new Strategy(algo, impurity, maxDepth, numClasses, maxBins, quantileCalculationStrategy, categoricalFeaturesInfo, minInstancesPerNode, minInfoGain, maxMemoryInMB, subsamplingRate, useNodeIdCache, checkpointInterval)
Parameters:
Name |
Type |
Description |
algo |
Algo
|
|
impurity |
Impurity
|
|
maxDepth |
number
|
|
numClasses |
number
|
|
maxBins |
number
|
|
quantileCalculationStrategy |
QuantileStrategy
|
|
categoricalFeaturesInfo |
Map
|
|
minInstancesPerNode |
number
|
|
minInfoGain |
number
|
|
maxMemoryInMB |
number
|
|
subsamplingRate |
number
|
|
useNodeIdCache |
boolean
|
|
checkpointInterval |
number
|
|
- Source:
Methods
Parameters:
Name |
Type |
Description |
algo |
Algo
|
|
- Source:
Returns:
-
Type
-
module:eclairjs/mllib/tree/configuration.Strategy
Construct a default set of parameters for DecisionTree
Parameters:
Name |
Type |
Description |
algo |
Algo
|
Algo.Classification or Algo.Regression |
- Source:
Returns:
-
Type
-
module:eclairjs/mllib/tree/configuration.Strategy
Construct a default set of parameters for DecisionTree
Parameters:
Name |
Type |
Description |
algo |
string
|
"Classification" or "Regression" |
- Source:
Returns:
-
Type
-
module:eclairjs/mllib/tree/configuration.Strategy
Returns a shallow copy of this instance.
- Source:
Returns:
-
Type
-
module:eclairjs/mllib/tree/configuration.Strategy
isMulticlassClassification() → {boolean}
- Source:
Returns:
-
Type
-
boolean
isMulticlassWithCategoricalFeatures() → {boolean}
- Source:
Returns:
-
Type
-
boolean
setAlgo(algo)
Sets Algorithm using a String.
Parameters:
Name |
Type |
Description |
algo |
string
|
|
- Source:
setCategoricalFeaturesInfo(categoricalFeaturesInfo)
Sets categoricalFeaturesInfo using a JavaScript object with simple key/value map.
Parameters:
Name |
Type |
Description |
categoricalFeaturesInfo |
object
|
simple key/value map |
- Source:
setMaxDepth(num)
Parameters:
Name |
Type |
Description |
num |
integer
|
|
- Source:
setNumClasses(num)
Parameters:
Name |
Type |
Description |
num |
integer
|
|
- Source: