Serverless machine learning operations
4 likes964 views
Any startup has to have a clear go-to-market strategy from the beginning. Similarly, any data science project has to have a go-to-production strategy from its first days, so it could go beyond proof-of-concept. Machine learning and artificial intelligence in production would result in hundreds of training pipelines and machine learning models that are continuously revised by teams of data scientists and seamlessly connected with web applications for tenants and users. In this demo-based talk we will walk through the best practices for simplifying machine learning operations across the enterprise and providing a serverless abstraction for data scientists and data engineers, so they could train, deploy and monitor machine learning models faster and with better quality.
1 of 40
Downloaded 47 times
Ad
Recommended
Multi runtime serving pipelines for machine learning
Multi runtime serving pipelines for machine learningStepan Pushkarev The talk I gave at Scale By The Bay.
Deploying, Serving and monitoring machine learning models built with different ML frameworks in production. Envoy proxy powered serving mesh. TensorFlow, Spark ML, Scikit-learn and custom functions on CPU and GPU.
Spark ML Pipeline serving
Spark ML Pipeline servingStepan Pushkarev My talk at Data Science Labs conference in Odessa.
Training a model in Apache Spark while having it automatically available for real-time serving is an essential feature for end-to-end solutions.
There is an option to export the model into PMML and then import it into a separated scoring engine. The idea of interoperability is great but it has multiple challenges, such as code duplication, limited extensibility, inconsistency, extra moving parts. In this talk we discussed an alternative solution that does not introduce custom model formats and new standards, not based on export/import workflow and shares Apache Spark API.
Spark and machine learning in microservices architecture
Spark and machine learning in microservices architectureStepan Pushkarev Modern architecture for machine learning and state-safe data processing.
Data Pipelines and machine learning models deployment in production.
Using PySpark to Process Boat Loads of Data
Using PySpark to Process Boat Loads of DataRobert Dempsey Learn how to use PySpark for processing massive amounts of data. Combined with the GitHub repo - https://github.com/rdempsey/pyspark-for-data-processing - this presentation will help you gain familiarity with processing data using Python and Spark.
If you're thinking about machine learning and not sure if it can help improve your business, but want to find out, set up a free 20-minute consultation with us: https://calendly.com/robertwdempsey/free-consultation
Next.ml Boston: Data Science Dev Ops
Next.ml Boston: Data Science Dev OpsEric Chiang This document discusses challenges in deploying machine learning models into production and potential solutions. It covers:
1. Issues with reproducibility due to dependencies and environment configurations when models are trained and deployed.
2. Problems with serializing models and transferring them between different versions of libraries and software stacks.
3. How containers can help address these issues by encapsulating the full runtime environment and dependencies of a model.
4. Managing both models and Docker containers is still required when using this approach.
Hamburg Data Science Meetup - MLOps with a Feature Store
Hamburg Data Science Meetup - MLOps with a Feature StoreMoritz Meister MLOps is a trend in machine learning (ML) engineering that unifies ML system development (Dev) and ML system operation (Ops). Some ML lifecycle frameworks, such as TensorFlow Extended, are based around end-to-end pipelines that start with raw data and end in production models. During this talk we will introduce the concept of a feature store as the missing piece of ML infrastructure that enables faster lower cost deployment of models. We will show how the Hopsworks Feature Store - factors monolithic end-to-end ML pipelines into feature and model training pipelines that can each run at different cadences. We will show examples of ingestion and training pipelines including hyperparameter optimization and model deployment.
Apache Liminal (Incubating)—Orchestrate the Machine Learning Pipeline
Apache Liminal (Incubating)—Orchestrate the Machine Learning PipelineDatabricks Apache Liminal is an end-to-end platform for data engineers & scientists, allowing them to build, train and deploy machine learning models in a robust and agile way. The platform provides the abstractions and declarative capabilities for data extraction & feature engineering followed by model training and serving; using standard tools and libraries (e.g. Airflow, K8S, Spark, scikit-learn, etc.).
“Houston, we have a model...” Introduction to MLOps
“Houston, we have a model...” Introduction to MLOpsRui Quintino 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.
Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning Models
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.
MLOps with a Feature Store: Filling the Gap in ML Infrastructure
MLOps with a Feature Store: Filling the Gap in ML InfrastructureData Science Milan A Feature Store enables machine learning (ML) features to be registered, discovered, and used as part of ML pipelines, thus making it easier to transform and validate the training data that is fed into machine learning systems. Feature stores can also enable consistent engineering of features between training and inference, but to do so, they need a common data processing platform. The first Feature Stores, developed at hyperscale AI companies such as Uber, Airbnb, and Facebook, enabled feature engineering using domain specific languages, providing abstractions tailored to the companies’ feature engineering domains. However, a general purpose Feature Store needs a general purpose feature engineering, feature selection, and feature transformation platform.
In this talk, we describe how we built a general purpose, open-source Feature Store for ML around dataframes and Apache Spark. We will demonstrate how data engineers can transform and engineers features from backend databases and data lakes, while data scientists can use PySpark to select and transform features into train/test data in a file format of choice (.tfrecords, .npy, .petastorm, etc) on a file system of choice (S3, HDFS). Finally, we will show how the Feature Store enables end-to-end ML pipelines to be factored into feature engineering and data science stages that each can run at different cadences.
Bio:
Fabio Buso is the head of engineering at Logical Clocks AB, where he leads the Feature Store development. Fabio holds a master's degree in cloud computing and services with a focus on data intensive applications, awarded by a joint program between KTH Stockholm and TU Berlin.
Topics: feature store, MLOps.
MLOps - Build pipelines with Tensor Flow Extended & Kubeflow
MLOps - Build pipelines with Tensor Flow Extended & KubeflowJan Kirenz Machine Learning Operations (MLOps) for Machine Learning Lifecycle Management: Building pipelines with Tensor Flow Extended (TFX) and Kubeflow.
MLOps and Data Quality: Deploying Reliable ML Models in Production
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
MLOps in action
MLOps in actionPieter de Bruin Machine Learning operations brings data science to the world of devops. Data scientists create models on their workstations. MLOps adds automation, validation and monitoring to any environment including machine learning on kubernetes. In this session you hear about latest developments and see it in action.
Moving a Fraud-Fighting Random Forest from scikit-learn to Spark with MLlib, ...
Moving a Fraud-Fighting Random Forest from scikit-learn to Spark with MLlib, ...Databricks This talk describes migrating a large random forest classifier from scikit-learn to Spark's MLlib. We cut training time from 2 days to 2 hours, reduced failed runs, and track experiments better with MLflow. Kount provides certainty in digital interactions like online credit card transactions. One of our scores uses a random forest classifier with 250 trees and 100,000 nodes per tree. We used scikit-learn to train using 60 million samples that each contained over 150 features. The in-memory requirements exceeded 750 GB, took 2 days, and were not robust to disruption in our database or training execution. To migrate workflow to Spark, we built a 6-node cluster with HDFS. This provides 1.35 TB of RAM and 484 cores. Using MLlib and parallelization, the training time for our random forests are now less than 2 hours. Training data stays in our production environment, which used to require a deploy cycle to move locally-developed code onto our training server. The new implementation uses Jupyter notebooks for remote development with server-side execution. MLflow tracks all input parameters, code, and git revision number, while the performance and model itself are retained as experiment artifacts. The new workflow is robust to service disruption. Our training pipeline begins by pulling from a Vertica database. Originally, this single connection took over 8 hours to complete with any problem causing a restart. Using sqoop and multiple connections, we pull the data in 45 minutes. The old technique used volatile storage and required the data for each experiment. Now, we pull the data from Vertica one time and then reload much faster from HDFS. While a significant undertaking, moving to the Spark ecosystem converted an ad hoc and hands-on training process into a fully repeatable pipeline that meets regulatory and business goals for traceability and speed.
Speaker: Josh Johnston
The A-Z of Data: Introduction to MLOps
The A-Z of Data: Introduction to MLOpsDataPhoenix 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/
NextGenML 
NextGenML Moldovan Radu Adrian Revolutionary container based hybrid cloud solution for MLPlatform
Ness' data science platform, NextGenML, puts the entire machine learning process: modelling, execution and deployment in the hands of data science teams.
The entire paradigm approaches collaboration around AI/ML, being implemented with full respect for best practices and commitment to innovation.
Kubernetes (onPrem) + Docker, Azure Kubernetes Cluster (AKS), Nexus, Azure Container Registry(ACR), GlusterFS
Workflow
Argo->Kubeflow
DevOps
Helm, kSonnet, Kustomize,Azure DevOps
Code Management & CI/CD
Git, TeamCity, SonarQube, Jenkins
Security
MS Active Directory, Azure VPN, Dex (K8s) integrated with GitLab
Machine Learning
TensorFlow (model training, boarding, serving), Keras, Seldon
Storage (Azure)
Storage Gen1 & Gen2, Data Lake, File Storage
ETL (Azure)
Databricks, Spark on K8, Data Factory (ADF), HDInsight (Kafka and Spark), Service Bus (ASB)
Lambda functions & VMs, Cache for Redis
Monitoring and Logging
Graphana, Prometeus, GrayLog
Model versioning done right: A ModelDB 2.0 Walkthrough
Model versioning done right: A ModelDB 2.0 WalkthroughManasi Vartak Slides for the Verta.ai ModelDB 2.0 webinar #1. Covers:
(1) Why model versioning
(2) ModelDB 2.0 Architecture
(3) Hands-on exercise with ModelDB
Robust MLOps with Open-Source: ModelDB, Docker, Jenkins, and Prometheus
Robust MLOps with Open-Source: ModelDB, Docker, Jenkins, and PrometheusManasi Vartak These are slides from Manasi Vartak's Strata Talk in March 2020 on Robust MLOps with Open-Source.
* Introduction to talk
* What is MLOps?
* Building an MLOps Pipeline
* Real-world Simulations
* Let’s fix the pipeline
* Wrap-up
Blind spots in big data erez koren @ forter
Blind spots in big data erez koren @ forterIdo Shilon 1) The document discusses challenges with big data analysis including ensuring complete data coverage from all relevant sources like devices, platforms and browser configurations.
2) It also discusses the challenge of effective monitoring to detect issues that could corrupt alerting data, giving examples of how the company Forter addresses these challenges through techniques like API monitoring and machine learning anomaly detection.
3) The key takeaways are to understand all parts of the data pipeline, log errors from both client and server, and flag any incidents affecting input data for data scientists.
Productionalizing Models through CI/CD Design with MLflow
Productionalizing Models through CI/CD Design with MLflowDatabricks Often times model deployment and integration consists of several moving parts that require intricate steps woven together. Automating this pipeline and feedback loop can be incredibly challenging, especially in lieu of varying model development techniques.
MLOps Virtual Event | Building Machine Learning Platforms for the Full Lifecycle
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.
Continuous Delivery of Deep Transformer-Based NLP Models Using MLflow and AWS...
Continuous Delivery of Deep Transformer-Based NLP Models Using MLflow and AWS...Databricks Transformer-based pretrained language models such as BERT, XLNet, Roberta and Albert significantly advance the state-of-the-art of NLP and open doors for solving practical business problems with high performance transfer learning. However, operationalizing these models with production-quality continuous integration/ delivery (CI/CD) end-to-end pipelines that cover the full machine learning life cycle stages of train, test, deploy and serve while managing associated data and code repositories is still a challenging task.
Nasscom ml ops webinar
Nasscom ml ops webinarSameer Mahajan This document outlines a presentation on revamping machine learning pipelines with MLOps. The presentation covers the machine learning lifecycle and challenges with ML productization. It provides examples of end-to-end ML platforms like Uber's Michaelangelo and Google's TFX. The presentation discusses MLOps best practices and methodologies, including build, retrain and release pipelines. It demonstrates MLflow and shows demos of Airflow, Tensorflow model serving, and TFX-based MLOps systems on Google Cloud and Azure.
KFServing, Model Monitoring with Apache Spark and a Feature Store
KFServing, Model Monitoring with Apache Spark and a Feature StoreDatabricks In recent years, MLOps has emerged to bring DevOps processes to the machine learning (ML) development process, aiming at more automation in the execution of repetitive tasks and at smoother interoperability between tools. Among the different stages in the ML lifecycle, model monitoring involves the supervision of model performance over time, involving the combination of techniques in four categories: outlier detection, data drift detection, explainability and adversarial attacks. Most existing model monitoring tools follow a scheduled batch processing approach or analyse model performance using isolated subsets of the inference data. However, for the continuous monitoring of models, stream processing platforms show several advantages, including support for continuous data analytics, scalable processing of large amounts of data and first-class support for window-based aggregations useful for concept drift detection.
In this talk, we present an open-source platform for serving and monitoring models at scale based on Kubeflow’s model serving framework, KFServing, the Hopsworks Online Feature Store for enriching feature vectors with transformer in KFServing, and Spark and Spark Streaming as general purpose frameworks for monitoring models in production.
We also show how Spark Streaming can use the Hopsworks Feature Store to implement continuous data drift detection, where the Feature Store provides statistics on the distribution of feature values in training, and Spark Streaming computes the statistics on live traffic to the model, alerting if the live traffic differs significantly from the training data. We will include a live demonstration of the platform in action.
Productionizing Deep Reinforcement Learning with Spark and MLflow
Productionizing Deep Reinforcement Learning with Spark and MLflowDatabricks Deep Reinforcement Learning has driven exciting AI breakthroughs like self-driving cars, beating the best Go players in the world and even winning at StarCraft. How can businesses harness this power for real world applications?
Managing the Machine Learning Lifecycle with MLflow
Managing the Machine Learning Lifecycle with MLflowDatabricks ML development brings many new complexities beyond the traditional software development lifecycle. MLflow is an open-source project from Databricks aiming to solve some of these challenges such as experiment tracking, reproducibility, model packaging, deployment, and governance, in order to manage and accelerate the lifecycle of your ML projects.
Developing ML-enabled Data Pipelines on Databricks using IDE & CI/CD at Runta...
Developing ML-enabled Data Pipelines on Databricks using IDE & CI/CD at Runta...Databricks Data & ML projects bring many new complexities beyond the traditional software development lifecycle. Unlike software projects, after they were successfully delivered and deployed, they cannot be abandoned but must be continuously monitored if model performance still satisfies all requirements. We can always get new data with new statistical characteristics that can break our pipelines or influence model performance.
Richard Coffey (x18140785) - Research in Computing CA2
Richard Coffey (x18140785) - Research in Computing CA2Richard Coffey The document discusses applying DevOps practices to machine learning algorithms through MLOps. It defines MLOps as combining DevOps practices with machine learning to improve the reliability and deployment of ML models. The document outlines using Microsoft's Azure ML tools to develop a custom ML application and deploy it using MLOps pipelines, then surveying ML professionals on the value of MLOps. It proposes tracking project progression through a Gantt chart.
Machine learning in production with scikit-learn
Machine learning in production with scikit-learnJeff Klukas Presented at PyOhio 2017: https://pyohio.org/schedule/presentation/284/
The Python data ecosystem provides amazing tools to quickly get up and running with machine learning models, but the path to stably serving them in production is not so clear. We'll discuss details of wrapping a minimal REST API around scikit-learn, training and persisting models in batch, and logging decisions, then compare to some other common approaches to productionizing models.
PostgreSQL + Kafka: The Delight of Change Data Capture
PostgreSQL + Kafka: The Delight of Change Data CaptureJeff Klukas PostgreSQL is an open source relational database. Kafka is an open source log-based messaging system. Because both systems are powerful and flexible, they’re devouring whole categories of infrastructure. And they’re even better together.
In this talk, you’ll learn about commit logs and how that fundamental data structure underlies both PostgreSQL and Kafka. We’ll use that basis to understand what Kafka is, what advantages it has over traditional messaging systems, and why it’s perfect for modeling database tables as streams. From there, we’ll introduce the concept of change data capture (CDC) and run a live demo of Bottled Water, an open source CDC pipeline, watching INSERT, UPDATE, and DELETE operations in PostgreSQL stream into Kafka. We’ll wrap up with a discussion of use cases for this pipeline: messaging between systems with transactional guarantees, transmitting database changes to a data warehouse, and stream processing.
Ad
More Related Content
What's hot (20)
Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning Models
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.
MLOps with a Feature Store: Filling the Gap in ML Infrastructure
MLOps with a Feature Store: Filling the Gap in ML InfrastructureData Science Milan A Feature Store enables machine learning (ML) features to be registered, discovered, and used as part of ML pipelines, thus making it easier to transform and validate the training data that is fed into machine learning systems. Feature stores can also enable consistent engineering of features between training and inference, but to do so, they need a common data processing platform. The first Feature Stores, developed at hyperscale AI companies such as Uber, Airbnb, and Facebook, enabled feature engineering using domain specific languages, providing abstractions tailored to the companies’ feature engineering domains. However, a general purpose Feature Store needs a general purpose feature engineering, feature selection, and feature transformation platform.
In this talk, we describe how we built a general purpose, open-source Feature Store for ML around dataframes and Apache Spark. We will demonstrate how data engineers can transform and engineers features from backend databases and data lakes, while data scientists can use PySpark to select and transform features into train/test data in a file format of choice (.tfrecords, .npy, .petastorm, etc) on a file system of choice (S3, HDFS). Finally, we will show how the Feature Store enables end-to-end ML pipelines to be factored into feature engineering and data science stages that each can run at different cadences.
Bio:
Fabio Buso is the head of engineering at Logical Clocks AB, where he leads the Feature Store development. Fabio holds a master's degree in cloud computing and services with a focus on data intensive applications, awarded by a joint program between KTH Stockholm and TU Berlin.
Topics: feature store, MLOps.
MLOps - Build pipelines with Tensor Flow Extended & Kubeflow
MLOps - Build pipelines with Tensor Flow Extended & KubeflowJan Kirenz Machine Learning Operations (MLOps) for Machine Learning Lifecycle Management: Building pipelines with Tensor Flow Extended (TFX) and Kubeflow.
MLOps and Data Quality: Deploying Reliable ML Models in Production
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
MLOps in action
MLOps in actionPieter de Bruin Machine Learning operations brings data science to the world of devops. Data scientists create models on their workstations. MLOps adds automation, validation and monitoring to any environment including machine learning on kubernetes. In this session you hear about latest developments and see it in action.
Moving a Fraud-Fighting Random Forest from scikit-learn to Spark with MLlib, ...
Moving a Fraud-Fighting Random Forest from scikit-learn to Spark with MLlib, ...Databricks This talk describes migrating a large random forest classifier from scikit-learn to Spark's MLlib. We cut training time from 2 days to 2 hours, reduced failed runs, and track experiments better with MLflow. Kount provides certainty in digital interactions like online credit card transactions. One of our scores uses a random forest classifier with 250 trees and 100,000 nodes per tree. We used scikit-learn to train using 60 million samples that each contained over 150 features. The in-memory requirements exceeded 750 GB, took 2 days, and were not robust to disruption in our database or training execution. To migrate workflow to Spark, we built a 6-node cluster with HDFS. This provides 1.35 TB of RAM and 484 cores. Using MLlib and parallelization, the training time for our random forests are now less than 2 hours. Training data stays in our production environment, which used to require a deploy cycle to move locally-developed code onto our training server. The new implementation uses Jupyter notebooks for remote development with server-side execution. MLflow tracks all input parameters, code, and git revision number, while the performance and model itself are retained as experiment artifacts. The new workflow is robust to service disruption. Our training pipeline begins by pulling from a Vertica database. Originally, this single connection took over 8 hours to complete with any problem causing a restart. Using sqoop and multiple connections, we pull the data in 45 minutes. The old technique used volatile storage and required the data for each experiment. Now, we pull the data from Vertica one time and then reload much faster from HDFS. While a significant undertaking, moving to the Spark ecosystem converted an ad hoc and hands-on training process into a fully repeatable pipeline that meets regulatory and business goals for traceability and speed.
Speaker: Josh Johnston
The A-Z of Data: Introduction to MLOps
The A-Z of Data: Introduction to MLOpsDataPhoenix 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/
NextGenML
NextGenML Moldovan Radu Adrian Revolutionary container based hybrid cloud solution for MLPlatform
Ness' data science platform, NextGenML, puts the entire machine learning process: modelling, execution and deployment in the hands of data science teams.
The entire paradigm approaches collaboration around AI/ML, being implemented with full respect for best practices and commitment to innovation.
Kubernetes (onPrem) + Docker, Azure Kubernetes Cluster (AKS), Nexus, Azure Container Registry(ACR), GlusterFS
Workflow
Argo->Kubeflow
DevOps
Helm, kSonnet, Kustomize,Azure DevOps
Code Management & CI/CD
Git, TeamCity, SonarQube, Jenkins
Security
MS Active Directory, Azure VPN, Dex (K8s) integrated with GitLab
Machine Learning
TensorFlow (model training, boarding, serving), Keras, Seldon
Storage (Azure)
Storage Gen1 & Gen2, Data Lake, File Storage
ETL (Azure)
Databricks, Spark on K8, Data Factory (ADF), HDInsight (Kafka and Spark), Service Bus (ASB)
Lambda functions & VMs, Cache for Redis
Monitoring and Logging
Graphana, Prometeus, GrayLog
Model versioning done right: A ModelDB 2.0 Walkthrough
Model versioning done right: A ModelDB 2.0 WalkthroughManasi Vartak Slides for the Verta.ai ModelDB 2.0 webinar #1. Covers:
(1) Why model versioning
(2) ModelDB 2.0 Architecture
(3) Hands-on exercise with ModelDB
Robust MLOps with Open-Source: ModelDB, Docker, Jenkins, and Prometheus
Robust MLOps with Open-Source: ModelDB, Docker, Jenkins, and PrometheusManasi Vartak These are slides from Manasi Vartak's Strata Talk in March 2020 on Robust MLOps with Open-Source.
* Introduction to talk
* What is MLOps?
* Building an MLOps Pipeline
* Real-world Simulations
* Let’s fix the pipeline
* Wrap-up
Blind spots in big data erez koren @ forter
Blind spots in big data erez koren @ forterIdo Shilon 1) The document discusses challenges with big data analysis including ensuring complete data coverage from all relevant sources like devices, platforms and browser configurations.
2) It also discusses the challenge of effective monitoring to detect issues that could corrupt alerting data, giving examples of how the company Forter addresses these challenges through techniques like API monitoring and machine learning anomaly detection.
3) The key takeaways are to understand all parts of the data pipeline, log errors from both client and server, and flag any incidents affecting input data for data scientists.
Productionalizing Models through CI/CD Design with MLflow
Productionalizing Models through CI/CD Design with MLflowDatabricks Often times model deployment and integration consists of several moving parts that require intricate steps woven together. Automating this pipeline and feedback loop can be incredibly challenging, especially in lieu of varying model development techniques.
MLOps Virtual Event | Building Machine Learning Platforms for the Full Lifecycle
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.
Continuous Delivery of Deep Transformer-Based NLP Models Using MLflow and AWS...
Continuous Delivery of Deep Transformer-Based NLP Models Using MLflow and AWS...Databricks Transformer-based pretrained language models such as BERT, XLNet, Roberta and Albert significantly advance the state-of-the-art of NLP and open doors for solving practical business problems with high performance transfer learning. However, operationalizing these models with production-quality continuous integration/ delivery (CI/CD) end-to-end pipelines that cover the full machine learning life cycle stages of train, test, deploy and serve while managing associated data and code repositories is still a challenging task.
Nasscom ml ops webinar
Nasscom ml ops webinarSameer Mahajan This document outlines a presentation on revamping machine learning pipelines with MLOps. The presentation covers the machine learning lifecycle and challenges with ML productization. It provides examples of end-to-end ML platforms like Uber's Michaelangelo and Google's TFX. The presentation discusses MLOps best practices and methodologies, including build, retrain and release pipelines. It demonstrates MLflow and shows demos of Airflow, Tensorflow model serving, and TFX-based MLOps systems on Google Cloud and Azure.
KFServing, Model Monitoring with Apache Spark and a Feature Store
KFServing, Model Monitoring with Apache Spark and a Feature StoreDatabricks In recent years, MLOps has emerged to bring DevOps processes to the machine learning (ML) development process, aiming at more automation in the execution of repetitive tasks and at smoother interoperability between tools. Among the different stages in the ML lifecycle, model monitoring involves the supervision of model performance over time, involving the combination of techniques in four categories: outlier detection, data drift detection, explainability and adversarial attacks. Most existing model monitoring tools follow a scheduled batch processing approach or analyse model performance using isolated subsets of the inference data. However, for the continuous monitoring of models, stream processing platforms show several advantages, including support for continuous data analytics, scalable processing of large amounts of data and first-class support for window-based aggregations useful for concept drift detection.
In this talk, we present an open-source platform for serving and monitoring models at scale based on Kubeflow’s model serving framework, KFServing, the Hopsworks Online Feature Store for enriching feature vectors with transformer in KFServing, and Spark and Spark Streaming as general purpose frameworks for monitoring models in production.
We also show how Spark Streaming can use the Hopsworks Feature Store to implement continuous data drift detection, where the Feature Store provides statistics on the distribution of feature values in training, and Spark Streaming computes the statistics on live traffic to the model, alerting if the live traffic differs significantly from the training data. We will include a live demonstration of the platform in action.
Productionizing Deep Reinforcement Learning with Spark and MLflow
Productionizing Deep Reinforcement Learning with Spark and MLflowDatabricks Deep Reinforcement Learning has driven exciting AI breakthroughs like self-driving cars, beating the best Go players in the world and even winning at StarCraft. How can businesses harness this power for real world applications?
Managing the Machine Learning Lifecycle with MLflow
Managing the Machine Learning Lifecycle with MLflowDatabricks ML development brings many new complexities beyond the traditional software development lifecycle. MLflow is an open-source project from Databricks aiming to solve some of these challenges such as experiment tracking, reproducibility, model packaging, deployment, and governance, in order to manage and accelerate the lifecycle of your ML projects.
Developing ML-enabled Data Pipelines on Databricks using IDE & CI/CD at Runta...
Developing ML-enabled Data Pipelines on Databricks using IDE & CI/CD at Runta...Databricks Data & ML projects bring many new complexities beyond the traditional software development lifecycle. Unlike software projects, after they were successfully delivered and deployed, they cannot be abandoned but must be continuously monitored if model performance still satisfies all requirements. We can always get new data with new statistical characteristics that can break our pipelines or influence model performance.
Richard Coffey (x18140785) - Research in Computing CA2
Richard Coffey (x18140785) - Research in Computing CA2Richard Coffey The document discusses applying DevOps practices to machine learning algorithms through MLOps. It defines MLOps as combining DevOps practices with machine learning to improve the reliability and deployment of ML models. The document outlines using Microsoft's Azure ML tools to develop a custom ML application and deploy it using MLOps pipelines, then surveying ML professionals on the value of MLOps. It proposes tracking project progression through a Gantt chart.
Viewers also liked (13)
Machine learning in production with scikit-learn
Machine learning in production with scikit-learnJeff Klukas Presented at PyOhio 2017: https://pyohio.org/schedule/presentation/284/
The Python data ecosystem provides amazing tools to quickly get up and running with machine learning models, but the path to stably serving them in production is not so clear. We'll discuss details of wrapping a minimal REST API around scikit-learn, training and persisting models in batch, and logging decisions, then compare to some other common approaches to productionizing models.
PostgreSQL + Kafka: The Delight of Change Data Capture
PostgreSQL + Kafka: The Delight of Change Data CaptureJeff Klukas PostgreSQL is an open source relational database. Kafka is an open source log-based messaging system. Because both systems are powerful and flexible, they’re devouring whole categories of infrastructure. And they’re even better together.
In this talk, you’ll learn about commit logs and how that fundamental data structure underlies both PostgreSQL and Kafka. We’ll use that basis to understand what Kafka is, what advantages it has over traditional messaging systems, and why it’s perfect for modeling database tables as streams. From there, we’ll introduce the concept of change data capture (CDC) and run a live demo of Bottled Water, an open source CDC pipeline, watching INSERT, UPDATE, and DELETE operations in PostgreSQL stream into Kafka. We’ll wrap up with a discussion of use cases for this pipeline: messaging between systems with transactional guarantees, transmitting database changes to a data warehouse, and stream processing.
Square's Machine Learning Infrastructure and Applications - Rong Yan
Square's Machine Learning Infrastructure and Applications - Rong YanHakka Labs 1) Square uses machine learning for fraud detection in payments and to power recommendations on its Square Market platform.
2) Random forests and gradient boosted trees are the primary algorithms used for fraud detection, achieving up to a 10-11% improvement over random forests alone.
3) Square has built scalable machine learning infrastructure including parallel environments, data transport systems, and a learning management system to support rapid model development and evaluation.
Python as part of a production machine learning stack by Michael Manapat PyDa...
Python as part of a production machine learning stack by Michael Manapat PyDa...PyData Over the course of three years, we've built Stripe from scratch and scaled it to process billions of dollars of transaction volume a year by making it easy and painless for merchants to get set up and start accepting payments. While the vast majority of transactions facilitated by Stripe are honest, we do need to protect our merchants from rogue individuals and groups seeing to "test" or "cash" stolen credit cards. To combat this sort of activity, Stripe uses Python (together with Scala and Ruby) as part of its production machine learning pipeline to detect and block fraud in real time. In this talk, I'll go through the scikit-based modeling process for a sample data set that is derived from production data to illustrate how we train and validate our models. We'll also walk through how we deploy the models and monitor them in our production environment and how Python has allowed us to do this at scale.
Building A Production-Level Machine Learning Pipeline
Building A Production-Level Machine Learning PipelineRobert Dempsey With so many options to choose from how do you select the right technologies to use for your machine learning pipeline? Do you purchase bare metal and hire a devops team, install Spark on EC2 instances, use EMR and other AWS services, combine Spark and Elasticsearch?! View this talk to get a first-hand experience of building ML pipelines: what options were looked at, how the final solution was selected, the tradeoffs made and the final results.
Production machine learning_infrastructure
Production machine learning_infrastructurejoshwills This document discusses building machine learning infrastructure to scale data science from the lab to production. It describes two types of data scientists - those focused on investigative analytics in the lab and those building production systems in the factory. Moving analytics from the lab to the factory requires a shift from question-driven and ad-hoc work to metric-driven and automated systems. The document outlines steps to begin this transition such as choosing a good problem, logging everything, and hiring more data scientists. It also describes tools and techniques for experimentation in production machine learning.
Machine learning in production
Machine learning in productionTuri, Inc. Tutorial for Machine Learning 101 (an all-day tutorial at Strata + Hadoop World, New York City, 2015)
The course is designed to introduce machine learning via real applications like building a recommender image analysis using deep learning.
In this talk we cover deployment of machine learning models.
Managing and Versioning Machine Learning Models in Python
Managing and Versioning Machine Learning Models in PythonSimon Frid Practical machine learning is becoming messy, and while there are lots of algorithms, there is still a lot of infrastructure needed to manage and organize the models and datasets. Estimators and Django-Estimators are two python packages that can help version data sets and models, for deployment and effective workflow.
Production and Beyond: Deploying and Managing Machine Learning Models
Production and Beyond: Deploying and Managing Machine Learning ModelsTuri, Inc. 1) Deploying machine learning models into production involves evaluating, monitoring, deploying, and managing models over their lifecycle.
2) Evaluation involves continuously tracking metrics on both historical and live data to determine when models need to be updated. Monitoring involves choosing between existing models, such as by using A/B testing or multi-armed bandits.
3) Dato provides tools to simplify each stage of the machine learning lifecycle from batch training to real-time predictions to continuous evaluation and management of models in production.
Machine Learning In Production
Machine Learning In ProductionSamir Bessalah This document discusses challenges in running machine learning applications in production environments. It notes that while Kaggle competitions focus on accuracy, real-world applications require balancing accuracy with interpretability, speed and infrastructure constraints. It also emphasizes that machine learning in production is as much a software and systems problem as a modeling problem. Key aspects that are discussed include flexible and scalable deployment architectures, model versioning, packaging and serving, online evaluation and experiments, and ensuring reproducibility of results.
A full Machine learning pipeline in Scikit-learn vs in scala-Spark: pros and ...
A full Machine learning pipeline in Scikit-learn vs in scala-Spark: pros and ...Jose Quesada (hiring) The machine learning libraries in Apache Spark are an impressive piece of software engineering, and are maturing rapidly. What advantages does Spark.ml offer over scikit-learn? At Data Science Retreat we've taken a real-world dataset and worked through the stages of building a predictive model -- exploration, data cleaning, feature engineering, and model fitting; which would you use in production?
The machine learning libraries in Apache Spark are an impressive piece of software engineering, and are maturing rapidly. What advantages does Spark.ml offer over scikit-learn?
At Data Science Retreat we've taken a real-world dataset and worked through the stages of building a predictive model -- exploration, data cleaning, feature engineering, and model fitting -- in several different frameworks. We'll show what it's like to work with native Spark.ml, and compare it to scikit-learn along several dimensions: ease of use, productivity, feature set, and performance.
In some ways Spark.ml is still rather immature, but it also conveys new superpowers to those who know how to use it.
Machine Learning Pipelines
Machine Learning Pipelinesjeykottalam This document discusses machine learning pipelines and introduces Evan Sparks' presentation on building image classification pipelines. It provides an overview of feature extraction techniques used in computer vision like normalization, patch extraction, convolution, rectification and pooling. These techniques are used to transform images into feature vectors that can be input to linear classifiers. The document encourages building simple, intermediate and advanced image classification pipelines using these techniques to qualitatively and quantitatively compare their effectiveness.
AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017
AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017Carol Smith What is machine learning? Is UX relevant in the age of artificial intelligence (AI)? How can I take advantage of cognitive computing? Get answers to these questions and learn about the implications for your work in this session. Carol will help you understand at a basic level how these systems are built and what is required to get insights from them. Carol will present examples of how machine learning is already being used and explore the ethical challenges inherent in creating AI. You will walk away with an awareness of the weaknesses of AI and the knowledge of how these systems work.
A full Machine learning pipeline in Scikit-learn vs in scala-Spark: pros and ...
A full Machine learning pipeline in Scikit-learn vs in scala-Spark: pros and ...Jose Quesada (hiring)
Ad
Similar to Serverless machine learning operations (20)
DevOps for DataScience
DevOps for DataScienceStepan Pushkarev A talk for SF big analytics meetup. Building, testing, deploying, monitoring and maintaining big data analytics services. http://hydrosphere.io/
mlflow: Accelerating the End-to-End ML lifecycle
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.
Data Agility—A Journey to Advanced Analytics and Machine Learning at Scale
Data Agility—A Journey to Advanced Analytics and Machine Learning at ScaleDatabricks Hari Subramanian presented on Uber's journey to enable data agility and advanced analytics at scale. He discussed Uber's large and growing data platform that processes millions of daily trips and terabytes of data. He then described Uber's Data Science Workbench, which aims to democratize data science by providing self-service access to infrastructure, tools, and data to support various users from data scientists to business analysts. Finally, he presented a case study on COTA, a deep learning model for customer support ticketing that was developed and deployed using Uber's data platform and workflow.
Hydrosphere.io Platform for AI/ML Operations Automation
Hydrosphere.io Platform for AI/ML Operations AutomationRustem Zakiev Simple and robust ML models deployment
Automated versioning
Easy models and versions management
Score the model from your app or microservice via REST, gRPC or Kafka stream API.
A/B and Canary testing on production traffic.
Hot-wing bumpless model replacement in production pipeline
An Insider’s Guide to Maximizing Spark SQL Performance
An Insider’s Guide to Maximizing Spark SQL PerformanceTakuya UESHIN This document provides an overview of optimizing Spark SQL performance. It begins with introducing the speaker and their background with Spark. It then discusses reading query plans, interpreting them to understand optimizations, and tuning plans by pushing down filters, avoiding implicit casts, and other techniques. It emphasizes tracking query execution through the Spark UI to analyze jobs, stages and tasks for bottlenecks. The document aims to help understand how to maximize Spark SQL performance.
Data Summer Conf 2018, “Monitoring AI with AI (RUS)” — Stepan Pushkarev, CTO ...
Data Summer Conf 2018, “Monitoring AI with AI (RUS)” — Stepan Pushkarev, CTO ...Provectus In this demo based talk we discuss a solution, tooling and architecture that allows machine learning engineer to be involved in delivery phase and take ownership over deployment and monitoring of machine learning pipelines. It allows data scientists to safely deploy early results as end-to-end AI applications in a self serve mode without assistance from engineering and operations teams. It shifts experimentation and even training phases from offline datasets to live production and closes a feedback loop between research and production.
Monitoring AI with AI
Monitoring AI with AIStepan Pushkarev Monitoring AI applications with AI
The best performing offline algorithm can lose in production. The most accurate model does not always improve business metrics. Environment misconfiguration or upstream data pipeline inconsistency can silently kill the model performance. Neither prodops, data science or engineering teams are skilled to detect, monitor and debug such types of incidents.
Was it possible for Microsoft to test Tay chatbot in advance and then monitor and adjust it continuously in production to prevent its unexpected behaviour? Real mission critical AI systems require advanced monitoring and testing ecosystem which enables continuous and reliable delivery of machine learning models and data pipelines into production. Common production incidents include:
Data drifts, new data, wrong features
Vulnerability issues, malicious users
Concept drifts
Model Degradation
Biased Training set / training issue
Performance issue
In this demo based talk we discuss a solution, tooling and architecture that allows machine learning engineer to be involved in delivery phase and take ownership over deployment and monitoring of machine learning pipelines.
It allows data scientists to safely deploy early results as end-to-end AI applications in a self serve mode without assistance from engineering and operations teams. It shifts experimentation and even training phases from offline datasets to live production and closes a feedback loop between research and production.
Technical part of the talk will cover the following topics:
Automatic Data Profiling
Anomaly Detection
Clustering of inputs and outputs of the model
A/B Testing
Service Mesh, Envoy Proxy, trafic shadowing
Stateless and stateful models
Monitoring of regression, classification and prediction models
How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....
How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....Databricks Richard Garris presented on ways to productionize machine learning models built with Apache Spark MLlib. He discussed serializing models using MLlib 2.X to save models for production use without reimplementation. This allows data scientists to build models in Python/R and deploy them directly for scoring. He also reviewed model scoring architectures and highlighted Databricks' private beta solution for deploying serialized Spark MLlib models for low latency scoring outside of Spark.
ODSC East 2020 Accelerate ML Lifecycle with Kubernetes and Containerized Da...
ODSC East 2020 Accelerate ML Lifecycle with Kubernetes and Containerized Da...Abhinav Joshi This deck provide an overview of containers and Kubernetes, and how these technologies can help solve the challenges faced by data scientists, ML engineers, and application developers. Next, it showcases the key capabilities required in a containers and kubernetes platform to help data scientists easily use technologies like Jupyter Notebooks, ML frameworks, programming languages to innovate faster. Finally it discusses the available platform options (e.g. KubeFlow, Open Data Hub, etc.), and some examples of how data scientists are accelerating their ML initiatives with containers and kubernetes platform.
Databricks Meetup @ Los Angeles Apache Spark User Group
Databricks Meetup @ Los Angeles Apache Spark User GroupPaco Nathan This document summarizes a presentation on Apache Spark and Spark Streaming. It provides an overview of Spark, describing it as an in-memory cluster computing framework. It then discusses Spark Streaming, explaining that it runs streaming computations as small batch jobs to provide low latency processing. Several use cases for Spark Streaming are presented, including from companies like Stratio, Pearson, Ooyala, and Sharethrough. The presentation concludes with a demonstration of Python Spark Streaming code.
End-to-End Data Pipelines with Apache Spark
End-to-End Data Pipelines with Apache SparkBurak Yavuz This presentation is about building a data product backed by Apache Spark. The source code for the demo can be found at http://brkyvz.github.io/spark-pipeline
Architecting an Open Source AI Platform 2018 edition
Architecting an Open Source AI Platform 2018 editionDavid Talby How to build a scalable AI platform using open source software. The end-to-end architecture covers data integration, interactive queries & visualization, machine learning & deep learning, deploying models to production, and a full 24x7 operations toolset in a high-compliance environment.
Serverless ML Workshop with Hopsworks at PyData Seattle
Serverless ML Workshop with Hopsworks at PyData SeattleJim Dowling 1. The document discusses building a minimal viable prediction service (MVP) to predict air quality using only Python and free serverless services in 90 minutes.
2. It describes creating feature, training, and inference pipelines to build an air quality prediction service using Hopsworks, Modal, and Streamlit/Gradio.
3. The pipelines would extract features from weather and air quality data, train a model, and deploy an inference pipeline to make predictions on new data.
Media_Entertainment_Veriticals
Media_Entertainment_VeriticalsPeyman Mohajerian This document discusses applying Apache Spark to data science challenges in media and entertainment. It introduces Spark as a unifying framework for content personalization using recommendation systems and streaming data, as well as social media analytics using GraphFrames. Specific use cases discussed include content personalization with recommendations, churn analysis, analyzing social networks with GraphFrames, sentiment analysis, and viewership prediction using topic modeling. The document also discusses continuous applications with Spark Streaming, and how Spark ML can be used for machine learning workflows and optimization.
Vinay Kumar [InfluxData] | InfluxDB API Overview | InfluxDays 2022![Vinay Kumar [InfluxData] | InfluxDB API Overview | InfluxDays 2022](https://cdn.slidesharecdn.com/ss_thumbnails/dem08apioverviewvinaykumar-221020211417-9f8004f0-thumbnail.jpg?width=560&fit=bounds)
Vinay Kumar [InfluxData] | InfluxDB API Overview | InfluxDays 2022InfluxData InfluxDB offers a common API across the entire platform. Discover what that means and how you can use the API to build solutions that deliver real value for developers.
MLflow: Infrastructure for a Complete Machine Learning Life Cycle with Mani ...
MLflow: Infrastructure for a Complete Machine Learning Life Cycle with Mani ...Databricks 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 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. In this deep-dive session, through a complete ML model life-cycle example, you will walk away with:
MLflow concepts and abstractions for models, experiments, and projects
How to get started with MLFlow
Understand aspects of MLflow APIs
Using tracking APIs during model training
Using MLflow UI to visually compare and contrast experimental runs with different tuning parameters and evaluate metrics
Package, save, and deploy an MLflow model
Serve it using MLflow REST API
What’s next and how to contribute
Running Apache Spark Jobs Using Kubernetes
Running Apache Spark Jobs Using KubernetesDatabricks Apache Spark has introduced a powerful engine for distributed data processing, providing unmatched capabilities to handle petabytes of data across multiple servers. Its capabilities and performance unseated other technologies in the Hadoop world, but while Spark provides a lot of power, it also comes with a high maintenance cost, which is why we now see innovations to simplify the Spark infrastructure.
PyCon Sweden 2022 - Dowling - Serverless ML with Hopsworks.pdf
PyCon Sweden 2022 - Dowling - Serverless ML with Hopsworks.pdfJim Dowling This document discusses building machine learning systems using serverless services and Python. It introduces the Iris flower classification dataset as a case study. The key steps outlined are to: create accounts on Hopsworks, Modal, and HuggingFace; build and run feature, training and inference pipelines on Modal to classify Iris flowers; and create a predictive user interface using Gradio on HuggingFace to allow users to input Iris flower properties and predict the variety. The document emphasizes that serverless infrastructure allows building operational and analytical ML systems without managing underlying infrastructure.
Grokking TechTalk #29: Building Realtime Metrics Platform at LinkedIn
Grokking TechTalk #29: Building Realtime Metrics Platform at LinkedInGrokking VN Bài techtalk của anh Khải Trần nói về hệ thống data pipeline của LinkedIn được dùng để thu thập hàng chục tỷ messages mỗi ngày, và cách họ chạy hệ thống real-time processing để thống kê lượng dữ liệu này cho mục đính metrics monitoring.
1 số điểm bài talk sẽ chia sẻ:
- Giới thiệu về hệ thống unified metrics platform của LinkedIn
- Cách LinkedIn setup hệ thống BigData pipeline dùng Kafka, HDFS, Apache Calcite và Apache Samza.
- Khái niệm nearline storage, và cách LinkedIn chuyển từ offline architecture sang nearline architecture.
Speaker: Khai Tran, Staff Software Engineer - LinkedIn.
- Hiện đang là staff software engineer ở LinkedIn, phụ trách hệ thống metrics monitoring system. Trước đây từng làm ở Amazon AWS và Oracle.
- PhD, University of Wisconsin-Madison, nghiên cứu về Database Systems.
03_aiops-1.pptx
03_aiops-1.pptxFarazulHoda2 This document discusses several tools that can be used to help manage and track machine learning projects from prototyping to production deployment. It describes Neptune, MLFlow, Kubeflow, and Pachyderm. Neptune and MLFlow are experiment trackers that can log parameters, results and artifacts from machine learning runs. Kubeflow provides notebooks, pipelines and model deployment on Kubernetes. Pachyderm is a versioning system for data pipelines that provides data provenance and tracks changes through containers on Kubernetes.
Ad
Recently uploaded (20)
Societal challenges of AI: biases, multilinguism and sustainability
Societal challenges of AI: biases, multilinguism and sustainabilityJordi Cabot Towards a fairer, inclusive and sustainable AI that works for everybody.
Reviewing the state of the art on these challenges and what we're doing at LIST to test current LLMs and help you select the one that works best for you
Download YouTube By Click 2025 Free Full Activated
Download YouTube By Click 2025 Free Full Activatedsaniamalik72555 Copy & Past Link 👉👉
https://dr-up-community.info/
"YouTube by Click" likely refers to the ByClick Downloader software, a video downloading and conversion tool, specifically designed to download content from YouTube and other video platforms. It allows users to download YouTube videos for offline viewing and to convert them to different formats.
Top 10 Client Portal Software Solutions for 2025.docx
Top 10 Client Portal Software Solutions for 2025.docxPortli Discover the top 10 client portal software solutions for 2025. Streamline communication, ensure security, and enhance client experience.
Why Orangescrum Is a Game Changer for Construction Companies in 2025
Why Orangescrum Is a Game Changer for Construction Companies in 2025Orangescrum Orangescrum revolutionizes construction project management in 2025 with real-time collaboration, resource planning, task tracking, and workflow automation, boosting efficiency, transparency, and on-time project delivery.
Not So Common Memory Leaks in Java Webinar
Not So Common Memory Leaks in Java WebinarTier1 app This SlideShare presentation is from our May webinar, “Not So Common Memory Leaks & How to Fix Them?”, where we explored lesser-known memory leak patterns in Java applications. Unlike typical leaks, subtle issues such as thread local misuse, inner class references, uncached collections, and misbehaving frameworks often go undetected and gradually degrade performance. This deck provides in-depth insights into identifying these hidden leaks using advanced heap analysis and profiling techniques, along with real-world case studies and practical solutions. Ideal for developers and performance engineers aiming to deepen their understanding of Java memory management and improve application stability.
Designing AI-Powered APIs on Azure: Best Practices& Considerations
Designing AI-Powered APIs on Azure: Best Practices& ConsiderationsDinusha Kumarasiri AI is transforming APIs, enabling smarter automation, enhanced decision-making, and seamless integrations. This presentation explores key design principles for AI-infused APIs on Azure, covering performance optimization, security best practices, scalability strategies, and responsible AI governance. Learn how to leverage Azure API Management, machine learning models, and cloud-native architectures to build robust, efficient, and intelligent API solutions
Interactive odoo dashboards for sales, CRM , Inventory, Invoice, Purchase, Pr...
Interactive odoo dashboards for sales, CRM , Inventory, Invoice, Purchase, Pr...AxisTechnolabs Interactive Odoo Dashboard for various business needs can provide users with dynamic, visually appealing dashboards tailored to their specific requirements. such a module that could support multiple dashboards for different aspects of a business
✅Visit And Buy Now : https://bit.ly/3VojWza
✅This Interactive Odoo dashboard module allow user to create their own odoo interactive dashboards for various purpose.
App download now :
Odoo 18 : https://bit.ly/3VojWza
Odoo 17 : https://bit.ly/4h9Z47G
Odoo 16 : https://bit.ly/3FJTEA4
Odoo 15 : https://bit.ly/3W7tsEB
Odoo 14 : https://bit.ly/3BqZDHg
Odoo 13 : https://bit.ly/3uNMF2t
Try Our website appointment booking odoo app : https://bit.ly/3SvNvgU
👉Want a Demo ?📧 [email protected]
➡️Contact us for Odoo ERP Set up : 091066 49361
👉Explore more apps: https://bit.ly/3oFIOCF
👉Want to know more : 🌐 https://www.axistechnolabs.com/
#odoo #odoo18 #odoo17 #odoo16 #odoo15 #odooapps #dashboards #dashboardsoftware #odooerp #odooimplementation #odoodashboardapp #bestodoodashboard #dashboardapp #odoodashboard #dashboardmodule #interactivedashboard #bestdashboard #dashboard #odootag #odooservices #odoonewfeatures #newappfeatures #odoodashboardapp #dynamicdashboard #odooapp #odooappstore #TopOdooApps #odooapp #odooexperience #odoodevelopment #businessdashboard #allinonedashboard #odooproducts
Requirements in Engineering AI- Enabled Systems: Open Problems and Safe AI Sy...
Requirements in Engineering AI- Enabled Systems: Open Problems and Safe AI Sy...Lionel Briand Keynote at RAISE workshop, ICSE 2025
Salesforce Data Cloud- Hyperscale data platform, built for Salesforce.
Salesforce Data Cloud- Hyperscale data platform, built for Salesforce.Dele Amefo Salesforce Data Cloud- Hyperscale data platform, built for Salesforce.
Automation Techniques in RPA - UiPath Certificate
Automation Techniques in RPA - UiPath CertificateVICTOR MAESTRE RAMIREZ Automation Techniques in RPA - UiPath Certificate
Adobe Illustrator Crack FREE Download 2025 Latest Version
Adobe Illustrator Crack FREE Download 2025 Latest Versionkashifyounis067 🌍📱👉COPY LINK & PASTE ON GOOGLE http://drfiles.net/ 👈🌍
Adobe Illustrator is a powerful, professional-grade vector graphics software used for creating a wide range of designs, including logos, icons, illustrations, and more. Unlike raster graphics (like photos), which are made of pixels, vector graphics in Illustrator are defined by mathematical equations, allowing them to be scaled up or down infinitely without losing quality.
Here's a more detailed explanation:
Key Features and Capabilities:
Vector-Based Design:
Illustrator's foundation is its use of vector graphics, meaning designs are created using paths, lines, shapes, and curves defined mathematically.
Scalability:
This vector-based approach allows for designs to be resized without any loss of resolution or quality, making it suitable for various print and digital applications.
Design Creation:
Illustrator is used for a wide variety of design purposes, including:
Logos and Brand Identity: Creating logos, icons, and other brand assets.
Illustrations: Designing detailed illustrations for books, magazines, web pages, and more.
Marketing Materials: Creating posters, flyers, banners, and other marketing visuals.
Web Design: Designing web graphics, including icons, buttons, and layouts.
Text Handling:
Illustrator offers sophisticated typography tools for manipulating and designing text within your graphics.
Brushes and Effects:
It provides a range of brushes and effects for adding artistic touches and visual styles to your designs.
Integration with Other Adobe Software:
Illustrator integrates seamlessly with other Adobe Creative Cloud apps like Photoshop, InDesign, and Dreamweaver, facilitating a smooth workflow.
Why Use Illustrator?
Professional-Grade Features:
Illustrator offers a comprehensive set of tools and features for professional design work.
Versatility:
It can be used for a wide range of design tasks and applications, making it a versatile tool for designers.
Industry Standard:
Illustrator is a widely used and recognized software in the graphic design industry.
Creative Freedom:
It empowers designers to create detailed, high-quality graphics with a high degree of control and precision.
F-Secure Freedome VPN 2025 Crack Plus Activation New Version
F-Secure Freedome VPN 2025 Crack Plus Activation New Versionsaimabibi60507 Copy & Past Link 👉👉
https://dr-up-community.info/
F-Secure Freedome VPN is a virtual private network service developed by F-Secure, a Finnish cybersecurity company. It offers features such as Wi-Fi protection, IP address masking, browsing protection, and a kill switch to enhance online privacy and security .
Solidworks Crack 2025 latest new + license code
Solidworks Crack 2025 latest new + license codeaneelaramzan63 Copy & Paste On Google >>> https://dr-up-community.info/
The two main methods for installing standalone licenses of SOLIDWORKS are clean installation and parallel installation (the process is different ...
Disable your internet connection to prevent the software from performing online checks during installation
Meet the Agents: How AI Is Learning to Think, Plan, and Collaborate
Meet the Agents: How AI Is Learning to Think, Plan, and CollaborateMaxim Salnikov Imagine if apps could think, plan, and team up like humans. Welcome to the world of AI agents and agentic user interfaces (UI)! In this session, we'll explore how AI agents make decisions, collaborate with each other, and create more natural and powerful experiences for users.
Adobe Marketo Engage Champion Deep Dive - SFDC CRM Synch V2 & Usage Dashboards
Adobe Marketo Engage Champion Deep Dive - SFDC CRM Synch V2 & Usage DashboardsBradBedford3 Join Ajay Sarpal and Miray Vu to learn about key Marketo Engage enhancements. Discover improved in-app Salesforce CRM connector statistics for easy monitoring of sync health and throughput. Explore new Salesforce CRM Synch Dashboards providing up-to-date insights into weekly activity usage, thresholds, and limits with drill-down capabilities. Learn about proactive notifications for both Salesforce CRM sync and product usage overages. Get an update on improved Salesforce CRM synch scale and reliability coming in Q2 2025.
Key Takeaways:
Improved Salesforce CRM User Experience: Learn how self-service visibility enhances satisfaction.
Utilize Salesforce CRM Synch Dashboards: Explore real-time weekly activity data.
Monitor Performance Against Limits: See threshold limits for each product level.
Get Usage Over-Limit Alerts: Receive notifications for exceeding thresholds.
Learn About Improved Salesforce CRM Scale: Understand upcoming cloud-based incremental sync.
Exceptional Behaviors: How Frequently Are They Tested? (AST 2025)
Exceptional Behaviors: How Frequently Are They Tested? (AST 2025)Andre Hora Exceptions allow developers to handle error cases expected to occur infrequently. Ideally, good test suites should test both normal and exceptional behaviors to catch more bugs and avoid regressions. While current research analyzes exceptions that propagate to tests, it does not explore other exceptions that do not reach the tests. In this paper, we provide an empirical study to explore how frequently exceptional behaviors are tested in real-world systems. We consider both exceptions that propagate to tests and the ones that do not reach the tests. For this purpose, we run an instrumented version of test suites, monitor their execution, and collect information about the exceptions raised at runtime. We analyze the test suites of 25 Python systems, covering 5,372 executed methods, 17.9M calls, and 1.4M raised exceptions. We find that 21.4% of the executed methods do raise exceptions at runtime. In methods that raise exceptions, on the median, 1 in 10 calls exercise exceptional behaviors. Close to 80% of the methods that raise exceptions do so infrequently, but about 20% raise exceptions more frequently. Finally, we provide implications for researchers and practitioners. We suggest developing novel tools to support exercising exceptional behaviors and refactoring expensive try/except blocks. We also call attention to the fact that exception-raising behaviors are not necessarily “abnormal” or rare.
Adobe Lightroom Classic Crack FREE Latest link 2025
Adobe Lightroom Classic Crack FREE Latest link 2025kashifyounis067 🌍📱👉COPY LINK & PASTE ON GOOGLE http://drfiles.net/ 👈🌍
Adobe Lightroom Classic is a desktop-based software application for editing and managing digital photos. It focuses on providing users with a powerful and comprehensive set of tools for organizing, editing, and processing their images on their computer. Unlike the newer Lightroom, which is cloud-based, Lightroom Classic stores photos locally on your computer and offers a more traditional workflow for professional photographers.
Here's a more detailed breakdown:
Key Features and Functions:
Organization:
Lightroom Classic provides robust tools for organizing your photos, including creating collections, using keywords, flags, and color labels.
Editing:
It offers a wide range of editing tools for making adjustments to color, tone, and more.
Processing:
Lightroom Classic can process RAW files, allowing for significant adjustments and fine-tuning of images.
Desktop-Focused:
The application is designed to be used on a computer, with the original photos stored locally on the hard drive.
Non-Destructive Editing:
Edits are applied to the original photos in a non-destructive way, meaning the original files remain untouched.
Key Differences from Lightroom (Cloud-Based):
Storage Location:
Lightroom Classic stores photos locally on your computer, while Lightroom stores them in the cloud.
Workflow:
Lightroom Classic is designed for a desktop workflow, while Lightroom is designed for a cloud-based workflow.
Connectivity:
Lightroom Classic can be used offline, while Lightroom requires an internet connection to sync and access photos.
Organization:
Lightroom Classic offers more advanced organization features like Collections and Keywords.
Who is it for?
Professional Photographers:
PCMag notes that Lightroom Classic is a popular choice among professional photographers who need the flexibility and control of a desktop-based application.
Users with Large Collections:
Those with extensive photo collections may prefer Lightroom Classic's local storage and robust organization features.
Users who prefer a traditional workflow:
Users who prefer a more traditional desktop workflow, with their original photos stored on their computer, will find Lightroom Classic a good fit.
How Valletta helped healthcare SaaS to transform QA and compliance to grow wi...
How Valletta helped healthcare SaaS to transform QA and compliance to grow wi...Egor Kaleynik This case study explores how we partnered with a mid-sized U.S. healthcare SaaS provider to help them scale from a successful pilot phase to supporting over 10,000 users—while meeting strict HIPAA compliance requirements.
Faced with slow, manual testing cycles, frequent regression bugs, and looming audit risks, their growth was at risk. Their existing QA processes couldn’t keep up with the complexity of real-time biometric data handling, and earlier automation attempts had failed due to unreliable tools and fragmented workflows.
We stepped in to deliver a full QA and DevOps transformation. Our team replaced their fragile legacy tests with Testim’s self-healing automation, integrated Postman and OWASP ZAP into Jenkins pipelines for continuous API and security validation, and leveraged AWS Device Farm for real-device, region-specific compliance testing. Custom deployment scripts gave them control over rollouts without relying on heavy CI/CD infrastructure.
The result? Test cycle times were reduced from 3 days to just 8 hours, regression bugs dropped by 40%, and they passed their first HIPAA audit without issue—unlocking faster contract signings and enabling them to expand confidently. More than just a technical upgrade, this project embedded compliance into every phase of development, proving that SaaS providers in regulated industries can scale fast and stay secure.
WinRAR Crack for Windows (100% Working 2025)
WinRAR Crack for Windows (100% Working 2025)sh607827 copy and past on google ➤ ➤➤ https://hdlicense.org/ddl/
WinRAR Crack Free Download is a powerful archive manager that provides full support for RAR and ZIP archives and decompresses CAB, ARJ, LZH, TAR, GZ, ACE, UUE, .
Download Wondershare Filmora Crack [2025] With Latest![Download Wondershare Filmora Crack [2025] With Latest](https://cdn.slidesharecdn.com/ss_thumbnails/neo4j-howkgsareshapingthefutureofgenerativeaiatawssummitlondonapril2024-240426125209-2d9db05d-250419-250428115407-a04afffa-thumbnail.jpg?width=560&fit=bounds)
Download Wondershare Filmora Crack [2025] With Latesttahirabibi60507 Copy & Past Link 👉👉
http://drfiles.net/
Wondershare Filmora is a video editing software and app designed for both beginners and experienced users. It's known for its user-friendly interface, drag-and-drop functionality, and a wide range of tools and features for creating and editing videos. Filmora is available on Windows, macOS, iOS (iPhone/iPad), and Android platforms.
Serverless machine learning operations
- 1. Serverless Machine Learning Operations by Stepan Pushkarev CTO of Hydrosphere.io
- 2. Mission: Accelerate Machine Learning to Production Opensource Products: - Mist: Spark Compute as a Service - ML Lambda: ML Function as a Service - Sonar: Data and ML Monitoring Business Model: Subscription services and hands-on consulting About
- 3. Ops folks here? Machine Learning nerds here? VP/Managers/Strategy?
- 4. Development Operations are well studied
- 5. Machine Learning operations are ad hoc ● Research phase -> productization phase ● Scripts driven ./bin/spark-submit python train.py ● Raw SQL / HiveQL / SQL on Hadoop ● Automated with Cron and/or Workflow Managers ● Hosted Notebooks culture
- 6. ML Project Time to Market
- 7. ML Project Time to Market
- 8. ML Project Time to Market
- 9. - Go to production strategy from the Day 1 - Training: Serverless Spark Compute - Serving/inferencing: Serverless ML Lambdas Agenda
- 10. Why does business hire data scientists?
- 11. Why do companies hire data scientists? To make products smarter.
- 12. What is a deliverable of data scientist? Academic paper ML Model R/Python script Jupiter Notebook BI Dashboard
- 13. How to move this to prod? Academic paper? ML Model? R/Python script? Jupiter Notebook? BI Dashboard?
- 14. Tragedy 1: Engineer to re-implement R/Python script
- 15. Tragedy 2: Notebook/scripts deployments
- 16. Tragedy 2: Run notebook/script as it is using cron
- 17. © Daniel Tunkelang - Where should you put your data scientists? - www.slideshare.net/dtunkelang/where-should-you-put-your-data-scientists Step 1 (management): Integrate data scientists into cross-functional teams
- 18. Step 2: Build/Deploy functions, not notebooks
- 19. Step 3: Monitor ML in production with other ML ● Data pipeline statistics ● Anomaly detection ● Pattern recognition ● Keep Data Scientist in the loop ● Treat data errors as Software bugs
- 22. From Vanilla Spark to serverless training and data processing ./bin/spark-submit - Spark Sessions Pool - Functions Registry - Multi-tenancy - REST API Framework - Data API Framework - Infrastructure Integration (EMR, Hortonworks, etc)
- 24. UX: Deploy Spark functions and trigger it from apps
- 25. Mist - Serverless proxy for Spark DEMO
- 26. Machine Learning: training + serving
- 27. pipeline Training (Estimation) pipeline trainpreprocess preprocess
- 29. cluster data model data scientist web app docker API libs model Local Spark ML Serving Library: https://github.com/Hydrospheredata/spark-ml-serving
- 30. Model Artifact
- 31. Models - Runtimes - Formats Zoo
- 32. API & Logistics - HTTP/1.1, HTTP/2, gRPC - Kafka, Flink, Kinesis - Protobuf, Avro - Service Discovery - Pipelining - Tracing - Monitoring - Autoscaling - Versioning - A/B, Canary - Testing - CPU, GPU
- 35. UX: Train anywhere and deploy as a Function
- 36. UX: Models and Applications Applications provide public endpoints for the models and compositions of the models.
- 37. UX: Streaming Applications + Batching
- 38. UX: Pipelines, Assembles and BestSLA Applications
- 39. ML Function as a Service Demo!!!
- 40. Thank you Looking for - Feedback - Advisors, mentors & partners - Pilots and early adopters Stay in touch - @hydrospheredata - https://github.com/Hydrospheredata - http://hydrosphere.io/ - [email protected]