components llm_rag_update_acs_index

LLM - Update ACS Index

llm_rag_update_acs_index

Overview

Uploads embeddings into Azure Cognitive Search instance specified in acs_config. The Index will be created if it doesn't exist.

The Index will have the following fields populated:

• "id", String, key=True
• "content", String
• "contentVector", Collection(Single)
• "category", String
• "url", String
• "filepath", String
• "content_hash", String
• "meta_json_string", String

"meta_json_string" contains all metadata for a document serialized as a JSON string.

Version: 0.0.69

Tags

Preview

View in Studio: https://ml.azure.com/registries/azureml/components/llm_rag_update_acs_index/version/0.0.69

Inputs

Name	Description	Type	Default	Optional	Enum
embeddings	Embeddings output produced from parallel_create_embeddings.	uri_folder
acs_config	JSON string containing the ACS configuration. e.g. {"endpoint": "https://my-cog.search.windows.net", "index_name": "my-index", "api_version": "2021-04-30-Preview", "endpoint_key_name": "ACS-PROD-KEY"}	string
index_connection_id	The connection id of the ACS Index provider to use.	string		True

Outputs

Name	Description	Type
index	Uri folder containing MLINdex yaml describing the newly created/updated ACS index	uri_folder

Environment

azureml:llm-rag-embeddings@latest