Skip to content

fuzzylabs/awesome-open-mlops

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 

History

44 Commits
ย 
ย 
ย 
ย 
ย 
ย 

Repository files navigation

Open Source MLOps

This is the Fuzzy Labs guide to the universe of free and open source MLOps tools.

Contents

What is MLOps anyway?

MLOps (machine learning operations) is a discipline that helps people to train, deploy and run machine learning models successfully in production environments. Because this is a new and rapidly-evolving field, there are a lot of tools out there, and new ones appear all the time. If we've missed any, then please do raise a pull request! (but please ensure it meets the definitions of open source below).

What counts as open source?

In deciding what to include in this list we apply 3 criteria.

๐Ÿ‘ฎโ€ Fits the definition

Rather than invent our own definition, we defer to the Open Source Initiative's definition of open source. The OSI have laid out a set of clear and unambiguous requirements for something to be considered free and open source.

โš– Open source license

So it follows that any OSI-approved license is acceptable, with examples including the various Apache, GPL, and BSD licenses. But this also means there are licenses that we don't include, such as the increasingly-popular Server Side Public License; to understand why, see the OSI's writeup on the subject.

๐Ÿ”‹ Batteries included

We only list tools that are fully-functional and able to stand on their own. Some vendors offer limited open source versions of their tools, but ultimately those vendors want to guide you towards a SaaS offering. Sometimes there's a genuinely open-source client, but a proprietary server, meaning that the tool is not truly stand-alone. Everything in this collection comes complete with "batteries included".

Data annotation

Data annotation (or labelling) tools help us create labelled training data for use in supervised learning. This is one of the most labour-intensive processes in machine learning, and one where good tooling can save a great deal of effort. Because data comes in a huge variety of types and formats, there are also a huge variety of different tools which specialise in certain types of data.

Supported Data Types Name License Description
Multiple Label Studio Apache 2.0 A Swiss army knife of data labelling.
Text doccano MIT A popular tool for text annotation to build datasets for NLP tasks such as sentiment analysis and named entity recognition. Easy to setup and support most languages.
Image labelme GPL-3 A graphical image annotation tool for object detection, segmentation, and classification. Supports polygon, circle, line, and point annotations.
Video Computer Vision Annotation Tool (CVAT) MIT One of the most popular video and image annotation tools used by professional data annotation teams. It can be used online so you can start labelling your data without any installations.
Audio Praat GPL-3 A speech analysis tool used for doing phonetics by computer. Some of its most prominent features are speech labelling and speech analysis.

For a more in-depth list of data annotation tools, check out Open Source Data Annotation & Labeling Tools by ZenML.

Data validation

The golden rule of machine learning is that a model is only as good as the data used to train it. Data validation is the process of verifying that data is accurate and consistent. As errors in data can severely impact the performance of your model, it is crucial to spot data errors early.

Sometimes your model might achieve good accuracy during training , but performs poorly on real-world data. This could be an indication that there are errors in your data which were not spotted during validation. Even a tiny mismatch in the format of the data provided to your machine learning model can significantly impact its performance.

Name License Description
Great Expectations Apache 2.0 An easy-to-use tool with over 280 assertions for validating, documenting, and profiling your data to maintain quality. Great Expectations works with a variety of data sources, including SQL databases, and various file formats
Data Validation Tool (DVT) Apache 2.0 A Python tool that allows you to build custom query validation, schema validation and much more. It also supports a wide range of databases such as MySql, Snowflake and BigQuery, just to name a few.
data-diff MIT An open-source command-line tool to compare rows across two different databases. An ideal tool to use for data migration.
Cerberus ISC A powerful yet simple data validation library for Python. Lightweight data validation functionality out-of-the-box and is designed to be easily extensible, allowing for custom validation.
deequ Apache 2.0 A library built on top of Apache Spark for defining "unit tests for data" in large datasets.

Data version control

Just like code, data grows and evolves over time. Data versioning tools help you to keep track of these changes.

You might wonder why you can't just store data in Git (or equivalent). There are a few reasons this doesn't work, but the main one is size: Git is designed for small text files, and typical datasets used in machine learning are just too big. Some tools, like DVC, store the data externally, but also integrate with Git so that data versions can be linked to code versions.

Name License Description
DVC Apache 2.0 One of the most popular general-purpose data versioning tools.
Delta Lake Apache 2.0 Data versioning for data warehouses.
LakeFS Apache 2.0 Transform your object storage into a Git-like repository.
Git LFS MIT Not specialised in machine learning use-cases, but another popular way to version datasets.

Experiment tracking

Machine learning involves a lot of experimentation. We end up training a lot of models, most of which are never intended to go into production, but represent progressive steps towards having something production-worthy. Experiment tracking tools are there to help us keep track of each experiment. What exactly do we need to track? typically this includes the code version, data version, input parameters, training performance metrics, as well as the final model assets.

Name License Description
Sacred MIT
CometLLM MIT Track, visualize, and evaluate your LLM prompts in one convenient UI.
Tensorboard Apache 2.0
Guild.AI Apache 2.0
MLFlow Apache 2.0
Kedro Apache 2.0 A Python framework for creating reproducible, maintainable and modular data science code.

Model training

Name License Description
MLFlow Apache 2.0
Kubeflow Apache 2.0 The machine learning toolkit for kubernetes
Metaflow Apache 2.0
envd Apache 2.0 The machine learning development environment for data science and AI/ML engineering teams.

Model validation

Data validation has been around for some time, but what about validating ML? Although ML models are very dependant on the data that you feed them, models must also be validated before being released into the real world.

Name License Description
Trubrics Apache 2.0 Validate machine learning with data science and domain expert feedback.

Model registries

Model registries are used to track the lifecycle of trained machine learning models.

Name License Description
MLFlow Apache 2.0 The ML Flow model registry allows you to store, annotate, discover, and manage models in a central repository.
modelstore Apache 2.0 A Python library for versioning, storing, and tracking ML model artefacts across several different types of storage.

Workflows

Name License Description
CML Apache 2.0 Continuous Machine Learning (CML) is CI/CD for Machine Learning Projects

Feature stores

Name License Description
Feast Apache 2.0 A complete open source feature store.
Hopsworks AGPL-3.0 A feature store, feature engineering, and more.
Feathr Apache 2.0 Enterprise-grade, high performance feature store

Feature Engineering

Name License Description
Hamilton BSD-3 CC A way to organize, curate, and scale your feature transform code.

Model deployment and serving

Model serving is the process of taking a trained model and presenting it behind a REST API, and this enables other software components to interact with a model. To make deployment of these model servers as simple as possible, it's commonplace to run them inside Docker containers and deploy them to a container orchestration system such as Kubernetes.

Name License Description
Seldon Core Apache 2.0 Turn your models into microservices to run on Kubernetes
BentoML Apache 2.0
Bodywork AGPL-3.0
KServe Apache 2.0
MLEM Apache 2.0 ๐Ÿถ Version and deploy your ML models following GitOps principles

Model monitoring

Monitoring means making sure that each deployed model is both functioning, and producing sensible results. We don't just want to check for errors in the traditional sense, but also for things like drift and signs of bias in the predictions and decisions that come from a model.

Name License Description
Evidently Apache 2.0
Boxkite ML Apache 2.0
Alibi Detect (by Seldon) Apache 2.0
whylogs Apache 2.0 whylogs allows users to generate data profiles, statistical summaries of data, that can be used as logs for an AI observability platform.
RecList MIT RecList is an open source library providing behavioral testing for recommender systems, to be used either offline or online for a better understanding of model "true" generalization abilities. RecList comes with a WWW paper, and several tutorials.

Data Catalogues

Tools for data discovery and data access control. Can be very important to workout and manage what data can be trained on or tested on, and how these rules vary from client to client. Great summary at https://atlan.com/open-source-data-governance-tools/. This space is pretty young with no real convergence on an expected set of features. They vary in focus from data discovery to data accesss management.

Name License Description
Magda Apache 2.0
Amundsen Apache 2.0
Apache Atlas Apache 2.0

Full stacks

Name License Description
Open MLOps MIT
You Don't Need a Bigger Boat MIT An end-to-end open project, from real-world data to an endpoint serving Transformer-based predictions, supported by videos, scholarly references and a TDS blog series on "Reasonable Scale" ML
Recommendations at "Reasonable Scale" MIT An end-to-end recommendation project leveraging NVIDIA Merlin library to train a deep learning model on top of the Modern Data Stack; the related NVIDIA keynote talk is also publicly available.
ZenML Apache 2.0 An extensible, open-source MLOps framework to create production-ready machine learning pipelines.

Governance

Name License Description
Bailo Apache 2.0 Managing the lifecycle of machine learning to support scalability, impact, collaboration, compliance and sharing.

More resources

Here are some more resources for MLOps, both open-source and proprietary