Fill
Fill nulls with a value, a neighbour or the mean.
Fills nulls in columns by strategy: a literal value, the previous or next non-null, zero, or the column mean. forward and backward follow the frame's current row order, so sort first.
Without groupBy, forward, backward and mean treat every row as one series. When the frame holds several series side by side (one per store, SKU or oven), set groupBy to the column naming that entity: each blank is then filled only from rows of its own group, so a store's first blank days stay blank instead of taking the last reading of the store sorted above it, and mean is that store's own mean. While no group is chosen, the inspector points out columns that look like such a key.
At a glance
Module type
Fill
Category
Primitives · Dataframe
Default tier
reactive
Runs on
reactive, durable
Starts a flow
no
Holds a body
no
Keeps state between runs
no
Available in
every application type
Parameters
value
string | number | boolean | null · required · defaults to null
The literal used by the value strategy.
columns
string[] · required · defaults to []
The columns whose nulls are filled.
strategy
"value" | "forward" | "backward" | "zero" | "mean" · required · defaults to "value"
What to fill with: a literal, a neighbour, zero, or the column mean.
groupBy
string[] · optional · defaults to []
The columns that name which series a row belongs to. forward, backward and mean read only rows with the same values in these columns, keeping the frame's row order inside each group. With columns empty, the group columns themselves are not filled. value and zero ignore it.
Inputs
on
event
no
Implicit on every module that does not itself start a flow: wire an event here to make this module the start of what that event runs.
data
dataframe
yes
The frame to work on.
Outputs
output
dataframe
dataframe, signal
The frame with those nulls filled.
Last updated