Upsample

The Upsample function is part of the HistoryFill class.

It's used to increase the resolution of low-res data by interpolating between the recorded points.

This is useful when you have unusually coarse data - such as hourly figures in a system where most of your data has 1s resolution. As such, this function is often used when bringing in information from….

• Enterprise Systems,
• Online APIs (ie. weather, stock prices etc.)
• Manual Observations,
• Informal Systems etc.

Note that if you're planning on upscaling your data, it's often a good idea to deliberately oversample your data.

For example, if you only have hourly information, it's often a good idea to request the previous and the next hours data when making a query. This way you'll be able to interpolate your analogue data to produce a smooth line rather than having a trend that contains abrupt changes.

The function takes one parameter…

This function is only for use on analogue data points. Do not use this function on discrete (ie. on/off) data.

#Record all of the points, building a list of the names as we go...
pointnamelist = []
for p in allpoints:
   if p.name not in pointnamelist:
      pointnamelist.append(p.name)
   history.RecordAnalogue(p.name,p.time,p.value)
 
#Upscale the low-resolution data
history.Upsample(pointnamelist)

You can also do the opposite job - Downsample is used to reduce the resolution of your data.