Managing Model Registries

Managing model registries is a crucial aspect of Machine Learning Operations (MLOps) that facilitates the organization and tracking of machine learning models throughout their lifecycle. A model registry serves as a centralized repository where models can be stored, versioned, and documented, ensuring that teams can efficiently manage their model artifacts. This process includes registering new models, maintaining metadata about training data and performance metrics, and promoting or demoting model versions based on their efficacy. By implementing effective model registry management, organizations can streamline collaboration, enhance reproducibility, and improve the overall deployment process of machine learning models.

Register and Deploy Models with SageMaker Model Registry

Towards Data Science

An Introduction To SageMaker Model Registry Continue reading on Towards Data Science

ML model registry — the “interface” that binds model experiments and model deployment

Towards Data Science

ML model registry — the “interface” that binds model experiments and model deployment. MLOps in Practice — A deep- dive into ML model registries, model versioning and model lifecycle management..

MLOps in a Nutshell: Model Registry, ML Metadata Store and Model Pipeline

Python in Plain English

The following is a collection of three shorter-form content pieces I’ve published on LinkedIn. They present three core MLOps (Machine Learning Operations) concepts in a concise manner: * Model Registr...

Model Monitoring Dashboards made easy (1/3).

Towards AI

I know the pain of model monitoring and retraining models is not very pleasant for any machine learning engineer, more so if there isn’t any easy way to keep track of all the models that are deployed…...

Advent of 2022, Day 14 – Registering the models

R-bloggers

In the series of Azure Machine Learning posts: Important asset is the “Models” in navigation bar. This feature allows you to work with different model types -__ custom, MLflow, and Triton. What you do...

Automate your Model Documentation using H2O AutoDoc

Towards Data Science

The Federal Reserve’s 2011 guidelines state that model risk assessment and management would be ineffective without adequate documentation. A similar requirement is put forward today by many…

How To Effectively Manage Deployed Models

Towards Data Science

Most models never make it to production. We previously looked at deploying Tensorflow models using Tensorflow Serving. Once that process is completed, we may think that our work is all done. In…

Working with Docker Registry: Managing Your Image Repositories

Python in Plain English

Docker Registry serves as a central repository for storing and distributing Docker images. Effectively managing your image repositories on Docker Registry is essential for organizing, sharing, and sec...

Models, MLFlow, and Microsoft Fabric

Towards Data Science

Fabric Madness part 5 Image by author and ChatGPT. “Design an illustration, with imagery representing multiple machine learning models, focusing on basketball data” prompt. ChatGPT, 4, OpenAI, 25th A...

MLOps: Model Monitoring 101

Towards Data Science

Model monitoring using a model metric stack is essential to put a feedback loop from a deployed ML model back to the model building stage so that ML models can constantly improve themselves under diff...

A Catalog of Models

Towards Data Science

There are many types of models--deterministic, empirical, probabilistic. You need to understand which type is best for your application.

MLOps With AWS Platform Part 2

Towards AI

AWS MLOps: Comprehensive Workflow Stages Stage 1: Data Management Stage 2: Model Development Stage 3: Model Registry Stage 4: CI/CD Pipeline Stage 5: Deployment Options Stage 6: Monitoring Stage 7: Se...