7.00.02 - Single_Tree_Drive Arguments - Aster Analytics

Teradata Aster® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Published
September 2017
Content Type
Programming Reference
User Guide
Publication ID
B700-1022-700K
Language
English (United States)
Last Update
2018-04-17
InputTable
[Required if you omit AttributeTableName and ResponseTableName.] Specifies the name of the table that contains the input data set.
AttributeTableName
[Required if you omit InputTable.] Specifies the name of the table that contains the attribute names and the values.
ResponseTableName
[Required if you omit InputTable.] Specifies the name of the table that contains the response values.
OutputTable
Specifies the name for the output table that is to contain the final decision tree (the model table). The output_table must not exceed 64 characters.
AttributeNameColumns
Specifies the names of the attribute table columns that define the attribute.
AttributeValueColumn
Specifies the names of the attribute table columns that define the value.
ResponseColumn
Specifies the name of the response table column that contains the response variable.
IDColumns
Specifies the names of the columns in the response and attribute tables that specify the ID of the instance.
CategoricalAttributeTableName
[Optional] Specifies the name of the input table that contains the categorical attributes.
SaveFinalResponseTableTo
[Optional] Specifies the name for the output table that is to contain the final PID and response pair from the response table and the node_id from the final single drive tree. Default behavior: The function does not output this table.
SplitsTable
[Optional] Specifies the name of the input table that contains the user-specified splits. Default behavior: The function creates new splits.
SplitsValueColumn
[Optional] If you specify SplitsTableName, this argument specifies the name of the column that contains the split value. Default: splits_valcol if UseApproximateSplits is 'true', node_column (the AttributeValueColumn argument value) otherwise.
NumSplits
[Optional] Specifies the number of splits to consider for each variable. Default: 10. The function does not consider all possible splits for all attributes.
ApproxSplits
[Optional] Specifies whether to use approximate percentiles (true) or exact percentiles (false). Default: 'true'. Internally, the function uses percentile values as split values.
IntermediateSplitsTable
[Optional] Specifies the name for the intermediate splits table, if it is to be saved. Default behavior: The function does not save the intermediate splits table.
DropTable
[Optional] Specifies whether to drop the output table (specified by OutputTableName) if it already exists. Default: 'false'.
MinNodeSize
[Optional] Specifies the decision tree stopping criterion and the minimum size of any particular node within each decision tree. Default: 100.
MaxDepth
Specifies a decision tree stopping criterion. If the tree reaches a depth past this value, the algorithm stops looking for splits. Decision trees can grow up to (2(max_depth+1) - 1) nodes. This stopping criteria has the greatest effect on function performance. The maximum value is 60. Default: 5.
Weighted
[Optional] Specifies whether to build a weighted decision tree. Default: 'false'. If you specify 'true', you must also specify the WeightColumn argument.
WeightColumn
[Optional] Specifies the name of the response table column that contains the weights of the attribute values.
SplitMeasure
[Optional] Specifies the impurity measurement to use while constructing the decision tree. Default: 'gini'. If the tree is weighted, this value cannot be 'chisquare'.
OutputResponseProbDist
[Optional] Specifies whether to output probabilities. Default: 'false'. If you specify 'true', the function outputs the probability distributions for each node in the output table.
Set this argument to 'true' if you intend to input the model to the Receiver Operating Characteristic (ROC) function.
ResponseProbDistType
[Optional] Specifies the probability distribution type. Default: 'laplace'.