Discover the essential properties missing in Azure AI Knowledge Store JSON configurations. Learn how to define table projections with examples to ace your AI-102 certification exam.
Table of Contents
Question
You need to create a projection for a table that will be a knowledge store for AI enriched content from the AI Search service in Azure Storage.
You create the following skillset definition for the JSON configuration table node:
“knowledgeStore” : {
“storageConnectionString”: “DefaultEndpointsProtocol=https;AccountName=<Acct Name>;AccountKey=<Acct Key>;”,
“projections” : [
{
“tables”: [
{ A : “Hotels”,
B : “HotelId”,
C : “/document/tableprojection” }
]
}
]
Which properties are missing in the JSON configuration table node?
Drag the property to the appropriate letter.
Property:
- source
- storageContainer
- tableName
- targetName
- generatedKeyName
Answer
A. tableName
B. generatedKeyName
C. source
Explanation
The following code creates a skillset definition for the JSON configuration table node:
“knowledgeStore” : {
“storageConnectionString”: “DefaultEndpointsProtocol=https;AccountName=<Acct Name>;AccountKey=<Acct Key>;”,
“projections” : [
{
“tables”: [
{ “tableName”: “Hotels”,
“generatedKeyName”: “HotelId”,
“source”: “/document/tableprojection” }
]
}
]
A table projection must contain three properties:
tablename: The name of the table.
generatedKeyName: The column name of the key that uniquely identifies each row in the table. The values of the key are system generated.
source: Specifies the path to the resource in a tree-like data structure. Typically, it is the output of a Shaper skill that specifies the shape of the table.
In this scenario, the tableName value is “Hotels”, the value for the generatedKeyName is “HotelId” which is the column name of the key, and the value for source is “/document/tableprojection”.
You would not specify the storageContainer property for the value of “Hotels”. This property can be used for a file projection or an object projection.
The following JSON script creates a file projection for images stored in a folder named “myImages”.
“projections”: [
{
“tables”: [ ],
“objects”: [ ],
“files”: [
{
“storageContainer”: “myImages”,
“source”: “/document/normalized_images/*”
}
]
}
]
You would not use the targetName property. This property is used to specify the destination of the output.
Microsoft Azure AI Engineer Associate AI-102 certification exam practice question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the Microsoft Azure AI Engineer Associate AI-102 exam and earn Microsoft Azure AI Engineer Associate AI-102 certification.