Training and Scoring Multiple Micro Models for Each Partition | Open Analytics Framework on VantageCloud Lake - Training and Scoring Multiple Micro Models for Each Partition

Use case: You want to create multiple micro models for each partition of the data (e.g., each product, each time period) and then score these models simultaneously.

Prerequisite steps:

Import the python library (teradataml) and specific environment setup module.

import getpass
import os
from collections import OrderedDict
from teradataml import create_context, remove_context, copy_to_sql, DataFrame, create_env, get_env, set_user_env, list_user_envs, remove_env
from teradataml.scriptmgmt import UserEnv, lls_utils
from teradataml.table_operators import Apply
from teradatasqlalchemy.types import FLOAT, BLOB

Connect from a client to a target VantageCloud Lake system where the training and scoring tasks will be performed.

print("Creating the context...")
host =  getpass.getpass("Host: ")
username = getpass.getpass("Username: ")
password = getpass.getpass("Password: ")
engine = create_context(host=host, username=username, password=password)

Generate the authentication token using set_auth_token API.

from teradataml import set_auth_token
ues_url = getpass.getpass("ues_url: ")
set_auth_token(ues_url=ues_url)

Training and Scoring Multiple Micro Models for Each Partition | Open Analytics Framework on VantageCloud Lake - Training and Scoring Multiple Micro Models for Each Partition - Teradata Vantage

Teradata® VantageCloud Lake