In some systems there may be non-numeric data that needs to be fed into a time series data model. In Koios, we treat all data fed in to the model as normalized floating point values. This can pose an issue when the value is a string some other datatype incapable of being coerced to a floating point number. Our solution is to allow the user to create an enumeration table to “lookup” values.
To enable this feature, go to the General
settings using the tag’s detail dropdown, and expand the Advanced Options
menu at the bottom of the form. From here, check the box that says Use Lookup Table
.
Once that is done, the Lookup Table
tab will appear for this tag.
In the case above, the tag is looking for a string value of “On” or “Off” coming in as an input. This is then being converted to the output of 0 or 1. These fields are editable and expandable into large lookup tables for complex enumerations. By clicking + Add Row
, you are prompted with this popup where you can add new rows. To delete rows, click on one and select Delete Row
.