HMMUnsupervised Output - 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 Vantageā„¢

The HMMUnsupervised function outputs a message and outputs three tables through JDBC.

Output Message Schema

Column Data Type Description
message VARCHAR Reports that the models will be saved to the output tables after the training process successfully completes.

InitStateTable Schema

The function saves this table to init_state_prob, whose default value is Pi in the current schema.

Column Data Type Description
model_column BIGINT, INTEGER, or VARCHAR [Column appears only with ModelColumn syntax element.] Model attribute.
state VARCHAR Hidden state of learned HMM.
probability DOUBLE PRECISION Initial state probability determined by function.

StateTransitionTable Schema

The function saves this table to state_transition_prob, whose default value is A in the current schema.

Column Data Type Description
model_column BIGINT, INTEGER, or VARCHAR [Column appears only with ModelColumn syntax element.] Model attribute.
from_state VARCHAR Hidden state of learned HMM, from which a transition emanates.
to_state VARCHAR Hidden state of learned HMM, to which a transition is made.
probability DOUBLE PRECISION Probability of a transition from from_state to to_state.

EmissionTable Schema

The function saves this table to emit_prob, whose default value is B in the current schema.

Column Data Type Description
model_column BIGINT, INTEGER, or VARCHAR [Column appears only with ModelColumn syntax element.] Model attribute.
state VARCHAR Hidden state of learned HMM.
observed VARCHAR Symbol in observed_id column of input table.
probability DOUBLE PRECISION Probability of emitting observed symbol, given that HMM is in hidden state.