Is it possible to provide multiple columns but only look for anomaly for one column?

[gautambak]

Hi There,

I've been playing with pyod and it's really amazing. Thank you for putting it together!

I have a weird problem, where I'm trying to find anomalies on an timeseries variable but I believe other variables can assess to detect the anomaly.

For example, if I'm trying to detect anomalies within temp. data but I believe that other observations can help also such as precipitation, then is it possible to provide both but have pyOD not look for anomalies in anything other than temp?