Andrew Maguire

all looking good to me

Yeah interval maybe should have to be part of the config of each query or job.

We had talked about the ability to add "Saved Views" onto the AA tab in some way such that you could just filter the space of stuff being looked at...

maybe its just a way for users to define and save netdata simple patterns that can then be applied like a filter in some easy way. https://learn.netdata.cloud/docs/agent/libnetdata/simple_pattern (obviously auto creating...

I'd imagine youd need to focus on some sort of multimodel feature representation in your pipeline before you then train a model using PyOD. https://en.wikipedia.org/wiki/Multimodal_learning [This course](https://www.youtube.com/watch?v=VIq5r7mCAyw) is a good...

Lets say each observation is an image and some text. You could train some embedding models for each - so lets assume now for the image you have 10 numbers...

For anyone who wants to use the version created by @FedericoCeratto i have made a copy of it in here: https://github.com/netdata/community/tree/main/collectors/python.d.plugin/clickhouse And added some instructions on how to use it...

So cool! if possible can you add some docstrings to some of the important functions to make it easier to try just jump in and understand them? I'm not 100%...

What would be a good way to try help test this out? Would it be feasible for me to try this "locally" in some way?

Wondering, for a big PR like this, is there a way or does make sense to be able to deploy this as a feature branch to it's own testing env...