giskard VS metaflow

Unchecked biases in LLMs can inadvertently perpetuate harmful stereotypes or produce misleading information, which in turn can produce severe consequences. In this article, we'll demonstrate how to evaluate your LLMs using an open source model testing framework, Giskard. 🤓

Hi! I’ve been working on this automatic scanner for ML models to detect issues like underperforming data slices, overconfidence in predictions, robustness problems, and others. It supports all main Python ML frameworks (sklearn, torch, xgboost, …) and integrates with the quality assurance solution we are building at Giskard AI (https://giskard.ai) to systematically test models before putting them in production.

It is still a beta and I would love to hear your feedback if you have the time to try it out.

We have quite a few tutorials in the docs with ready-made colab notebooks to make it easy to get started.

If you are interested in the code:

https://github.com/Giskard-AI/giskard/tree/main/python-clien...

Sure! Benjamini-Hochberg is a very good recommendation, much simpler than the alpha investing procedures I mentioned which makes it easily to implement in our case. I will give it a try, if there’s an easy way to set this up it could be included in some of the next releases. I’ll let you know. FYI, I added this to our issue tracker.

This is super interesting! Thanks for sharing. We're also working on this research field from an open-source angle (https://github.com/Giskard-AI/giskard)

Code repository: https://github.com/Giskard-AI/giskard

I would recommend the following: - https://www.mage.ai/ - https://dagster.io/ - https://www.prefect.io/ - https://metaflow.org/ - https://zenml.io/home

1) I've been looking into [Metaflow](https://metaflow.org/), which connects nicely to AWS, does a lot of heavy lifting for you, including scheduling.

even for people who don't have an ML background there's now a lot of very fully-featured model deployment environments that allow self-hosting (kubeflow has a good self-hosting option, as do mlflow and metaflow), handle most of the complicated stuff involved in just deploying an individual model, and work pretty well off the shelf.

They had to figure out video compression that worked at the volume that they wanted to deliver. They had to build and maintain their own CDN to be able to have a always available and consistent viewing experience. Don’t even get me started on the resiliency tools like hystrix that they were kind enough to open source. I mean, they have their own fucking data science framework and they’re looking into using neural networks to downscale video.. Sound familiar? That’s cause that’s practically the same thing as Nvidia’s DLSS (which upscales instead of downscales).

Check out Metaflow by Netflix

Github Actions, Metaflow and AWS SageMaker are awesome technologies by themselves however they are seldom used together in the same sentence, even less so in the same Machine Learning project.

It's undeniable that leadership is instrumental in any company and project success, however I was intrigued with one of their ML tool choices that helped them reach their goal. I was so curious about this choice that I just had to learn more about it, so in this article will be talking about a sound strategy of effectively scaling your AI/ML undertaking and a tool that makes this possible - Metaflow.