embedding-tool.md

title	titleSuffix	description	services	ms.service	ms.subservice	ms.custom	ms.topic	author	ms.author	ms.reviewer	ms.date
Embedding tool in Azure Machine Learning prompt flow	Azure Machine Learning	The prompt flow embedding tool uses OpenAI's embedding models to convert text into dense vector representations for various natural language processing tasks.	machine-learning	machine-learning	prompt-flow	ignite-2023	reference	wangchao1230	CLWAN	lagayhar	11/02/2023

Embedding tool

OpenAI's embedding models convert text into dense vector representations for various natural language processing tasks. For more information, see the OpenAI Embeddings API.

Prerequisites

Create OpenAI resources:

• OpenAI:

  • Sign up your account on the OpenAI website.
  • Sign in and find your personal API key.

• Azure OpenAI Service:

  Create Azure OpenAI resources with these instructions.

Connections

Set up connections to provide resources in the embedding tool.

Type	Name	API key	API type	API version
OpenAI	Required	Required	-	-
AzureOpenAI	Required	Required	Required	Required

Inputs

Name	Type	Description	Required
input	string	Input text to embed.	Yes
connection	string	Connection for the embedding tool used to provide resources.	Yes
model/deployment_name	string	Instance of the text-embedding engine to use. Fill in the model name if you use an OpenAI connection. Insert the deployment name if you use an Azure OpenAI connection.	Yes

Outputs

Return type	Description
list	Vector representations for inputs

Here's an example response that the embedding tool returns:

Output

[-0.005744616035372019,
-0.007096089422702789,
-0.00563855143263936,
-0.005272455979138613,
-0.02355326898396015,
0.03955197334289551,
-0.014260607771575451,
-0.011810848489403725,
-0.023170066997408867,
-0.014739611186087132,
...]