Datatype Change in Accumulated Columns | Teradata Vantage - Datatype Change in Accumulated Columns - Teradata Vantage

Teradata® VantageCloud Lake

Deployment
VantageCloud
Edition
Lake
Product
Teradata Vantage
Published
January 2023
Language
English (United States)
Last Update
2024-02-17
dita:mapPath
phg1621910019905.ditamap
dita:ditavalPath
pny1626732985837.ditaval
dita:id
phg1621910019905
The following functions change the data type of a column in the input table to a different data type in the output table:
  • TD_NumApply

    If you set the Inplace argument as false and specify the target columns in the Accumulate argument, then the data type of target columns in the output can be REAL or FLOAT.

  • TD_StrApply

    If you set the Inplace argument as false and specify the target columns in the Accumulate argument, then for the following StringOperation argument values, the data type of target columns is VARCHAR (UNICODE) in the output:

    STRINGCON, STRINGLIKE, STRINGPAD, STRINGTRIM, STRINGINDEX ]

  • TD_NonLinearCombineTransform

    If you specify the target columns in the Accumulate argument, then the data type of the target columns in the output can be REAL or FLOAT.

  • TD_KMeansPredict

    If you specify the target columns in the Accumulate argument, then the data type of the target columns in the output can be REAL or FLOAT.

  • TD_Silhouette

    If you specify the target columns in the Accumulate argument, then the data type of the target columns in the output can be REAL or FLOAT.

  • TD_RandomProjectionTransform

    If you specify the target columns in the Accumulate argument, then the data type of the target columns in the output can be REAL or FLOAT.

  • TD_SentimentExtractor

    If you specify the Text column in the Accumulate argument, then the data type of the Text column in the output is VARCHAR (UNICODE).

  • TD_FunctionTransform

    If a numeric column is not specified in the IDColumns argument, then the data type of the numeric column in the output can be REAL or FLOAT.

  • Pack

    If you set the Colcast argument as True and specify the target columns in the Accumulate argument, then the data type of target columns in the output can be VARCHAR.

  • TD_Qqnorm and TD_PolynomialFeaturesTransform

    The target columns are by default included in the output. The data type of the target columns included in the output can change to Double Precision or FLOAT.

  • TD_NBTCT

    If you specify the datatype for the token or doccategory columns as Char or VARCHAR, then the data type of token or category in the output can be VARCHAR Unicode.

  • TD_Histogram

    If you specify the datatype for label from minmaxtable table schema as BYTEINT, SMALLINT, INTEGER, or BIGINT, then the data type of the label in the output can be BIGINT. If you specify CHAR or VARCHAR as the datatype for the label, then the data type of the label in the output can be VARCHAR Unicode.