Replacing Missing Values - Developer Documentation
Skip to content

Replacing Missing Values

This transformation allows you to replace missing values in your dataset at the column level. Here are some important points about replacing missing values:

  • You must specify a replacement type for each column in which you want to replace missing values.
  • You can only define one replacement value for each column.
  • This feature is not available for String and Boolean data types.
  • The drop-down list contains these replacement options:
    • Mean (default)—replaces missing values with the average of all the values in the selected column; available for double data types only.
    • Median—replaces missing values with the median of all the values for the selected column; available for double data types only.
    • Zero—replaces missing values with zero; available for all numeric data types.
  • After specifying the replacement values, select the columns to include in the dataset, preview the results, and register the dataset.

Replace Missing Values Transformation Panel

Here is an example of the Replace Missing Values transformation panel.

Replace Missing Values Transformation Panel

How to Replace Missing Values

Follow these steps to replace missing values in your dataset using the Replace Missing Values transformation:

  1. Navigate to the Manage Analytics Workspaces page and start a cluster. The cluster starts and displays "Cluster Running".
  2. Click Create a New Workspace or open an existing workspace. The Workspace opens on a new page.
  3. Click Add Transformation Panel. The Select Transformation dialog box opens.
  4. Select Replace Missing Values and click Select. A Replace Missing Values panel opens.
  5. Click Select a Dataset and select a dataset from the Select a Dataset pop-up window.
  6. On the Define Replacements tab, click Select and choose a column.
  7. Select an option (Mean, Median, Zero) for the Replace With field.
  8. Click the +Add Column button, select a column and replacement value. Repeat as needed for other columns.

Except where otherwise noted, content on this site is licensed under the Development License Agreement.

Last update: June 15, 2023