Output - Aster Analytics

Teradata AsterĀ® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Published
September 2017
Language
English (United States)
Last Update
2018-04-17
dita:mapPath
uce1497542673292.ditamap
dita:ditavalPath
AA-notempfilter_pdf_output.ditaval
dita:id
B700-1022
lifecycle
previous
Product Category
Software

The Forest_Drive function outputs a message (to the screen) and a table.

Forest_Drive Output Message Schema
Column Name Data Type Description
message VARCHAR
Computing n1 classification trees.
Each worker is computing n2 trees.
Each tree will contain approximately n3 points.
Poisson sampling parameter: n4
Query finished in n5 seconds.
Out of bagging estimate of error rate: n6 %
Decision forest created in table "table".
Forest_Drive Output Table Schema
Column Name Data Type Description
worker_ip VARCHAR IP address of the worker that produced the decision tree.
task_index INTEGER Identifier of the worker that produced the decision tree.
tree_num INTEGER Identifier of the decision tree.
tree VARCHAR JSON representation of the decision tree. For JSON types that can appear in the tree representation, see the following table.
JSON Types in JSON Representation of Decision Tree
JSON Type Description
responseCounts Appears only for classification trees. Number of observations in each class at the node identified by id.
sum Appears only for regression trees. Sum of the values of the response variable at the node identified by id.
sumSq Appears only for regression trees. Sum of the squared values of the response variable at the node identified by id.
size Total number of observations at the node identified by id.
id Node identifier.
maxDepth Maximum possible depth of the tree, starting from the node identified by id. For the root node, this is the value of the MaxDepth argument in the function call. For leaf nodes, this value is 0. For other nodes, this value is the maximum possible depth of the tree, starting from that node.
split Indicates the start of a JSON item describing a split at the node identified by id.
attr Attribute (predictor) on which the algorithm split at the node identified by id.
leftCategories Appears only for splits where the attribute (identified by attr) is categorical. Categories assigned to the left child of the split.
rightCategories Appears only for splits where the attribute is categorical. Categories assigned to the right child of the split.
score Gini score of the node identified by id.
type Type of tree and split. Possible values:
  • CLASSIFICATION_CATEGORICAL_SPLIT
  • CLASSIFICATION_NUMERIC_SPLIT
  • REGRESSION_CATEGORICAL_SPLIT
  • REGRESSION_NUMERIC_SPLIT
leftNodeSize Number of observations assigned to the left node of the split.
rightNodeSize Number of observations assigned to the right node of the split.
scoreImprove Score improvement at the node identified by id.
leftChild Indicates the start of a JSON item describing the left child of the node identified by id.
rightChild Indicates the start of a JSON item describing the right child of the node identified by id.
nodeType Type of the node identified by id. Possible values:
  • CLASSIFICATION_NODE
  • CLASSIFICATION_LEAF
  • REGRESSION_NODE
  • REGRESSION_LEAF