Interface RandomForestProperties

Default values :

  • maxFeatures: 1.0,
  • replacement: true,
  • nEstimators: 50,
  • seed: 42,
  • useSampleBagging: true,
  • noOOB: false

See

delbot.RandomForestBaseOptions

Hierarchy

Properties

batchSize?: number
consoleInfo?: boolean
dataTraining: DataTraining
epoch?: number
isClassifier: boolean
maxFeatures: number

the number of features used on each estimator. * if is an integer it selects maxFeatures elements over the sample features. * if is a float between (0, 1), it takes the percentage of features.

nEstimators: number

number of estimator to use.

nameToLoad?: string
nameToSave?: string
noOOB: boolean

Don't calculate Out-Of-Bag predictions. Improves performance if set to true.

replacement: boolean

use replacement over the sample features.

seed: number

seed for feature and samples selection, must be a 32-bit integer.

selectionMethod?: "mean" | "median"

the way to calculate the prediction from estimators, "mean" and "median" are supported.

treeOptions: object

options for the tree classifier, see [ml-cart]https://mljs.github.io/decision-tree-cart/

useSampleBagging: boolean

use bagging over training samples.

useTfjsVis?: boolean

Generated using TypeDoc