Jack Kelly

I haven't checked the dataset but I suspect the problem is not with the code, but rather in my dataset, UK-DALE :) If you use `meter.power_series()` (i.e. without specifying a...

> I also believe that the problem is in the dataset Sure. Although, just to be clear: pretty much all datasets have missing data. It's not just UK-DALE :) >...

I'd still recommend zeroing-out the NaNs in the appliance data. Something like `power_series(resample_kwargs={"fill_value": 0})` might work. (not tested)

I must admit I don't fully understand what you count as a 'a set of labelled fingerprints for appliances'. NILMTK already has a way to store multiple submeter streams (e.g....

Our HDF files contain a DataFrame (2D matrix) for each meter. Each DataFrame contains every sample recorded by that meter. So if, as is the case with pretty much every...

> One of the things I believe is currently missing from the Hart implementation in nilmtk is step 'H' above i.e. the linking of the centroid models to any appliance...

@nipunreddevil in commit ff8f566d000c99d36ed0a104b98b0c08d00cecc9 you added 'power factor' to the list of `PHYSICAL_QUANTITIES`. Thank you! Should we now remove 'pf' and standardise on 'power factor'?

what term should we use for 'phase'? 'phase'? Or 'phase angle'? Or 'phase difference'? And what unit should we standardise on? Degrees? Radians? (This stuff is a bit beyond my...

Cool. Shall we standardise on 'phase angle' and degrees as the unit? Do any datasets currently include this information?

OK, I've added phase angle. I've forgotten: what units were we planning to use for 'energy' and 'cumulative energy'? kWh? Or Wh?