations) and the need to operationalize machine learning models beyond just model deployment. It discusses challenges like data and model drift, retraining models, software dependencies, monitoring models in production, and the need for automation, testing, and reproducibility across the full machine learning lifecycle from data to deployment. An example MLOps workflow is shown using GitHub and Azure ML to enable experiment tracking, automation, and continuous integration and delivery of models.

Rui Quintino
AI & Analytics Strategy Advisor
DevScope
20210219
“Houston,
we have a model...”
Introduction to MLOps

We have a model... It’s great!
YOU
ARE
HERE

Easy right?
https://cloud.google.com/solutions/machine-learning/mlops-continuous-delivery-and-automation-pipelines-in-machine-learning

Or Data Drift -> causing Model Performance degradation?
https://github.com/Azure/MachineLearningNotebooks/blob/master/how-to-use-azureml/work-with-data/datadrift-tutorial/datadrift-tutorial.ipynb

Plus...
• How/who can we retrain if needed (ex: data drift)?
• What are the software requirements?
• How can we later update lib/os versions? (ex: security)
• How can we monitor model performance in production?
• How do we ensure data checks in production?
• How was the model trained? What was the performance?
• How to scale? And Reproduce?
• ...

Model & ML code just tip of the iceberg...

Model & ML code just tip of the iceberg...
https://cloud.google.com/solutions/machine-learning/mlops-continuous-delivery-and-automation-pipelines-in-machine-learning

Introducing MLOps – Machine Learning & Operations

DevOps MLOps
Code testing
Code reproducibility
App deployment
Model retraining
Model validation
Model reproducibility
Model deployment

Build app
Collaborate Test app Deploy app Monitor app
App developer

Model reproducibility Model retraining
Model deployment
Model validation
Train model Validate
model
Deploy
model
Monitor
model
Retrain model
Build app
Collaborate Test app Release app Monitor app
Audit /interpretability
App developer
Data scientist

“Houston, we have a model...” Introduction to MLOps

Lets try a new hyper param configuration...

Compute, Experiment, Model, Deployment in Azure ML

Tracking critical for MLOps, but also for Experimenting

With GitHub & CML
• Automation for
Experimentation
• Ex: Smoke Test Locally (Fast)
• Queue Reproducible in Github
• Using notebooks
Experimentation

GitHub Workflows to support Experiment Queues
Check Continuous ML (CML) framework in Github

Branch for new experiment, change code, queue in GitHub

Commit & Push branch to queue workflow in GitHub

Notebook execution using papermill on Github agent

Train performance results on commits (also PRs)

MLOps Recap
•From Models to Pipelines
•Execution of Pipelines produce models
•Train for Deployment is fully automated/reproducible
•Pipelines enable auto model retraining
•Automation, Automation, Automation...
•Monitoring for Data Drift, Concept Drift, Model Drift
•Tracking, Metadata & Artifact Stores

Mature MLOps Example (Level 1)
https://cloud.google.com/solutions/machine-learning/mlops-continuous-delivery-and-automation-pipelines-in-machine-learning

Not best fit for interative/model development

Final notes
•MLOps doesn’t guarantee ML success
•MLOps should improve feedback and learning
(not block DataScientists from business/data understanding)
•Dont reinvent the wheel...
•Still a young field
Dont believe everything you read in docs (and ppt...)
•When/If to introduce MLOps
(it depends...)

https://mlops.githubapp.com/

Also
• https://azure.microsoft.com/en-us/services/machine-
learning/mlops/
• https://cloud.google.com/solutions/machine-learning/mlops-
continuous-delivery-and-automation-pipelines-in-machine-
learning
• https://github.com/microsoft/MLOps
• https://databricks.com/blog/2020/10/13/using-mlops-with-
mlflow-and-azure.html
• https://github.com/Azure/MachineLearningNotebooks

© 2016 DevScope. All rights reserved.
Rua Passos Manuel Nº 223 – 4º Andar
4000-385 Porto
T. +351 223 751 350/51
F. +351 223 751 352
Av. Sidónio Pais, Nº 2 – 3º Andar
1050-214 5 Lisboa
info@devscope.net
www.devscope.net

Команда Data Phoenix Events приглашает всех, 17 августа в 19:00, на первый вебинар из серии "The A-Z of Data", который будет посвящен MLOps. В рамках вводного вебинара, мы рассмотрим, что такое MLOps, основные принципы и практики, лучшие инструменты и возможные архитектуры. Мы начнем с простого жизненного цикла разработки ML решений и закончим сложным, максимально автоматизированным, циклом, который нам позволяет реализовать MLOps. https://dataphoenix.info/the-a-z-of-data/ https://dataphoenix.info/the-a-z-of-data-introduction-to-mlops/

From Data Science to MLOpsCarl W. Handlin

The document discusses moving from data science to MLOps. It defines MLOps as extending DevOps methodology to include machine learning, data science, and data engineering assets. Key concepts of MLOps include iterative development, automation, continuous integration and delivery, versioning, testing, reproducibility, monitoring, source control, and model/feature stores. MLOps helps address challenges of moving models to production like the deployment gap by establishing best practices and tools for testing, deploying, managing, and monitoring models.

MLOps in actionPieter de Bruin

Ml ops intro sessionAvinash Patil

MLOps for production-level machine learningcnvrg.io AI OS - Hands-on ML Workshops

MLOps (a compound of “machine learning” and “operations”) is a practice for collaboration and communication between data scientists and operations professionals to help manage the production machine learning lifecycle. Similar to the DevOps term in the software development world, MLOps looks to increase automation and improve the quality of production ML while also focusing on business and regulatory requirements. MLOps applies to the entire ML lifecycle - from integrating with model generation (software development lifecycle, continuous integration/continuous delivery), orchestration, and deployment, to health, diagnostics, governance, and business metrics. To watch the full presentation click here: https://info.cnvrg.io/mlopsformachinelearning In this webinar, we’ll discuss core practices in MLOps that will help data science teams scale to the enterprise level. You’ll learn the primary functions of MLOps, and what tasks are suggested to accelerate your teams machine learning pipeline. Join us in a discussion with cnvrg.io Solutions Architect, Aaron Schneider, and learn how teams use MLOps for more productive machine learning workflows. - Reduce friction between science and engineering - Deploy your models to production faster - Health, diagnostics and governance of ML models - Kubernetes as a core platform for MLOps - Support advanced use-cases like continual learning with MLOps

MLOps with Azure DevOpsMarco Parenzan

MLOps refers to applying DevOps practices and principles to machine learning. This allows for machine learning models and projects to be developed and deployed using automated pipelines for continuous integration and delivery. MLOps benefits include making machine learning work reproducible and auditable, enabling validation of models, and providing observability through monitoring of models after deployment. MLOps uses the same development practices as software engineering to ensure quality control for machine learning.

Using MLOps to Bring ML to Production/The Promise of MLOpsWeaveworks

In this final Weave Online User Group of 2019, David Aronchick asks: have you ever struggled with having different environments to build, train and serve ML models, and how to orchestrate between them? While DevOps and GitOps have made huge traction in recent years, many customers struggle to apply these practices to ML workloads. This talk will focus on the ways MLOps has helped to effectively infuse AI into production-grade applications through establishing practices around model reproducibility, validation, versioning/tracking, and safe/compliant deployment. We will also talk about the direction for MLOps as an industry, and how we can use it to move faster, with more stability, than ever before. The recording of this session is on our YouTube Channel here: https://youtu.be/twsxcwgB0ZQ Speaker: David Aronchick, Head of Open Source ML Strategy, Microsoft Bio: David leads Open Source Machine Learning Strategy at Azure. This means he spends most of his time helping humans to convince machines to be smarter. He is only moderately successful at this. Previously, David led product management for Kubernetes at Google, launched GKE, and co-founded the Kubeflow project. David has also worked at Microsoft, Amazon and Chef and co-founded three startups. Sign up for a free Machine Learning Ops Workshop: http://bit.ly/MLOps_Workshop_List Weaveworks will cover concepts such as GitOps (operations by pull request), Progressive Delivery (canary, A/B, blue-green), and how to apply those approaches to your machine learning operations to mitigate risk.

MLOps Using MLflowDatabricks

MLflow is an MLOps tool that enables data scientist to quickly productionize their Machine Learning projects. To achieve this, MLFlow has four major components which are Tracking, Projects, Models, and Registry. MLflow lets you train, reuse, and deploy models with any library and package them into reproducible steps. MLflow is designed to work with any machine learning library and require minimal changes to integrate into an existing codebase. In this session, we will cover the common pain points of machine learning developers such as tracking experiments, reproducibility, deployment tool and model versioning. Ready to get your hands dirty by doing quick ML project using mlflow and release to production to understand the ML-Ops lifecycle.

Seamless MLOps with Seldon and MLflowDatabricks

The document provides an overview of seamless MLOps using Seldon and MLflow. It discusses how MLOps is challenging due to the wide range of requirements across the ML lifecycle. MLflow helps with training by allowing experiment tracking and model versioning. Seldon Core helps with deployment by providing servers to containerize models and infrastructure for monitoring, A/B testing, and feedback. The demo shows training models with MLflow, deploying them to Seldon for A/B testing, and collecting feedback to optimize models.

MLOps and Data Quality: Deploying Reliable ML Models in ProductionProvectus

Looking to build a robust machine learning infrastructure to streamline MLOps? Learn from Provectus experts how to ensure the success of your MLOps initiative by implementing Data QA components in your ML infrastructure. For most organizations, the development of multiple machine learning models, their deployment and maintenance in production are relatively new tasks. Join Provectus as we explain how to build an end-to-end infrastructure for machine learning, with a focus on data quality and metadata management, to standardize and streamline machine learning life cycle management (MLOps). Agenda - Data Quality and why it matters - Challenges and solutions of Data Testing - Challenges and solutions of Model Testing - MLOps pipelines and why they matter - How to expand validation pipelines for Data Quality

Apply MLOps at Scale by H&MDatabricks

H&M uses machine learning for various use cases including logistics, production, sales, marketing, and design/buying. MLOps principles like model versioning, reproducibility, scalability, and automated training are applied to manage the machine learning lifecycle. The technical stack includes Kubernetes, Docker, Azure Databricks for interactive development, Airflow for automated training, and Seldon for model serving. The goal is to apply MLOps at scale for various prediction scenarios through a continuous integration/continuous delivery pipeline.

Databricks Overview for MLOpsDatabricks

1) Databricks provides a machine learning platform for MLOps that includes tools for data ingestion, model training, runtime environments, and monitoring. 2) It offers a collaborative data science workspace for data engineers, data scientists, and ML engineers to work together on projects using notebooks. 3) The platform provides end-to-end governance for machine learning including experiment tracking, reproducibility, and model governance.

MLOps with Kubeflow Saurabh Kaushik

This document discusses MLOps and Kubeflow. It begins with an introduction to the speaker and defines MLOps as addressing the challenges of independently autoscaling machine learning pipeline stages, choosing different tools for each stage, and seamlessly deploying models across environments. It then introduces Kubeflow as an open source project that uses Kubernetes to minimize MLOps efforts by enabling composability, scalability, and portability of machine learning workloads. The document outlines key MLOps capabilities in Kubeflow like Jupyter notebooks, hyperparameter tuning with Katib, and model serving with KFServing and Seldon Core. It describes the typical machine learning process and how Kubeflow supports experimental and production phases.

MLOps.pptxAllenPeter7

This document discusses MLOps, which aims to standardize and streamline machine learning model development and deployment through continuous delivery. MLOps applies agile principles to machine learning projects and treats models and datasets as first-class citizens within CI/CD systems. The document outlines three levels of MLOps implementation from manual to fully automated pipelines. It also describes common MLOps platform tools for data management, modeling, and operationalization, including tools for data labeling, versioning, experiment tracking, hyperparameter optimization, model deployment, and monitoring.

ML-Ops: From Proof-of-Concept to Production ApplicationHunter Carlisle

Ml ops past_present_futureNisha Talagala

MLOps Virtual Event | Building Machine Learning Platforms for the Full LifecycleDatabricks

This document summarizes a webinar on building machine learning platforms. It discusses how operating ML models is complex, requiring tasks like monitoring performance, handling data drift, and ensuring governance and security. It then outlines common components of ML platforms, including data management, model management, and code/deployment management. The webinar will demonstrate how different organizations handle these components and include demos from four companies. It will also cover Databricks' approach to providing an ML platform that integrates various tools and simplifies the full ML lifecycle from data preparation to deployment.

ML-Ops how to bring your data science to productionHerman Wu

This document discusses end-to-end machine learning (ML) workflows and operations (MLOps) on Azure. It provides an overview of the ML lifecycle including developing and training models, validating models, deploying models, packaging models, and monitoring models. It also discusses how Azure services like Azure Machine Learning and Azure DevOps can be used to implement MLOps practices for continuous integration, delivery, and deployment of ML models. Real-world examples of automating energy demand forecasting and computer vision models are also presented.

Apply MLOps at ScaleDatabricks

MLOps - The Assembly Line of MLJordan Birdsell

The catalyst for the success of automobiles came not through the invention of the car but rather through the establishment of an innovative assembly line. History shows us that the ability to mass produce and distribute a product is the key to driving adoption of any innovation, and machine learning is no different. MLOps is the assembly line of Machine Learning and in this presentation we will discuss the core capabilities your organization should be focused on to implement a successful MLOps system.

MLOps Bridging the gap between Data Scientists and Ops.Knoldus Inc.

MLOps by Sasha RosenbaumSasha Rosenbaum

This document discusses MLOps, which is applying DevOps practices and principles to machine learning to enable continuous delivery of ML models. It explains that ML models need continuous improvement through retraining but data scientists currently lack tools for quick iteration, versioning, and deployment. MLOps addresses this by providing ML pipelines, model management, monitoring, and retraining in a reusable workflow similar to how software is developed. Implementing even a basic CI/CD pipeline for ML can help iterate models more quickly than having no pipeline at all. The document encourages building responsible AI through practices like ensuring model performance and addressing bias.

Managing the Complete Machine Learning Lifecycle with MLflowDatabricks

ML development brings many new complexities beyond the traditional software development lifecycle. Unlike in traditional software development, ML developers want to try multiple algorithms, tools and parameters to get the best results, and they need to track this information to reproduce work. In addition, developers need to use many distinct systems to productionize models. To solve for these challenges, Databricks unveiled last year MLflow, an open source project that aims at simplifying the entire ML lifecycle. MLflow introduces simple abstractions to package reproducible projects, track results, and encapsulate models that can be used with many existing tools, accelerating the ML lifecycle for organizations of any size. In the past year, the MLflow community has grown quickly: over 120 contributors from over 40 companies have contributed code to the project, and over 200 companies are using MLflow. In this tutorial, we will show you how using MLflow can help you: Keep track of experiments runs and results across frameworks. Execute projects remotely on to a Databricks cluster, and quickly reproduce your runs. Quickly productionize models using Databricks production jobs, Docker containers, Azure ML, or Amazon SageMaker. We will demo the building blocks of MLflow as well as the most recent additions since the 1.0 release. What you will learn: Understand the three main components of open source MLflow (MLflow Tracking, MLflow Projects, MLflow Models) and how each help address challenges of the ML lifecycle. How to use MLflow Tracking to record and query experiments: code, data, config, and results. How to use MLflow Projects packaging format to reproduce runs on any platform. How to use MLflow Models general format to send models to diverse deployment tools. Prerequisites: A fully-charged laptop (8-16GB memory) with Chrome or Firefox Python 3 and pip pre-installed Pre-Register for a Databricks Standard Trial Basic knowledge of Python programming language Basic understanding of Machine Learning Concepts

Introduction to MLflowDatabricks

Intro to Vertex AI, unified MLOps platform for Data Scientists & ML EngineersDaniel Zivkovic

This document introduces ServerlessToronto.org and provides information about upcoming events. It discusses how adopting a serverless mindset can help companies accelerate by shifting the focus from infrastructure to business outcomes. It promotes bridging the gap between business and IT through serverless consulting services and knowledge sharing events. Upcoming events are listed, and there is an offer to be a raffle winner for a Manning e-book. The final sections provide information about an upcoming presentation on Google's Vertex AI platform for machine learning.

Machine Learning Operations & AzureErlangen Artificial Intelligence & Machine Learning Meetup

Pythonsevilla2019 - Introduction to MLFlowFernando Ortega Gallego

mlflow: Accelerating the End-to-End ML lifecycleDatabricks

Building and deploying a machine learning model can be difficult to do once. Enabling other data scientists (or yourself, one month later) to reproduce your pipeline, to compare the results of different versions, to track what’s running where, and to redeploy and rollback updated models is much harder. In this talk, I’ll introduce MLflow, a new open source project from Databricks that simplifies the machine learning lifecycle. MLflow provides APIs for tracking experiment runs between multiple users within a reproducible environment, and for managing the deployment of models to production. MLflow is designed to be an open, modular platform, in the sense that you can use it with any existing ML library and development process. MLflow was launched in June 2018 and has already seen significant community contributions, with over 50 contributors and new features including language APIs, integrations with popular ML libraries, and storage backends. I’ll show how MLflow works and explain how to get started with MLflow.

Slides-Артем Коваль-Cloud-Native MLOps Framework - DataFest 2021.pdfvitm11

Artem Koval presented on cloud-native MLOps frameworks. MLOps is a process for deploying and monitoring machine learning models through continuous integration and delivery. It addresses fairness, explainability, model monitoring, and human intervention. Modern MLOps frameworks focus on these areas as well as data labeling, testing, and observability. Different levels of MLOps are needed depending on an organization's size, from lightweight for small teams to enterprise-level for large companies with many models. Human-centered AI should be incorporated at all levels by involving humans throughout the entire machine learning process.

Trenowanie i wdrażanie modeli uczenia maszynowego z wykorzystaniem Google Clo...Sotrender

Okej, mam już mój świetny model w Notebooku, co dalej? Większość kursów i źródeł dotyczących uczenia maszynowego dobrze przygotowuje nas do implementacji algorytmów uczenia maszynowego i budowy mniej lub bardziej skomplikowanych modeli. Jednak w większości przypadków model jest jedynie małym fragmentem większego systemu, a jego wdrożenie i utrzymywanie okazuje się w praktyce procesem czasochłonnym i generującym rozmaite błędy. Problem potęguje się kiedy mamy do sproduktyzowania nie jeden, a więcej modeli. Choć z roku na rok powstaje coraz więcej narzędzi i platform do usprawnienia tego procesu, jest to zagadnienie któremu wciąż poświęca się stosunkowo mało uwagi. W mojej prezentacji przedstawię jakich podejść, dobrych praktyk oraz narzędzi i usług Google Cloud Platform używamy w Sotrender do efektywnego trenowania i produktyzacji naszych modeli ML, służących do analizy danych z mediów społecznościowych. Omówię na które aspekty DevOps zwracamy uwagę w kontekście wytwarzania produktów opartych o modele ML (MLOps) i jak z wykorzystaniem Google Cloud Platform można je w łatwy sposób wdrożyć w swoim startupie lub firmie. Prezentacja Macieja Pieńkosza z Sotrendera poczas Data Science Summit 2020

More Related Content

What's hot (20)

Seamless MLOps with Seldon and MLflowDatabricks

MLOps and Data Quality: Deploying Reliable ML Models in ProductionProvectus

Apply MLOps at Scale by H&MDatabricks

Databricks Overview for MLOpsDatabricks

MLOps with Kubeflow Saurabh Kaushik

MLOps.pptxAllenPeter7

ML-Ops: From Proof-of-Concept to Production ApplicationHunter Carlisle

Ml ops past_present_futureNisha Talagala

MLOps Virtual Event | Building Machine Learning Platforms for the Full LifecycleDatabricks

ML-Ops how to bring your data science to productionHerman Wu

Apply MLOps at ScaleDatabricks

MLOps - The Assembly Line of MLJordan Birdsell

MLOps Bridging the gap between Data Scientists and Ops.Knoldus Inc.

MLOps by Sasha RosenbaumSasha Rosenbaum

Managing the Complete Machine Learning Lifecycle with MLflowDatabricks

Introduction to MLflowDatabricks

Intro to Vertex AI, unified MLOps platform for Data Scientists & ML EngineersDaniel Zivkovic

Machine Learning Operations & AzureErlangen Artificial Intelligence & Machine Learning Meetup

Pythonsevilla2019 - Introduction to MLFlowFernando Ortega Gallego

mlflow: Accelerating the End-to-End ML lifecycleDatabricks

Seamless MLOps with Seldon and MLflowDatabricks

MLOps and Data Quality: Deploying Reliable ML Models in ProductionProvectus

Apply MLOps at Scale by H&MDatabricks

Databricks Overview for MLOpsDatabricks

MLOps with Kubeflow Saurabh Kaushik

MLOps.pptxAllenPeter7

ML-Ops: From Proof-of-Concept to Production ApplicationHunter Carlisle

Ml ops past_present_futureNisha Talagala

MLOps Virtual Event | Building Machine Learning Platforms for the Full LifecycleDatabricks

ML-Ops how to bring your data science to productionHerman Wu

Apply MLOps at ScaleDatabricks

MLOps - The Assembly Line of MLJordan Birdsell

MLOps Bridging the gap between Data Scientists and Ops.Knoldus Inc.

MLOps by Sasha RosenbaumSasha Rosenbaum

Managing the Complete Machine Learning Lifecycle with MLflowDatabricks

Introduction to MLflowDatabricks

Intro to Vertex AI, unified MLOps platform for Data Scientists & ML EngineersDaniel Zivkovic

Machine Learning Operations & AzureErlangen Artificial Intelligence & Machine Learning Meetup

Pythonsevilla2019 - Introduction to MLFlowFernando Ortega Gallego

mlflow: Accelerating the End-to-End ML lifecycleDatabricks

Similar to “Houston, we have a model...” Introduction to MLOps (20)

Slides-Артем Коваль-Cloud-Native MLOps Framework - DataFest 2021.pdfvitm11

Trenowanie i wdrażanie modeli uczenia maszynowego z wykorzystaniem Google Clo...Sotrender

Building a MLOps Platform Around MLflow to Enable Model Productionalization i...Databricks

Getting machine learning models to production is notoriously difficult: it involves multiple teams (data scientists, data and machine learning engineers, operations, …), who often does not speak to each other very well; the model can be trained in one environment but then productionalized in completely different environment; it is not just about the code, but also about the data (features) and the model itself… At DataSentics, as a machine learning and cloud engineering studio, we see this struggle firsthand – on our internal projects and client’s projects as well.

artificggggggggggggggialintelligence.pdftt4765690

This document provides an overview of MLOps principles and practices based on the author's experiences developing and deploying machine learning systems. It discusses key concepts like machine learning, models, algorithms, and ground-truth data. The document then explains that operationalizing machine learning involves both data scientists developing algorithms on historical data and ML engineering teams integrating models into operational systems and data flows. It outlines the typical steps of initial model development, integration/deployment, monitoring performance, and updating models. Several principles of MLOps are also presented, including having solid data foundations with accessible, high-quality ground-truth data for data scientists and maintainers to use.

Mohamed Sabri: Operationalize machine learning with KubeflowLviv Startup Club

This document summarizes a hands-on workshop on Kubeflow Pipeline. The workshop will cover requirements, an introduction to the presenter Mohamed Sabri, and their approach of strategizing, shaping, and spreading knowledge. It then discusses operationalizing machine learning (MLOps) and provides an analysis, design, coaching, and implementation framework. Deliverables include an implemented MLOps environment, training sessions, design documents, and a recommendations roadmap. The rest of the document discusses MLOps architectures, challenges, example technologies and tools, a use case, and deployment workflows from notebooks to production.

Mohamed Sabri: Operationalize machine learning with KubeflowEdunomica

CNCF-Istanbul-MLOps for Devops Engineers.pptxcansukavili1

AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)dtz001

While the adoption of machine learning and deep learning techniques continue to grow, many organizations find it difficult to actually deploy these sophisticated models into production. It is common to see data scientists build powerful models, yet these models are not deployed because of the complexity of the technology used or lack of understanding related to the process of pushing these models into production. As part of this talk, I will review several deployment design patterns for both real-time and batch use cases. I’ll show how these models can be deployed as scalable, distributed deployments within the cloud, scaled across hadoop clusters, as APIs, and deployed within streaming analytics pipelines. I will also touch on topics related to security, end-to-end governance, pitfalls, challenges, and useful tools across a variety of platforms. This presentation will involve demos and sample code for the the deployment design patterns.

Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...All Things Open

Apache Spark Model Deployment Databricks

Tech-Talk at Bay Area Spark Meetup Apache Spark(tm) has rapidly become a key tool for data scientists to explore, understand and transform massive datasets and to build and train advanced machine learning models. The question then becomes, how do I deploy these model to a production environment. How do I embed what I have learned into customer facing data applications. Like all things in engineering, it depends. In this meetup, we will discuss best practices from Databricks on how our customers productionize machine learning models and do a deep dive with actual customer case studies and live demos of a few example architectures and code in Python and Scala. We will also briefly touch on what is coming in Apache Spark 2.X with model serialization and scoring options.

Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning ModelsAnyscale

Apache Spark has rapidly become a key tool for data scientists to explore, understand and transform massive datasets and to build and train advanced machine learning models. The question then becomes, how do I deploy these model to a production environment? How do I embed what I have learned into customer facing data applications? In this webinar, we will discuss best practices from Databricks on how our customers productionize machine learning models do a deep dive with actual customer case studies, show live tutorials of a few example architectures and code in Python, Scala, Java and SQL.

Machine Learning Operations Cababilitiesdavidsh11

Making Netflix Machine Learning Algorithms ReliableJustin Basilico

This document discusses making Netflix machine learning algorithms reliable. It describes how Netflix uses machine learning for tasks like personalized ranking and recommendation. The goals are to maximize member satisfaction and retention. The models and algorithms used include regression, matrix factorization, neural networks, and bandits. The key aspects of making the models reliable discussed are: automated retraining of models, testing training pipelines, checking models and inputs online for anomalies, responding gracefully to failures, and training models to be resilient to different conditions and failures.

Consolidating MLOps at One of Europe’s Biggest AirportsDatabricks

At Schiphol airport we run a lot of mission critical machine learning models in production, ranging from models that predict passenger flow to computer vision models that analyze what is happening around the aircraft. Especially now in times of Covid it is paramount for us to be able to quickly iterate on these models by implementing new features, retraining them to match the new dynamics and above all to monitor them actively to see if they still fit the current state of affairs. To achieve those needs we rely on MLFlow but have also integrated that with many of our other systems. So have we written Airflow operators for MLFlow to ease the retraining of our models, have we integrated MLFlow deeply with our CI pipelines and have we integrated it with our model monitoring tooling. In this talk we will take you through the way we rely on MLFlow and how that enables us to release (sometimes) multiple versions of a model per week in a controlled fashion. With this set-up we are achieving the same benefits and speed as you have with a traditional software CI pipeline.

Enabling Scalable Data Science Pipeline with Mlflow at Thermo Fisher ScientificDatabricks

Mykola Mykytenko: MLOps: your way from nonsense to valuable effect (approache...Lviv Startup Club

How To Build Auto-Adaptive Machine Learning Models with Kubernetescnvrg.io AI OS - Hands-on ML Workshops

This webinar will instruct data scientists and machine learning engineers how to build manage and deploy auto-adaptive machine learning models in production. Data is ever changing, leaving your models outdated and built on old data. This can lead to underperforming models and a lot of manual work to fix it. By allowing your models to continually learn you’ll ensure that they run at peak performance. Using state of the art Kubernetes infrastructure, we’ll show you how to automatically track and manage your auto-adaptive machine learning models while in production. By building auto-adaptive machine learning models, data engineers can bridge the gap between research and production. After this webinar you’ll be able to build and deploy machine learning pipelines that automatically adapt and retrain based on any validation trigger you choose. Key webinar takeaways: How to build auto-adaptive machine learning pipelines How to use Kubernetes to manage and scale models in production How to automatically monitor for peak performance How to set up continuous deployment of ML pipeline Watch all our webinars at https://cnvrg.io/webinars-and-workshops/

Productionizing Machine Learning with a Microservices ArchitectureDatabricks

[DSC Europe 23] Petar Zecevic - ML in Production on DatabricksDataScienceConferenc1

Deploying ML models in production, with or without CI/CD, is significantly more complicated than deploying traditional applications. That is mainly because ML models do not just consist of the code used for their training, but they also depend on the data they are trained on and on the supporting code. Monitoring ML models also adds additional complexity beyond what is usually done for traditional applications. This talk will cover these problems and best practices for solving them, with special focus on how it's done on the Databricks platform.

MLOPS By Amazon offered and free downloadpouyan533

Slides-Артем Коваль-Cloud-Native MLOps Framework - DataFest 2021.pdfvitm11

Trenowanie i wdrażanie modeli uczenia maszynowego z wykorzystaniem Google Clo...Sotrender

Building a MLOps Platform Around MLflow to Enable Model Productionalization i...Databricks

artificggggggggggggggialintelligence.pdftt4765690

Mohamed Sabri: Operationalize machine learning with KubeflowLviv Startup Club

Mohamed Sabri: Operationalize machine learning with KubeflowEdunomica

CNCF-Istanbul-MLOps for Devops Engineers.pptxcansukavili1

AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)dtz001

Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...All Things Open

Apache Spark Model Deployment Databricks

Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning ModelsAnyscale

Machine Learning Operations Cababilitiesdavidsh11

Making Netflix Machine Learning Algorithms ReliableJustin Basilico

Consolidating MLOps at One of Europe’s Biggest AirportsDatabricks

Enabling Scalable Data Science Pipeline with Mlflow at Thermo Fisher ScientificDatabricks

Mykola Mykytenko: MLOps: your way from nonsense to valuable effect (approache...Lviv Startup Club

How To Build Auto-Adaptive Machine Learning Models with Kubernetescnvrg.io AI OS - Hands-on ML Workshops

Productionizing Machine Learning with a Microservices ArchitectureDatabricks

[DSC Europe 23] Petar Zecevic - ML in Production on DatabricksDataScienceConferenc1

MLOPS By Amazon offered and free downloadpouyan533

More from Rui Quintino (14)

Power BI for Data Science and Machine Learning - Data Science Portugal meetupRui Quintino

Session recording at https://youtu.be/gm5nohV30fE Abstract: Power BI is a widely successful tool for Business Intelligence projects, but there’s lots of value for Data Scientists too. This session will explore how to leverage Power BI for Data Science & Machine Learning workloads. From powerful & effortless EDA, built-in anomaly detection to R & Python integration, and even AutoML capabilities, we’ll see how these and other features can contribute to better AI project productivity and outcomes. Can Power BI be a superpower for Data Scientists? We’ll find out!

Empowering you - Power BI, Power Platform & AI BuilderRui Quintino

Jupyter Notebooks: Introduction, Tips & ToolsRui Quintino

Kaggle Days Porto 2019 - 1st place presentation by team DevScopeRui Quintino

The document summarizes Team DevScope's approach for the Kaggle competition KDM Porto 2019, which involved forecasting hourly agent headcounts at a call center. Their approach included blending regression, time series, and human knowledge models; addressing non-stationarity and a small test set; and extensive post-processing and validation to improve forecasts, especially around holidays. While their blend of approaches achieved first place, they discuss opportunities to better use additional features and automated time series tools in future forecasts.

DataSciencePT #27 - Fifty Shades of Automated Machine LearningRui Quintino

Is "the sexiest job of 21st century", the Data Scientist, about to be automated? How & when can AutoML tools help on a typical machine learning lifecycle? What AutoML challenges are still open & what ML work will remain in the foreseeable future? Most importantly… will robots get all the fun & sex appeal? :) Some questions we'll try to tackle on this session. *-Robots are not allowed in this session

Docker & Containers for Big Data, Data Science, Machine Learning & Deep LearningRui Quintino

PortoData Meetup http://www.portodata.net/xxxi-evento-porto-data-25-maio-2017-uptec/ Docker e Containers são das tecnologias atuais com maior crescimento e aceitação. Depois de um breve refresh ao Docker e vantagens dos containers, vamos ver casos práticos de como o Docker pode ajudar em workloads de desenvolvimento, testes ou mesmo produção para Big Data, Data Science, Machine Learning ou mesmo Deep Learning. Seja no posto de trabalho, on-prem ou Azure. Por último a facilidade de criação de clusters com Docker Swarm.

Microsoft Cognitive Services & Bot Framework - Universidade Fernando PessoaRui Quintino

This document discusses Microsoft's Cognitive Services and related technologies like Computer Vision, Face and Emotion APIs, and the Bot Framework. Code samples are provided to demonstrate how to use Computer Vision APIs to analyze images and detect faces. The document also promotes Microsoft's Cognitive Services, Bot Framework, and Cortana Intelligence Suite for building intelligent bots and applications using AI and machine learning technologies.

Open Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBMRui Quintino

Session at Microsoft Open Source Camp 2017 https://msoscamp.io/ Introducing Microsoft Cognitive Toolkit, a deep-learning framework for neural networks to model, train & use popular neural networks architectures for large range of machine learning scenarios.We’ll see how to start with custom image classification & computer vision scenarios using CNTK, running on CPU and GPUs, together with LightGBM (Microsoft open source gradient boosting framework).

Data Science Portugal Meetup 7 - Machine Learning & Data Science Safety Remi...Rui Quintino

Fasten your seat-belt first & other safety reminders for Data Science & Machine Learning The age of Big Data, Internet of Things, Deep Learning, increasingly effective Cognitive & Artificial Intelligence, we seem to have technology & computational resources for unlimited data mining. Still, extracting valuable insights from data remains a real challenge. We'll tackle some of the typical issues & obstacles we face when trying to extract meaning & insights from observational data. The never ending data quality challenges, why correlation is not causation is just the start… Why it's so hard to extract real causes & factors from data, and... what is observational data anyway? Things you'll be glad to know before presenting your first data project insights.

Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...Rui Quintino

The document discusses machine learning with SQL Server 2016 and R Services. It provides an overview of machine learning, R programming language, and the challenges of using R with SQL databases prior to SQL Server 2016. SQL Server 2016 introduces R Services, which allows running R code directly in the database for high performance, scalable machine learning. R Services integrates R with SQL Server through in-database deployment and parallel processing capabilities. This eliminates data movement and scaling issues while leveraging existing R and SQL skills.

Sql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbasRui Quintino

First steps with R & RStudio for SQL Developers & DBAs Are you a BI Developer? DBA? BI Analyst or a data/analytics lover? Come along, join us in this session as we explore this powerful, sometimes even strange!... language , showing main differences from the typical SQL/BI development. This is a beginner level session to guide you on why & when to use R, and help with the very first steps & typical challenges. http://www.sqlsaturday.com/583/Sessions/Details.aspx?sid=58218

The Power of Now! Azure Stream Analytics - Microsoft ITPro AirLiftRui Quintino

The document discusses Azure Stream Analytics, a fully managed stream processing service. It describes how Stream Analytics allows processing of streaming data using simple SQL queries. Examples are provided of using temporal windows like tumbling, hopping and sliding windows to analyze streaming data from scenarios like toll stations. Components of a Stream Analytics application are also outlined.

SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...Rui Quintino

"Faster than the speed of light"... with #MSBI What-if insights from your business data were just a few seconds/clicks away? What-if we could instantly explore our data with rich & fully interactive dashboards using the amazing new features of Excel 2013, PowerPivot and PowerView? Without needing to build auxiliary reference tables, complex dax or powerpivot modelling? What-if instead self-service we could be "instantly-served"?

"De histórias mal contadas..."Rui Quintino