GMM Input - Teradata Vantage

Machine Learning Engine Analytic Function Reference
Product
Teradata Vantage
Release Number
9.02
9.01
2.0
1.3
Published
February 2022
Language
English (United States)
Last Update
2022-02-10
dita:mapPath
rnn1580259159235.ditamap
dita:ditavalPath
ybt1582220416951.ditaval
dita:id
B700-4003
lifecycle
previous
Product Category
Teradata Vantage™
Table Description
InputTable Contains input data to cluster.
InitialValues [Optional] Table or view that contains initial values for weight, mean, and covariance of each cluster. If this table or view is empty, function terminates without executing.

If initial mean values have precision beyond 8 decimal places, result may be nondeterministic.

If you do not specify a table or view, you must use query ON (SELECT 1), which causes function to determine initial values as follows:
Characteristic Initial Value
weight Same for all clusters.
mean Selected from multivariate standard normal distribution centered at origin.
covariance Covariance matrix computed by function and assigned to each cluster.

InputTable Schema

Column Data Type Description
id Any Data point identifier.
dim_n Any numeric SQL data type [Column appears once for each dimension.] Data in dimension i.

InitialValues Table Schema

Column Data Type Description
weight Any numeric SQL data type Initial weight of cluster.
dim_n Any numeric SQL data type [Column appears once for each dimension.] Initial mean of cluster. If initial mean values have precision beyond 8 decimal places, result may be nondeterministic.
covariance VARCHAR Initial covariance of cluster. Possible values depend on CovarianceType.
CovarianceType Possible Values
'spherical' Positive numeric value (for example, 1.0.)
'diagonal' JSON representation of DOUBLE PRECISION array (for example, [1.0,2.0,3.0,4.0])
'tied' or 'full' JSON representation of two-dimensional DOUBLE PRECISION array (for example, [[1.0,2.0],[2.0,4.0]])