Simplifying Model Management with MLflow

•

5 likes•2,257 views

<p>Last summer, Databricks launched MLflow, an open source platform to manage the machine learning lifecycle, including experiment tracking, reproducible runs and model packaging. MLflow has grown quickly since then, with over 120 contributors from dozens of companies, including major contributions from R Studio and Microsoft. It has also gained new capabilities such as automatic logging from TensorFlow and Keras, Kubernetes integrations, and a high-level Java API. In this talk, we’ll cover some of the new features that have come to MLflow, and then focus on a major upcoming feature: model management with the MLflow Model Registry. Many organizations face challenges tracking which models are available in the organization and which ones are in production. The MLflow Model Registry provides a centralized database to keep track of these models, share and describe new model versions, and deploy the latest version of a model through APIs. We’ll demonstrate how these features can simplify common ML lifecycle tasks.</p>

Data & Analytics

ML development is harder than
traditional software development

Traditional Software Machine Learning
Goal: Optimize a metric (e.g., accuracy)
• Constantly experiment to improve it
Quality depends on input data
and tuning parameters
Compare + combine many libraries,
models & algorithms for the same task
Goal: Meet a functional specification
Quality depends only on code
Typically pick one software stack

Production ML is Even Harder
Data Prep
Training
Deployment
Raw Data
ML ENGINEER
APPLICATION
DEVELOPER
DATA
ENGINEER
ML apps must be fed new data
to keep working
Design, retraining & inference
done by different people

Solution: Machine Learning Platforms
Software to manage the ML lifecycle
Examples: Uber Michelangelo,
Google TFX, Facebook FBLearner
Data Prep
Training
Deployment
Raw Data
Versioning, CI/CD, QA,
ops, monitoring, etc

MLflow: An Open Source ML Platform
Three components:
• Tracking: experiment tracking
• Projects: reproducible runs
• Models: model packaging
140 contributors, 800K downloads/month
Works with any ML library, programming language, deployment tool

MLflow Tracking: Experiments
Notebooks
Local Apps
Cloud Jobs
Tracking Server
UI
API
mlflow.log_param(“alpha”, 0.5)
mlflow.log_metric(“accuracy”, 0.9)
...
REST API

MLflow Projects:
Reproducible Runs
Project Spec
Code DataConfig
Local
Execution
Remote
Cluster
MLflow Models:
Model Packaging
Model Format
ONNX Flavor
Python Flavor
Model Logic
Batch & Stream Scoring
REST Serving
Packaging Format
. . .
Evaluation & Debugging
LIME
TCAV

New in Last 6 Months
MLflow 1.0 (and 1.1, 1.2, 1.3)
Autologging in TensorFlow & Keras
DataFrame search API
Kubernetes, HDFS & Seldon integrations

MLflow Autologging
model = keras.models.Sequential()
model.add(layers.Dense(hidden_units, ...))
model.fit(X_train, y_train)
test_loss = model.evaluate(X_test, y_test)

MLflow Autologging
with mlflow.start_run():
model = keras.models.Sequential()
model.add(layers.Dense(hidden_units, ...))
model.fit(X_train, y_train)
test_loss = model.evaluate(X_test, y_test)
mlflow.log_param(“hidden_units”, hidden_units)
mlflow.log_param(“learning_rate”, learning_rate)
mlflow.log_metric(“train_loss”, train_loss)
mlflow.log_metric(“test_loss”, test_loss)
mlflow.keras.log_model(model)
mlflow.keras.autolog()
model = keras.models.Sequential()
model.add(layers.Dense(hidden_units, ...))
model.fit(X_train, y_train)
test_loss = model.evaluate(X_test, y_test)

The Model Management Problem
When you’re working on one ML app alone, storing your
models in files is manageable
MODEL
DEVELOPER
classifier_v1.h5
classifier_v2.h5
classifier_v3_sept_19.h5
classifier_v3_new.h5
…

The Model Management Problem
When you work in a large organization with many models,
management becomes a major challenge:
• Where can I find the best version of this model?
• How was this model trained?
• How can I track docs for each model?
• How can I review models?
MODEL
DEVELOPER
REVIEWER
MODEL
USER
???

MLflow Model Registry
Repository of named, versioned
models with comments & tags
Track each model’s stage: dev,
staging, production, archived
Easily load a specific version

Model Registry Workflow
Model Registry
MODEL
DEVELOPER
DOWNSTREAM
USERS
AUTOMATED JOBS
REST SERVING
REVIEWERS,
CI/CD TOOLS

Model Registry Availability
Pull request available: tinyurl.com/registry-pr
Available to Databricks customers

Modeling wind power availability
Weather
forecast
Power
forecast
ML model

Modeling wind power availability
Hourly
job
Weather
forecast
Power
forecast
ML model

MLflow Model Registry
Weather
forecast
Power
forecast
Easy to deploy bad models
Limited model information
No audit trail
Model administration and review
Model tracking with MLflow
Centralized activity logs and comments

Getting Started with
pip install mlflow
Docs and tutorials: mlflow.org
Databricks Community Edition: databricks.com/try

Simplifying Model Management with MLflow

What's hot

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

Managing the Complete Machine Learning Lifecycle with MLflow

Databricks

What is MLOps

Henrik Skogström

From Data Science to MLOps

Carl W. Handlin

MLflow with Databricks

Liangjun Jiang

Because MLflow is an API-first platform, there are many patterns for using it in complex workflows and integrating it with existing tools. In this talk, we’ll demo a few best practices for using MLflow in a more complex workflow. These include: * Run multi-step workflows on MLflow, such as data preparation steps followed by training, and organizing your projects so you can automatically reuse past work. * Tune Hyperparameter on MLflow with open source hyperparameter tuning packages. * Save a model in MLflow (eg, from a new machine learning library) and deploying it to the existing deployment tools.

Advanced MLflow: Multi-Step Workflows, Hyperparameter Tuning and Integrating ...

Databricks

Specialized tools for machine learning development and model governance are becoming essential. MlFlow is an open source platform for managing the machine learning lifecycle. Just by adding a few lines of code in the function or script that trains their model, data scientists can log parameters, metrics, artifacts (plots, miscellaneous files, etc.) and a deployable packaging of the ML model. Every time that function or script is run, the results will be logged automatically as a byproduct of those lines of code being added, even if the party doing the training run makes no special effort to record the results. MLflow application programming interfaces (APIs) are available for the Python, R and Java programming languages, and MLflow sports a language-agnostic REST API as well. Over a relatively short time period, MLflow has garnered more than 3,300 stars on GitHub , almost 500,000 monthly downloads and 80 contributors from more than 40 companies. Most significantly, more than 200 companies are now using MLflow. We will demo MlFlow Tracking , Project and Model components with Azure Machine Learning (AML) Services and show you how easy it is to get started with MlFlow on-prem or in the cloud.

Introducing MlFlow: An Open Source Platform for the Machine Learning Lifecycl...

DataWorks Summit

MLOps in action

Pieter de Bruin

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 for production-level machine learning

cnvrg.io AI OS - Hands-on ML Workshops

Apply MLOps at Scale by H&M

Databricks

Ml ops past_present_future

Nisha Talagala

Managing the Machine Learning Lifecycle with MLflow

Databricks

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.

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

Weaveworks

MLflow: A Platform for Production Machine Learning

Matei Zaharia

MLflow: Infrastructure for a Complete Machine Learning Life Cycle

Databricks

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 - The Assembly Line of ML

Jordan Birdsell

MLOps with Azure DevOps

Marco Parenzan

Machine Learning 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 address these problems, many companies are building custom “ML platforms” that automate this lifecycle, but even these platforms are limited to a few supported algorithms and to each company’s internal infrastructure. In this session, we introduce MLflow, a new open-source project from Databricks that aims to design an open ML platform where organizations can use any ML library and development tool of their choice to reliably build and share ML applications. 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.

"Managing the Complete Machine Learning Lifecycle with MLflow"

Databricks

Managing the Machine Learning Lifecycle with MLOps

Fatih Baltacı

MLOps Virtual Event: Automating ML at Scale

Databricks

MLOps Bridging the gap between Data Scientists and Ops.

Knoldus Inc.

What's hot (20)

Managing the Complete Machine Learning Lifecycle with MLflow

What is MLOps

From Data Science to MLOps

MLflow with Databricks

Advanced MLflow: Multi-Step Workflows, Hyperparameter Tuning and Integrating ...

Introducing MlFlow: An Open Source Platform for the Machine Learning Lifecycl...

MLOps in action

MLOps for production-level machine learning

Apply MLOps at Scale by H&M

Ml ops past_present_future

Managing the Machine Learning Lifecycle with MLflow

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

MLflow: A Platform for Production Machine Learning

MLflow: Infrastructure for a Complete Machine Learning Life Cycle

MLOps - The Assembly Line of ML

MLOps with Azure DevOps

"Managing the Complete Machine Learning Lifecycle with MLflow"

Managing the Machine Learning Lifecycle with MLOps

MLOps Virtual Event: Automating ML at Scale

MLOps Bridging the gap between Data Scientists and Ops.

Similar to Simplifying Model Management with MLflow

Successfully 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 45 contributors and new features new multiple language APIs, integrations with popular ML libraries, and storage backends. I’ll go through some of the newly released features and explain how to get started with MLflow.

Accelerating Production Machine Learning with MLflow with Matei Zaharia

Databricks

Scaling up Machine Learning Development

Matei Zaharia

Mlflow est un projet opensource pour administrer le cycle de vie des projets machine learning (de l’expérimentation jusqu’au déploiement) afin de mieux les intégrer dans l’écosystème qui les entoure. Durant cette présentation nous montrerons les différentes composantes de MLflow et ferons une démonstration de son utilisation à la fois dans le contexte d’une plateforme Databricks et d’un IDE local.

Utilisation de MLflow pour le cycle de vie des projet Machine learning

Paris Data Engineers !

ML-Ops how to bring your data science to production

Herman Wu

What are the Unique Challenges and Opportunities in Systems for ML?

Matei Zaharia

The data science lifecycle consists of multiple iterative steps: data collection, data cleaning/exploration, feature engineering, model training, model deployment and scoring among others. The process is often tedious and error-prone and requires considerable human effort. Apart from these challenges, when it comes to leveraging ML in enterprise applications, especially in regulated environments, the level of scrutiny for data handling, model fairness, user privacy, and debuggability is very high. In this talk, we present the basic features of Flock, an end-to-end platform that facilitates adoption of ML in enterprise applications. We refer to this new class of applications as Enterprise Grade Machine Learning (EGML). Flock leverages MLflow to simplify and automate some of the steps involved in supporting EGML applications, allowing data scientists to spend most of their time on improving their ML models. Flock makes use of MLflow for model and experiment tracking but extends and complements it by providing automatic logging, deeper integration with relational databases that often store confidential data, model optimizations and support for the ONNX model format and the ONNX Runtime for inference. We will also present our ongoing work on automatically tracking lineage between data and ML models which is crucial in regulated environments. We will showcase Flock’s features through a demo using Microsoft’s Azure Data Studio and MLflow.

Improving the Life of Data Scientists: Automating ML Lifecycle through MLflow

Databricks

As machine learning evolves from experimentation to serving production workloads, so does the need to effectively manage the end-to-end training and production workflow including model management, versioning, and serving. Clemens Mewald offers an overview of TensorFlow Extended (TFX), the end-to-end machine learning platform for TensorFlow that powers products across all of Alphabet. Many TFX components rely on the Beam SDK to define portable data processing workflows. This talk motivates the development of a Spark runner for Beam Python.

TensorFlow Extended: An End-to-End Machine Learning Platform for TensorFlow

Databricks

Track 6 Session 3_如何藉由 AWS AI 和機器學習平台搭建多功能的 AI 解決方案.pptx

Amazon Web Services

Track 6 Session 3_如何藉由 AWS AI 和機器學習平台搭建多功能的 AI 解決方案.pptx

Amazon Web Services

Machine Learning Pipelines - Joseph Bradley - Databricks

Spark Summit

201909 Automated ML for Developers

Mark Tabladillo

MLflow 1.0 is coming soon as the first stable release of MLflow. It also packs many cleanups and improvements, such as simpler metadata management, search APIs and HDFS support. In this talk, we’ll present these new features in detail, and then discuss additional MLflow components that Databricks and other companies are working on for the rest of 2019. These new tools include a model registry to share and track models, as well as a multi-step workflow abstraction, both of which were announced at Spark + AI Summit 2019.

MLFlow 1.0 Meetup

Databricks

Machine Learning for .NET Developers - ADC21

Gülden Bilgütay

This talk was given at the Cologne AI and Machine Learning Meetup on April 13, 2023 (https://www.meetup.com/de-DE/cologne-ai-and-machine-learning-meetup/events/291513393/) by Dr. Andreas Weiden, Co-Lead Cloud / Data Engineering at skillbyte: MLOps pipelines using MLFlow - From training to production In this talk we explore the world of MLOps pipelines and how MLFlow can be used to facilitate workflows for getting your machine learning models from training to production. We will briefly delve into the tracking aspects of MLFlow and how to store experiments and runs. Next, we will move on to an actual use case that involves managing artefacts generated by multiple training pipelines running on a daily schedule. These artefacts are used in prediction services but also in managed vector search engines such as ElasticSearch and Google VertexAI. A simple microservice that polls the MLFlow registry is used to update both REST-APIs running in Kubernetes and to ingest the models into the vector search services. Finally, we will compare different alternatives that were considered.

MLOps pipelines using MLFlow - From training to production

Fabian Hadiji

[REPEAT] Deep Learning Applications Using TensorFlow (AIM401-R) - AWS re:Inve...

Amazon Web Services

Successfully building a machine learning model is hard enough. Reproducing your results at scale — enabling others to reproduce pipelines, comparing results from other versions, moving models into production, redeploying and rolling out updated models — is exponentially harder. To address these challenges and accelerate innovation, many companies are building custom “ML platforms” to automate the end-to-end ML lifecycle. Watch a replay of this MLOps Virtual Event to hear more about the latest developments and best practices for managing the full ML lifecycle on Databricks with MLflow. We covered a checklist of capabilities you’ll need, common pitfalls, technological and organizational challenges, and how to overcome them. https://www.youtube.com/playlist?list=PLTPXxbhUt-YUFNBwBsSIlknoNbS7GExZw

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

Databricks

Automated machine learning (automated ML) automates feature engineering, algorithm and hyperparameter selection to find the best model for your data. The mission: Enable automated building of machine learning with the goal of accelerating, democratizing and scaling AI. This presentation covers some recent announcements of technologies related to Automated ML, and especially for Azure. The demonstrations focus on Python with Azure ML Service and Azure Databricks. This presentation is the fourth of four related to ML.NET and Automated ML. The presentation will be recorded with video posted to this YouTube Channel: http://bit.ly/2ZybKwI

201906 04 Overview of Automated ML June 2019

Mark Tabladillo

Introduction to ML.NET

Gianni Rosa Gallina

AWS re:Invent 2018 - AIM401 - Deep Learning using Tensorflow

Julien SIMON

DutchMLSchool. ML Automation

BigML, Inc

Similar to Simplifying Model Management with MLflow (20)

Accelerating Production Machine Learning with MLflow with Matei Zaharia

Scaling up Machine Learning Development

Utilisation de MLflow pour le cycle de vie des projet Machine learning

ML-Ops how to bring your data science to production

What are the Unique Challenges and Opportunities in Systems for ML?

Improving the Life of Data Scientists: Automating ML Lifecycle through MLflow

TensorFlow Extended: An End-to-End Machine Learning Platform for TensorFlow

Track 6 Session 3_如何藉由 AWS AI 和機器學習平台搭建多功能的 AI 解決方案.pptx

Machine Learning Pipelines - Joseph Bradley - Databricks

201909 Automated ML for Developers

MLFlow 1.0 Meetup

Machine Learning for .NET Developers - ADC21

MLOps pipelines using MLFlow - From training to production

[REPEAT] Deep Learning Applications Using TensorFlow (AIM401-R) - AWS re:Inve...

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

201906 04 Overview of Automated ML June 2019

Introduction to ML.NET

AWS re:Invent 2018 - AIM401 - Deep Learning using Tensorflow

DutchMLSchool. ML Automation

More from Databricks

DW Migration Webinar-March 2022.pptx

Databricks

The world of data architecture began with applications. Next came data warehouses. Then text was organized into a data warehouse. Then one day the world discovered a whole new kind of data that was being generated by organizations. The world found that machines generated data that could be transformed into valuable insights. This was the origin of what is today called the data lakehouse. The evolution of data architecture continues today. Come listen to industry experts describe this transformation of ordinary data into a data architecture that is invaluable to business. Simply put, organizations that take data architecture seriously are going to be at the forefront of business tomorrow. This is an educational event. Several of the authors of the book Building the Data Lakehouse will be presenting at this symposium.

Data Lakehouse Symposium | Day 1 | Part 1

Databricks

Data Lakehouse Symposium | Day 1 | Part 2

Databricks

Data Lakehouse Symposium | Day 2

Databricks

Data Lakehouse Symposium | Day 4

Databricks

In this session, learn how to quickly supplement your on-premises Hadoop environment with a simple, open, and collaborative cloud architecture that enables you to generate greater value with scaled application of analytics and AI on all your data. You will also learn five critical steps for a successful migration to the Databricks Lakehouse Platform along with the resources available to help you begin to re-skill your data teams.

5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop

Databricks

Bad data leads to bad decisions and broken customer experiences. Organizations depend on complete and accurate data to power their business, maintain efficiency, and uphold customer trust. With thousands of datasets and pipelines running, how do we ensure that all data meets quality standards, and that expectations are clear between producers and consumers? Investing in shared, flexible components and practices for monitoring data health is crucial for a complex data organization to rapidly and effectively scale. At Zillow, we built a centralized platform to meet our data quality needs across stakeholders. The platform is accessible to engineers, scientists, and analysts, and seamlessly integrates with existing data pipelines and data discovery tools. In this presentation, we will provide an overview of our platform’s capabilities, including: Giving producers and consumers the ability to define and view data quality expectations using a self-service onboarding portal Performing data quality validations using libraries built to work with spark Dynamically generating pipelines that can be abstracted away from users Flagging data that doesn’t meet quality standards at the earliest stage and giving producers the opportunity to resolve issues before use by downstream consumers Exposing data quality metrics alongside each dataset to provide producers and consumers with a comprehensive picture of health over time

Democratizing Data Quality Through a Centralized Platform

Databricks

Data scientists face numerous challenges throughout the data science workflow that hinder productivity. As organizations continue to become more data-driven, a collaborative environment is more critical than ever — one that provides easier access and visibility into the data, reports and dashboards built against the data, reproducibility, and insights uncovered within the data.. Join us to hear how Databricks’ open and collaborative platform simplifies data science by enabling you to run all types of analytics workloads, from data preparation to exploratory analysis and predictive analytics, at scale — all on one unified platform.

Learn to Use Databricks for Data Science

Databricks

Application performance monitoring (APM) has become the cornerstone of software engineering allowing engineering teams to quickly identify and remedy production issues. However, as the world moves to intelligent software applications that are built using machine learning, traditional APM quickly becomes insufficient to identify and remedy production issues encountered in these modern software applications. As a lead software engineer at NewRelic, my team built high-performance monitoring systems including Insights, Mobile, and SixthSense. As I transitioned to building ML Monitoring software, I found the architectural principles and design choices underlying APM to not be a good fit for this brand new world. In fact, blindly following APM designs led us down paths that would have been better left unexplored. In this talk, I draw upon my (and my team’s) experience building an ML Monitoring system from the ground up and deploying it on customer workloads running large-scale ML training with Spark as well as real-time inference systems. I will highlight how the key principles and architectural choices of APM don’t apply to ML monitoring. You’ll learn why, understand what ML Monitoring can successfully borrow from APM, and hear what is required to build a scalable, robust ML Monitoring architecture.

Why APM Is Not the Same As ML Monitoring

Databricks

Autonomy and ownership are core to working at Stitch Fix, particularly on the Algorithms team. We enable data scientists to deploy and operate their models independently, with minimal need for handoffs or gatekeeping. By writing a simple function and calling out to an intuitive API, data scientists can harness a suite of platform-provided tooling meant to make ML operations easy. In this talk, we will dive into the abstractions the Data Platform team has built to enable this. We will go over the interface data scientists use to specify a model and what that hooks into, including online deployment, batch execution on Spark, and metrics tracking and visualization.

The Function, the Context, and the Data—Enabling ML Ops at Stitch Fix

Databricks

In this talk, I will dive into the stage level scheduling feature added to Apache Spark 3.1. Stage level scheduling extends upon Project Hydrogen by improving big data ETL and AI integration and also enables multiple other use cases. It is beneficial any time the user wants to change container resources between stages in a single Apache Spark application, whether those resources are CPU, Memory or GPUs. One of the most popular use cases is enabling end-to-end scalable Deep Learning and AI to efficiently use GPU resources. In this type of use case, users read from a distributed file system, do data manipulation and filtering to get the data into a format that the Deep Learning algorithm needs for training or inference and then sends the data into a Deep Learning algorithm. Using stage level scheduling combined with accelerator aware scheduling enables users to seamlessly go from ETL to Deep Learning running on the GPU by adjusting the container requirements for different stages in Spark within the same application. This makes writing these applications easier and can help with hardware utilization and costs. There are other ETL use cases where users want to change CPU and memory resources between stages, for instance there is data skew or perhaps the data size is much larger in certain stages of the application. In this talk, I will go over the feature details, cluster requirements, the API and use cases. I will demo how the stage level scheduling API can be used by Horovod to seamlessly go from data preparation to training using the Tensorflow Keras API using GPUs. The talk will also touch on other new Apache Spark 3.1 functionality, such as pluggable caching, which can be used to enable faster dataframe access when operating from GPUs.

Stage Level Scheduling Improving Big Data and AI Integration

Databricks

In this talk, I would like to introduce an open-source tool built by our team that simplifies the data conversion from Apache Spark to deep learning frameworks. Imagine you have a large dataset, say 20 GBs, and you want to use it to train a TensorFlow model. Before feeding the data to the model, you need to clean and preprocess your data using Spark. Now you have your dataset in a Spark DataFrame. When it comes to the training part, you may have the problem: How can I convert my Spark DataFrame to some format recognized by my TensorFlow model? The existing data conversion process can be tedious. For example, to convert an Apache Spark DataFrame to a TensorFlow Dataset file format, you need to either save the Apache Spark DataFrame on a distributed filesystem in parquet format and load the converted data with third-party tools such as Petastorm, or save it directly in TFRecord files with spark-tensorflow-connector and load it back using TFRecordDataset. Both approaches take more than 20 lines of code to manage the intermediate data files, rely on different parsing syntax, and require extra attention for handling vector columns in the Spark DataFrames. In short, all these engineering frictions greatly reduced the data scientists’ productivity. The Databricks Machine Learning team contributed a new Spark Dataset Converter API to Petastorm to simplify these tedious data conversion process steps. With the new API, it takes a few lines of code to convert a Spark DataFrame to a TensorFlow Dataset or a PyTorch DataLoader with default parameters. In the talk, I will use an example to show how to use the Spark Dataset Converter to train a Tensorflow model and how simple it is to go from single-node training to distributed training on Databricks.

Simplify Data Conversion from Spark to TensorFlow and PyTorch

Databricks

There is no doubt Kubernetes has emerged as the next generation of cloud native infrastructure to support a wide variety of distributed workloads. Apache Spark has evolved to run both Machine Learning and large scale analytics workloads. There is growing interest in running Apache Spark natively on Kubernetes. By combining the flexibility of Kubernetes and scalable data processing with Apache Spark, you can run any data and machine pipelines on this infrastructure while effectively utilizing resources at disposal. In this talk, Rajesh Thallam and Sougata Biswas will share how to effectively run your Apache Spark applications on Google Kubernetes Engine (GKE) and Google Cloud Dataproc, orchestrate the data and machine learning pipelines with managed Apache Airflow on GKE (Google Cloud Composer). Following topics will be covered: – Understanding key traits of Apache Spark on Kubernetes- Things to know when running Apache Spark on Kubernetes such as autoscaling- Demonstrate running analytics pipelines on Apache Spark orchestrated with Apache Airflow on Kubernetes cluster.

Scaling your Data Pipelines with Apache Spark on Kubernetes

Databricks

Pipelines have become ubiquitous, as the need for stringing multiple functions to compose applications has gained adoption and popularity. Common pipeline abstractions such as “fit” and “transform” are even shared across divergent platforms such as Python Scikit-Learn and Apache Spark. Scaling pipelines at the level of simple functions is desirable for many AI applications, however is not directly supported by Ray’s parallelism primitives. In this talk, Raghu will describe a pipeline abstraction that takes advantage of Ray’s compute model to efficiently scale arbitrarily complex pipeline workflows. He will demonstrate how this abstraction cleanly unifies pipeline workflows across multiple platforms such as Scikit-Learn and Spark, and achieves nearly optimal scale-out parallelism on pipelined computations. Attendees will learn how pipelined workflows can be mapped to Ray’s compute model and how they can both unify and accelerate their pipelines with Ray.

Scaling and Unifying SciKit Learn and Apache Spark Pipelines

Databricks

In this talk about zipline, we will introduce a new type of windowing construct called a sawtooth window. We will describe various properties about sawtooth windows that we utilize to achieve online-offline consistency, while still maintaining high-throughput, low-read latency and tunable write latency for serving machine learning features.We will also talk about a simple deployment strategy for correcting feature drift – due operations that are not “abelian groups”, that operate over change data.

Sawtooth Windows for Feature Aggregations

Databricks

We want to present multiple anti patterns utilizing Redis in unconventional ways to get the maximum out of Apache Spark.All examples presented are tried and tested in production at Scale at Adobe. The most common integration is spark-redis which interfaces with Redis as a Dataframe backing Store or as an upstream for Structured Streaming. We deviate from the common use cases to explore where Redis can plug gaps while scaling out high throughput applications in Spark. Niche 1 : Long Running Spark Batch Job – Dispatch New Jobs by polling a Redis Queue · Why? o Custom queries on top a table; We load the data once and query N times · Why not Structured Streaming · Working Solution using Redis Niche 2 : Distributed Counters · Problems with Spark Accumulators · Utilize Redis Hashes as distributed counters · Precautions for retries and speculative execution · Pipelining to improve performance

Redis + Apache Spark = Swiss Army Knife Meets Kitchen Sink

Databricks

In the era of microservices, decentralized ML architectures and complex data pipelines, data quality has become a bigger challenge than ever. When data is involved in complex business processes and decisions, bad data can, and will, affect the bottom line. As a result, ensuring data quality across the entire ML pipeline is both costly, and cumbersome while data monitoring is often fragmented and performed ad hoc. To address these challenges, we built whylogs, an open source standard for data logging. It is a lightweight data profiling library that enables end-to-end data profiling across the entire software stack. The library implements a language and platform agnostic approach to data quality and data monitoring. It can work with different modes of data operations, including streaming, batch and IoT data. In this talk, we will provide an overview of the whylogs architecture, including its lightweight statistical data collection approach and various integrations. We will demonstrate how the whylogs integration with Apache Spark achieves large scale data profiling, and we will show how users can apply this integration into existing data and ML pipelines.

Re-imagine Data Monitoring with whylogs and Spark

Databricks

Machine learning (ML) models are typically part of prediction queries that consist of a data processing part (e.g., for joining, filtering, cleaning, featurization) and an ML part invoking one or more trained models. In this presentation, we identify significant and unexplored opportunities for optimization. To the best of our knowledge, this is the first effort to look at prediction queries holistically, optimizing across both the ML and SQL components. We will present Raven, an end-to-end optimizer for prediction queries. Raven relies on a unified intermediate representation that captures both data processing and ML operators in a single graph structure. This allows us to introduce optimization rules that (i) reduce unnecessary computations by passing information between the data processing and ML operators (ii) leverage operator transformations (e.g., turning a decision tree to a SQL expression or an equivalent neural network) to map operators to the right execution engine, and (iii) integrate compiler techniques to take advantage of the most efficient hardware backend (e.g., CPU, GPU) for each operator. We have implemented Raven as an extension to Spark’s Catalyst optimizer to enable the optimization of SparkSQL prediction queries. Our implementation also allows the optimization of prediction queries in SQL Server. As we will show, Raven is capable of improving prediction query performance on Apache Spark and SQL Server by up to 13.1x and 330x, respectively. For complex models, where GPU acceleration is beneficial, Raven provides up to 8x speedup compared to state-of-the-art systems. As part of the presentation, we will also give a demo showcasing Raven in action.

Raven: End-to-end Optimization of ML Prediction Queries

Databricks

Semantic segmentation is the classification of every pixel in an image/video. The segmentation partitions a digital image into multiple objects to simplify/change the representation of the image into something that is more meaningful and easier to analyze [1][2]. The technique has a wide variety of applications ranging from perception in autonomous driving scenarios to cancer cell segmentation for medical diagnosis. Exponential growth in the datasets that require such segmentation is driven by improvements in the accuracy and quality of the sensors generating the data extending to 3D point cloud data. This growth is further compounded by exponential advances in cloud technologies enabling the storage and compute available for such applications. The need for semantically segmented datasets is a key requirement to improve the accuracy of inference engines that are built upon them. Streamlining the accuracy and efficiency of these systems directly affects the value of the business outcome for organizations that are developing such functionalities as a part of their AI strategy. This presentation details workflows for labeling, preprocessing, modeling, and evaluating performance/accuracy. Scientists and engineers leverage domain-specific features/tools that support the entire workflow from labeling the ground truth, handling data from a wide variety of sources/formats, developing models and finally deploying these models. Users can scale their deployments optimally on GPU-based cloud infrastructure to build accelerated training and inference pipelines while working with big datasets. These environments are optimized for engineers to develop such functionality with ease and then scale against large datasets with Spark-based clusters on the cloud.

Processing Large Datasets for ADAS Applications using Apache Spark

Databricks

At Adobe Experience Platform, we ingest TBs of data every day and manage PBs of data for our customers as part of the Unified Profile Offering. At the heart of this is a bunch of complex ingestion of a mix of normalized and denormalized data with various linkage scenarios power by a central Identity Linking Graph. This helps power various marketing scenarios that are activated in multiple platforms and channels like email, advertisements etc. We will go over how we built a cost effective and scalable data pipeline using Apache Spark and Delta Lake and share our experiences. What are we storing? Multi Source – Multi Channel Problem Data Representation and Nested Schema Evolution Performance Trade Offs with Various formats Go over anti-patterns used (String FTW) Data Manipulation using UDFs Writer Worries and How to Wipe them Away Staging Tables FTW Datalake Replication Lag Tracking Performance Time!

Massive Data Processing in Adobe Using Delta Lake

Databricks

More from Databricks (20)

DW Migration Webinar-March 2022.pptx

Data Lakehouse Symposium | Day 1 | Part 1

Data Lakehouse Symposium | Day 1 | Part 2

Data Lakehouse Symposium | Day 2

Data Lakehouse Symposium | Day 4

5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop

Democratizing Data Quality Through a Centralized Platform

Learn to Use Databricks for Data Science

Why APM Is Not the Same As ML Monitoring

The Function, the Context, and the Data—Enabling ML Ops at Stitch Fix

Stage Level Scheduling Improving Big Data and AI Integration

Simplify Data Conversion from Spark to TensorFlow and PyTorch

Scaling your Data Pipelines with Apache Spark on Kubernetes

Scaling and Unifying SciKit Learn and Apache Spark Pipelines

Sawtooth Windows for Feature Aggregations

Redis + Apache Spark = Swiss Army Knife Meets Kitchen Sink

Re-imagine Data Monitoring with whylogs and Spark

Raven: End-to-end Optimization of ML Prediction Queries

Processing Large Datasets for ADAS Applications using Apache Spark

Massive Data Processing in Adobe Using Delta Lake

Recently uploaded

Gen AI on Enterprise Cloud Apache NiFi Milvus Apache Kafka Apache Flink Cloudera Machine Learning Cloudera DataFlow https://medium.com/@tspann/building-a-milvus-connector-for-nifi-34372cb3c7fa https://www.meetup.com/futureofdata-princeton/events/300737266/ https://lu.ma/q7pcfyjn?source=post_page-----34372cb3c7fa--------------------------------&tk=TTyakY If you're interested in working with Generative AI on the cloud, this virtual workshop is for you. Tim Spann from Cloudera and Yujian Tang from Zilliz will cover how you can implement your own GenAI workflows on the cloud at enterprise scale. 9:00 - 9:05: Intro 9:05 - 9:15: What is Milvus 9:15 - 9:25: Cloudera Development Platform 9:25 - 10:00: Demo Location https://www.youtube.com/watch?v=IfWIzKsoHnA https://github.com/tspannhw/SpeakerProfile https://www.linkedin.com/in/yujiantang/

Generative AI on Enterprise Cloud with NiFi and Milvus

Timothy Spann

Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Service Bangalore Booking Contact Details :- WhatsApp Chat :- +91-7737669865 2-May-2024(SMW) Call Girls In Model Towh Bangalore +91-7737669865 !! Best Woman Seeking Man Call Girls Service, Escorts Service in Home Hotel in Bangalore NCR 24 Hours Available Service Call Girls, Contact Us +91-7737669865 (Any Time. Any Where) Call Girls in Bangalore, Noida, Gurgaon, Ghaziabad,Sexy Indian Female Escorts Service Bangalore NCRWelcome To Bangalore Escorts Service – An All Over New Bangalore Very Sexy Hot Call Girls Agency Service Escorts In South BangaloreNCRBangalore’s No. 1 High Profile Independent Female Escorts Service. We Provide Good Quality Educated Profile At Very Regnebal Price 100% Safe And Original.We Are Provide Escorts Service All OYO Hotels ,3*,4*,5* Star Hotel And Home Flat, Apartment. Guest-House. Services In -Call And Out – Call Both Are Services Available. 24Hrs. Any Time Any Where. In All Over Bangalore Noida Gurgaon Ghaziabad Faridabad.More Information And Contact Profile Real Pic Visit Our Website City Wise Escorts Service Agency.Good Looking Cheap And Best Models Girls U Can Get Best Click On Link……Night Call Girls Now In Hotel Le Meridien Gurgaon Near Female Escort One Shot — 5000/in call (time 1 hour), 6000/out call Two shot with one girl — 8000/in call (time 2 hour), 10000/out call Body to body massage with sex- 8000/in call (time 1 hour) Full night Service for one person– 12000/in call, 13000/out call (shot limit 3-4 shots) Full night Service for more than 1 person — please contact Us —7737669865 We are available 24*7 all days of the year. Call us — 7737669865 Thank you for Visiting.

Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...

amitlee9823

Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call Booking Contact Details :- WhatsApp Chat :- +91-9711199171 27-April-2024(SMW) Best Escorts Service in Delhi Call Us — 9711199171 —Are you Looking for Call Girls in Delhi then This's a perfect place for you. We brings Many Models and Independent Call girls who Can fulfill all your Sexual Desires by Their Complete package of sex Service and It's Absolutely safe and secure. Book Your One night Stand Call Girls : 9711199171 -SERVICES- Our Delhi Call Girls are fully cooperative and understand your needs and they Give you their complete package of Sex service like- Real girl friend experience, Lip Kiss, Lip Lock, Smooch, Sucking without Condom, Oral, Erotic Massage, Lap Dance, Threesome, 69 Licking, Sex in all Position, Anal Sex etc, -AVAILABLE GIRLS- We Bring Many Good looking Slim, Busty, Hot and Sexy Call Girls as per our client's Choice like Housewives, College girls, Russian girls, Muslim girls, Afghani girls, Bengali girls, Working girls, south Indian girls, Punjabi girls, Big Boobs Call Girls etc, -PAYMENT METHOD- C O D. If you Want to Book our Call girls Service in Delhi . We don,t ask anything in Advance, you Can simply pay to her Cash on Delivery or via any upi. In-Call: — You can reach at our Place in Delhi Our place Hotel and Flat Which Is Very Clean Hygiene And 100% safe Accommodation. Out-Call: — You Can pick up the Girl from my Place if not we Provide Door-Step Call Girls Delivery in Delhi. NOTE: — Pic Collectors Time Passers and Bargainers please Stay Away As We Respect The Value For Your Money Time And Expect The Same From You. OUR SERVICES RATE: – let you know the our Call girls price in Delhi are very affordable and Best rate in genuine market of Call girls. But as all you knows that Quality has comes with its own Price tag. One Shot — 5000/in call (time 1 hour), 6000/out call Two shot with one girl — 8000/in call (time 2 hour), 10000/out call Body to body massage with sex- 8000/in call (time 1 hour) Full night Service for one person– 12000/in call, 13000/out call (shot limit 3-4 shots) Full night Service for more than 1 person — please contact Us —9711199171 We are available 24*7 all days of the year. Call us — 9711199171 Thank you for Visiting.

Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call

shivangimorya083

Determinants of health, dimensions of health, positive health and spectrum of...

shambhavirathore45

BabyOno dropshipping via API with DroFx.pptx

olyaivanovalion

Week-01-2.ppt BBB human Computer interaction

fulawalesam

(NEHA) Call Girls Katra Call Now: 8617697112 Katra Escorts Booking Contact Details WhatsApp Chat: +91-8617697112 Katra Escort Service includes providing maximum physical satisfaction to their clients as well as engaging conversation that keeps your time enjoyable and entertaining. Plus, they look fabulously elegant, making an impression. Independent Escorts Katra understands the value of confidentiality and discretion; they will go the extra mile to meet your needs. Simply contact them via text messaging or through their online profiles; they'd be more than delighted to accommodate any request or arrange a romantic date or fun-filled night together. We provide:

(NEHA) Call Girls Katra Call Now 8617697112 Katra Escorts 24x7

Call Girls in Nagpur High Profile Call Girls

VidaXL dropshipping via API with DroFx.pptx

olyaivanovalion

(Vivek)Call Us, 8448380779,Call girls in Delhi NCr – We Offer best in class call girls. escort Service At Affordable Price At low Rate with Space Night 8000 We Are One Of The Oldest Escort and Call girls Agencies in Delhi. You Will Find That Our Female Escorts Are Full Of Fun, Sexy And They Would Love Enjoy Your Company. We Have A Fantastic Selection Of Escort Ladies Available For In-Calls As Well As Out-Calls. Our Escorts Are Not Only Beautiful But All Have Great Personalities Making Them The Perfect Companion For Any Occasion. In-Call:- You Can Come At Our Place in Delhi Our place Which Is Very Clean Hygienic 100% safe Accommodation. Out-Call:- You have To Come Pick The Girl From My Place We Are Also Provide Door Step Services (Delhi Ncr, Noida, Gurgaon, Faridabad, Ghaziabad Note:- Pic Collectors Time Passers Bargainers Stay Away As We Respect The Value For Your Money Time And Expect The Same From You Hygienic:- Full Ac room And Clean Rooms Available In Hotel 24 * 7 Hourly In Delhi NCR More Details, With WhatsApp Number, +91-8448380779

BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service

Delhi Call girls

BPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptx

MohammedJunaid861692

Accredited-Transport-Cooperatives-Jan-2021-Web.pdf

adriantubila

Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Bangalore Booking Contact Details :- WhatsApp Chat :- +91-7737669865 2-May-2024(SMW) Call Girls In Model Towh Bangalore +91-7737669865 !! Best Woman Seeking Man Call Girls Service, Escorts Service in Home Hotel in Bangalore NCR 24 Hours Available Service Call Girls, Contact Us +91-7737669865 (Any Time. Any Where) Call Girls in Bangalore, Noida, Gurgaon, Ghaziabad,Sexy Indian Female Escorts Service Bangalore NCRWelcome To Bangalore Escorts Service – An All Over New Bangalore Very Sexy Hot Call Girls Agency Service Escorts In South BangaloreNCRBangalore’s No. 1 High Profile Independent Female Escorts Service. We Provide Good Quality Educated Profile At Very Regnebal Price 100% Safe And Original.We Are Provide Escorts Service All OYO Hotels ,3*,4*,5* Star Hotel And Home Flat, Apartment. Guest-House. Services In -Call And Out – Call Both Are Services Available. 24Hrs. Any Time Any Where. In All Over Bangalore Noida Gurgaon Ghaziabad Faridabad.More Information And Contact Profile Real Pic Visit Our Website City Wise Escorts Service Agency.Good Looking Cheap And Best Models Girls U Can Get Best Click On Link……Night Call Girls Now In Hotel Le Meridien Gurgaon Near Female Escort One Shot — 5000/in call (time 1 hour), 6000/out call Two shot with one girl — 8000/in call (time 2 hour), 10000/out call Body to body massage with sex- 8000/in call (time 1 hour) Full night Service for one person– 12000/in call, 13000/out call (shot limit 3-4 shots) Full night Service for more than 1 person — please contact Us —7737669865 We are available 24*7 all days of the year. Call us — 7737669865 Thank you for Visiting.

Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...

amitlee9823

Edukaciniai dropshipping via API with DroFx

olyaivanovalion

Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure

Pooja Nehwal

Introduction-to-Machine-Learning (1).pptx

firstjob4

April 2024 - Crypto Market Report's Analysis

manisha194592

Sampling (random) method and Non random.ppt

Dr. Soumendra Kumar Patra

Digital advertising, or paid media, encompasses the strategic deployment of online advertisements to reach target audiences efficiently and effectively. This includes any digital platform that supports advertising to deliver unique messages for any objective. Understanding the mechanics of digital advertising platforms, along with insights into audience behaviors and preferences, allows marketers to optimize their ad spend and achieve significant engagement and conversion rates. This lecture is for Advanced Digital & Social Media Strategy (MGMTX 466.05) at UCLA Extension.

Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...

Valters Lauzums

Discover Why Less is More in B2B Research

michael115558

Midocean dropshipping via API with DroFx

olyaivanovalion

Recently uploaded (20)

Generative AI on Enterprise Cloud with NiFi and Milvus

Call Girls Bannerghatta Road Just Call 👗 7737669865 👗 Top Class Call Girl Ser...

Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call

Determinants of health, dimensions of health, positive health and spectrum of...

BabyOno dropshipping via API with DroFx.pptx

Week-01-2.ppt BBB human Computer interaction

(NEHA) Call Girls Katra Call Now 8617697112 Katra Escorts 24x7

VidaXL dropshipping via API with DroFx.pptx

BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service

BPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptx

Accredited-Transport-Cooperatives-Jan-2021-Web.pdf

Call Girls Hsr Layout Just Call 👗 7737669865 👗 Top Class Call Girl Service Ba...

Edukaciniai dropshipping via API with DroFx

Call me @ 9892124323 Cheap Rate Call Girls in Vashi with Real Photo 100% Secure

Introduction-to-Machine-Learning (1).pptx

April 2024 - Crypto Market Report's Analysis

Sampling (random) method and Non random.ppt

Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...

Discover Why Less is More in B2B Research

Midocean dropshipping via API with DroFx

Simplifying Model Management with MLflow

1. Simplifying ML Model Management with .

2. ML development is harder than traditional software development

3. Traditional Software Machine Learning Goal: Optimize a metric (e.g., accuracy) • Constantly experiment to improve it Quality depends on input data and tuning parameters Compare + combine many libraries, models & algorithms for the same task Goal: Meet a functional specification Quality depends only on code Typically pick one software stack

4. Production ML is Even Harder Data Prep Training Deployment Raw Data ML ENGINEER APPLICATION DEVELOPER DATA ENGINEER ML apps must be fed new data to keep working Design, retraining & inference done by different people

5. Solution: Machine Learning Platforms Software to manage the ML lifecycle Examples: Uber Michelangelo, Google TFX, Facebook FBLearner Data Prep Training Deployment Raw Data Versioning, CI/CD, QA, ops, monitoring, etc

6. MLflow: An Open Source ML Platform Three components: • Tracking: experiment tracking • Projects: reproducible runs • Models: model packaging 140 contributors, 800K downloads/month Works with any ML library, programming language, deployment tool

7. MLflow Tracking: Experiments Notebooks Local Apps Cloud Jobs Tracking Server UI API mlflow.log_param(“alpha”, 0.5) mlflow.log_metric(“accuracy”, 0.9) ... REST API

8. Tracking UI: Inspecting Runs

9. MLflow Projects: Reproducible Runs Project Spec Code DataConfig Local Execution Remote Cluster MLflow Models: Model Packaging Model Format ONNX Flavor Python Flavor Model Logic Batch & Stream Scoring REST Serving Packaging Format . . . Evaluation & Debugging LIME TCAV

10. MLflow Talks at This Summit

11. New in Last 6 Months MLflow 1.0 (and 1.1, 1.2, 1.3) Autologging in TensorFlow & Keras DataFrame search API Kubernetes, HDFS & Seldon integrations

12. MLflow Autologging model = keras.models.Sequential() model.add(layers.Dense(hidden_units, ...)) model.fit(X_train, y_train) test_loss = model.evaluate(X_test, y_test)

13. MLflow Autologging with mlflow.start_run(): model = keras.models.Sequential() model.add(layers.Dense(hidden_units, ...)) model.fit(X_train, y_train) test_loss = model.evaluate(X_test, y_test) mlflow.log_param(“hidden_units”, hidden_units) mlflow.log_param(“learning_rate”, learning_rate) mlflow.log_metric(“train_loss”, train_loss) mlflow.log_metric(“test_loss”, test_loss) mlflow.keras.log_model(model) mlflow.keras.autolog() model = keras.models.Sequential() model.add(layers.Dense(hidden_units, ...)) model.fit(X_train, y_train) test_loss = model.evaluate(X_test, y_test)

14. MLflow’s Next Goal: Model Management

15. The Model Management Problem When you’re working on one ML app alone, storing your models in files is manageable MODEL DEVELOPER classifier_v1.h5 classifier_v2.h5 classifier_v3_sept_19.h5 classifier_v3_new.h5 …

16. The Model Management Problem When you work in a large organization with many models, management becomes a major challenge: • Where can I find the best version of this model? • How was this model trained? • How can I track docs for each model? • How can I review models? MODEL DEVELOPER REVIEWER MODEL USER ???

17. MLflow Model Registry Repository of named, versioned models with comments & tags Track each model’s stage: dev, staging, production, archived Easily load a specific version

18. Model Registry Workflow Model Registry MODEL DEVELOPER DOWNSTREAM USERS AUTOMATED JOBS REST SERVING REVIEWERS, CI/CD TOOLS

19. Model Registry Availability Pull request available: tinyurl.com/registry-pr Available to Databricks customers

20.

21. Wind directionWind speed Power + =

22.

23. Modeling wind power availability Weather forecast Power forecast ML model

24. Modeling wind power availability Hourly job Weather forecast Power forecast ML model

25. MLflow Model Registry Weather forecast Power forecast Easy to deploy bad models Limited model information No audit trail Model administration and review Model tracking with MLflow Centralized activity logs and comments

26. Thank you

27. Getting Started with pip install mlflow Docs and tutorials: mlflow.org Databricks Community Edition: databricks.com/try

Simplifying Model Management with MLflow

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Simplifying Model Management with MLflow

Similar to Simplifying Model Management with MLflow (20)

More from Databricks

More from Databricks (20)

Recently uploaded

Recently uploaded (20)

Simplifying Model Management with MLflow