Filling missing values

Text column link

Fill with value link

You can easily fill missing values of a particular of the dataset using value. In the following example I am replacing all missing values with the value “Unknown”.

  [[preprocess.columns]]
name = 'short_name'
type = 'string'
operations = [
    {op = "fillna", value = "Unknown"}
]
  

Numeric column link

For numeric columns, there are few missing value replacement.

Fill with value link

  [[preprocess.columns]]
name = 'weight_kg'
type = 'int'
operations = [
    {op = "fillna", value = 0}
]
  

method = mean link

Replaces missing values in a numeric column with the mean of the non-missing values. This maintains dataset size while minimizing bias introduced by missing data. Suitable when data is missing at random and the distribution is approximately symmetric.

  [[preprocess.columns]]
name = 'weight_kg'
type = 'int'
operations = [
    {op = "fillna", method = "mean"}
]
  

method = median link

Replaces missing values in a numeric column with the median of the non-missing values. Useful for skewed distributions or when outliers are present, as the median is more robust to extreme values than the mean.

  [[preprocess.columns]]
name = 'weight_kg'
type = 'int'
operations = [
    {op = "fillna", method = "median"}
]