Install UDFs on ML Engine | Teradata Vantage - Install UDFs - Teradata Vantage

Machine Learning Engine User Guide

Product
Teradata Vantage
Release Number
9.02
9.2.4
9.2.3
2.3.2
1.3
Published
May 2022
Language
English (United States)
Last Update
2022-05-04
dita:mapPath
kaz1597341649653.ditamap
dita:ditavalPath
ehs1594682156756.ditaval
dita:id
B700-4004
lifecycle
previous
Product Category
Teradata Vantageā„¢

You can upload dtw.zip (the zip file containing the function or file) from any client machine. Specify the schema to which the installation should occur. You can only install in your private schema or public schema.

A unique uuid or udfid will be generated for each of the UDFs installed. The uuid will be available in the response.

REST API

POST https://<CLUSTER_DOMAIN>/mlengine/udfmanager/api/v1/udfs

Example Request

curl \
   -X POST \
   -H "Accept: application/json" \
   -H "Content-Type: multipart/form-data" \
   -H "Authorization: Bearer <JWT token>" \
   -F 'file=@/home/user/dtw.zip' \
   -F 'schema=alice' (OR) 'schema=public' \
   "https://<CLUSTER_DOMAIN>/mlengine/udfmanager/api/v1/udfs"

Example Response

HTTP/1.1 201 OK
Content-Type: "application/json"
{
  "filename":"dtw.zip",
  "result":"200",
  "owner":"alice",
  "message":"function dtw.zip installed successfully",
  "id":"e4349670-c7e3-4214-b133-66c5708a1787",
  "schema":"alice"
}

You can optionally create a function alias for your UDF. See Function Alias.