CustomTransformationConfiguration
Settings for customizing steps in the data source content ingestion pipeline.
You can configure the data source to process documents with a Lambda function after they are parsed and converted into chunks. When you add a post-chunking transformation, the service stores chunked documents in an S3 bucket and invokes a Lambda function to process them.
To process chunked documents with a Lambda function, define an S3 bucket path for input and output objects, and a transformation that specifies the Lambda function to invoke. You can use the Lambda function to customize how chunks are split, and the metadata for each chunk.
Contents
- intermediateStorage
-
An S3 bucket path for input and output objects.
Type: IntermediateStorage object
Required: Yes
- transformations
-
A Lambda function that processes documents.
Type: Array of Transformation objects
Array Members: Fixed number of 1 item.
Required: Yes
See Also
For more information about using this API in one of the language-specific AWS SDKs, see the following:
