Skip to main content

schema_fit.png Schema Fit Tool

The Schema Fit tool transforms Source data to match the format of Target data. Use the Schema Fit tool to quickly add data from new sources to your existing workflows.

注意

The GenAI Tools are currently in Public Preview. Learn how to join the Public Preview and get started with AI-powered workflows!

Tool Components

The Schema Fit tool has 5 anchors (3 input and 2 output):

  • M input anchor: Use the M input anchor to connect the model connection settings from the LLM Override tool.

  • T input anchor: Use the T input anchor to connect your reference formatted Target data.

  • S input anchor: Use the S input anchor to connect your unformatted Source data.

  • T output anchor: Use the T output anchor to pass the steps used to transform your Source data. These steps are written in Python.

  • D output anchor: Use the D output anchor to pass the transformed data downstream.

Configure the Tool

  1. Add a Schema Fit tool to the canvas.

  2. Connect your Target data to the T input anchor. The Schema Fit tool references this data for the Source data transformation.

  3. Connect your Source data to the S input anchor. This is the data that you want to transform to match the Target data.

  4. Connect the M input anchor to an LLM Override tool.

  5. Enter the Number of transformations to make. This parameter sets the maximum number of transformation steps the tool might attempt. Note that more transformation steps might not result in improved performance. Use this setting to limit the max number of requests made to the LLM.

  6. If you only want to include the columns from your Target data in the output, select the Only include target schema columns checkbox. This option excludes columns from your Source data that don’t match your Target data.

  7. Run the workflow.

Output

The Schema Fit tool outputs 2 types of data:

  • The T output anchor outputs two columns…

    • Transformation Step: Description of the transformation step performed by the Code column.

    • Code: Contains the Python code used to transform your data. Each row relates to a different data transformation step.

  • The D output anchor outputs your transformed Source data with the same formatting as your Target data.

本节内容如下: