NerReplace Input - Teradata Vantage

Teradata Vantage™ - Bring Your Own Model User Guide
Deployment
VantageCloud
VantageCore
Edition
VMware
Enterprise
IntelliFlex
Lake
Product
Teradata Vantage
Release Number
20.00
Published
May 2026
ft:locale
en-US
ft:lastEdition
2026-06-18
dita:mapPath
fee1607120608274.ditamap
dita:ditavalPath
ayr1485454803741.ditaval
dita:id
fee1607120608274
Input Table Description
InputTable Table that contains the following columns:
  • txt (VARCHAR or CLOB): The input text to process.
  • logits (BLOB or VARBYTE): The output of the model run with ONNXEmbeddings.
  • Any additional columns as required.

Supports Unicode input data.
ModelTable Table that contains the following columns:
  • model_id
  • model (BLOB)

When ModelTable contains more than one model, you must use a SELECT query with a WHERE clause to select the model, otherwise an error occurs.

ModelTable must be specified as DIMENSION, otherwise an error occurs.
TokenizerTable Table that contains the following columns:
  • model_id
  • tokenizer (BLOB): Contains the contents of the tokenizer.json file for the model used in ONNXEmbeddings.

When TokenizerTable contains more than one tokenizer, you must use a SELECT query with a WHERE clause to select the tokenizer, otherwise an error occurs.

TokenizerTable must be specified as DIMENSION, otherwise an error occurs.
ModelConfigurationsTable Table that contains the following columns:
  • model_id
  • config (BLOB): Contains the contents of the config.json file for the model used in ONNXEmbeddings.

When ModelConfigurationsTable contains more than one configuration, you must use a SELECT query with a WHERE clause to select the configuration, otherwise an error occurs.

ModelConfigurationsTable must be specified as DIMENSION, otherwise an error occurs.