Andrew Maguire

Assuming we can still just have the singe anomaly rates special chart then I'd be ok with this (but will defer to @vkalintiris on this one obviously). This chart: https://london.my-netdata.io/api/v1/data?chart=anomaly_detection.anomaly_rates...

Thanks @svrnwnsch - this is more like a little grey area in our docs. That page you mention was planned to be like a central landing page for all things...

hmm - thanks @svrnwnsch i'll try recreate a similar set up on such Azure VM and see what i can see. > I assume over time more and more data...

adding PR to clean up the original docs as reference: https://github.com/netdata/netdata/pull/13238

I have a b1s and b2s node running on Azure and have ML enabled. Going to give it 24 hours and see how things look then in Netdata and Azure...

I have had both a b1s (dev) and b2s (dev-b2s) running for 24 hours now just with default set up - looks like about 1500 metrics on each by default....

issue i mentioned about adding some better telemetry to be more proactive around stuff like this: https://github.com/netdata/netdata/issues/13303

Hmm I see this recent discussion that looks like some work ongoing https://forum.netgate.com/topic/172459/netdata-doesn-t-work-in-pfsense-2-6-0 Unsure of specifics myself, might be more details in that thread as I do see one of...

I should add - if i just skip the nan with something like this in the loop: ``` for X, y in tqdm(iterator.iter(X_stream, y_stream)): # Stream data. X = preprocessor.fit_transform_partial(X)...

My bad - I left that part out. I am doing this: ``` # set up grid search gs_clf = GridSearchCV(full_pipeline, full_params, n_jobs=-1) ``` And then: ``` # do the...