Skip to main content

Varicent ELT Help Center

Oversample

Abstract

Generate more data or address classification problems in your data.

To address the imbalance in data, supplement the data with multiple copies of some of the minority classes.

A new column is added to indicate that Oversample generated the record.

When to use this tool

Use the Oversample tool to generate more data or to address classification problems in your data.

Configuration

Use the following configuration options to help configure the Oversample tool.

Configuring the Oversample tool

  1. In Varicent ELT, go to the Pipes module.

  2. On the Pipes tab, find the pipe you want to work with. Click the pipe to open.

  3. In your Pipe builder add your data source.

  4. On the canvas toolbar, click symon_add_icon.png + Tool.

  5. In the Tools modal search bar, type Oversample.

    Tip

    You can also find the Oversample tool in the Data section.

  6. Click + Add Tool.

  7. Connect the tool to your data set.

  8. In the configuration pane, enter the following information:

    Table 59. Oversample tool configuration

    Field

    Description

    Target column

    Select the column to use for the target column.

    Oversample column

    Enter the name to use for the oversample column.