API_Request In-database Function Input | Google Cloud Vertex AI | Vantage - API_Request In-database Function Input - Teradata Vantage

Teradata® VantageCloud Lake

Deployment
VantageCloud
Edition
Lake
Product
Teradata Vantage
Published
January 2023
ft:locale
en-US
ft:lastEdition
2024-12-11
dita:mapPath
phg1621910019905.ditamap
dita:ditavalPath
pny1626732985837.ditaval
dita:id
phg1621910019905

When API_Request in-database function is running with Google Vertex AI, it has the following input requirements.

Input table columns are used to form the model input for Google Vertex AI deployed models.

START_KEY_INDEX can be used to make a few columns pass through without going to Google Vertex AI models.

Along with the input table, ENDPOINT, AUTHORIZATION are required arguments.

Supported Input Data Types

API_Request supports following data types.

In CSV and JSON content types, all columns are serialized into a string and then a csv, json, or json-with-keys structure is formed.

Data Type Category Teradata Data Type
Numeric
  • BYTEINT
  • SMALLINT
  • INTEGER
  • BIGINT
  • DECIMAL/NUMERIC
  • FLOAT/REAL/DOUBLE PRECISION
Character
  • CHAR
  • VARCHAR
DateTime
  • DATE
  • TIME
  • TIMESTAMP
  • TIME WITH TIME ZONE
  • TIMESTAMP WITH TIME ZONE