Just a short snippet on Data Catalogues. Almost started trying to build out a custom one because I didn't know they existed. Therefore feel it's important to make some noise about them.

Open to feedback, can add more detail once I know if you guys think they fit the MLOps category. Only didn't put them under the Data Governance bin because those that focus on Data Discovery don't have so much to do with it. Happy to shuffle the content over though.

This PR adds Feathr

About Feathr:

Feathr is an open source enterprise-grade, high performance feature store, hosted in incubation in the LF AI & Data Foundation.

@archena Please let me know if any changes are needed for the description.

Hello, we have recently launched Trubrics and I'd love to add it to your great list!

Trubrics helps Data Scientists validate their models, by providing them with a framework to write validations. Validations can be built purely with data science knowledge, or with feedback collected from business users on ML models.

Thanks :grinning:

Hello! We have a new MLOPs tool we'd love to add to the awesome-open-mlops!

You can find the repo here: https://github.com/iterative/mlem Read the blog post: https://iterative.ai/blog/MLEM-release Watch the video: https://youtu.be/7h0fiZNwCnA

Let me know if you need anything else or would like to collaborate in some way! Best regards...

Added few hopefully interesting repos, all OS focused. I used monitoring for RecList even if (interestingly) is probably in his own category (with CheckList for example): "model testing", or something like that!

Picked out the most interesting tools from this repo built by ZenML. Added a new section to our repo for data annotation. Include a link to ZenML’s repo as well.

I'd like to share envd with the community!

envd is a machine learning development environment for data science and AI/ML engineering teams.

🐍 No Docker, only Python - Focus on writing Python code, we will take care of Docker and development environment setup.

🖨️ Built-in Jupyter/VSCode - First-class support for Jupyter and VSCode remote extension.

⏱️ Save time - Better cache management to save your time, keep the focus on the model, instead of dependencies.

☁️ Local & cloud - envd integrates seamlessly with Docker so that you can easily share, version, and publish envd environments with Docker Hub or any other OCI image registries.

🔁 Repeatable builds & reproducible results - You can reproduce the same dev environment on your laptop, public cloud VMs, or Docker containers, without any change in setup.

Signed-off-by: Ce Gao [email protected]

👋 Hello!

This PR adds a new section for model registries and I've added both ML Flow and modelstore to it. Disclosure: I'm the author of the latter.

When reading through the other sections, I saw that this might overlap slightly with experiment tracking, which is described as:

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.

Happy to change this around as you see fit.

Thank you for considering this contribution 🙏

Hamilton was created to help wrangle a feature engineering code base. It forces decoupling of feature transform logic from materialization, and results in code that is always unit testable, reusable, and documentation friendly.

I didn't see an appropriate section to add it, so I created a feature engineering section - putting it in feature stores wouldn't be the right place.

I've synced the forks. I have no idea why my previous data catalogue addition is listed again as a new change? Will correct that if anyone can tell me where I've gone wrong.

I've hit my limit of GitHub lists so will be making greater use of this repo. Adding in image analysis tools (fiftyone - the dominant player, and dendromap - looks very powerful).

Will add links, and try to get round to adding descriptions for data catalogues soon

Deepchecks is open source tool for testing and validating machine learning models and data. The 3 checks supported in different phases of ML pipeline are

• Data Integrity (between ingestion and preprocessing step)
• Train-Test Validation (Distribution and Methodology Checks) (between preprocessing and training step)
• Model Performance Evaluation (evaluation step)