Perceptron

Perceptron

Perceptron to train and predict

Constructor

new Perceptron()

Source:
See:
  • of
Example
const m =  Perceptron.of([0.2365698, 0.4567897, 0.95989898, -0.457878787])

Members

type :string

Source:
Type:
  • string

Methods

(static) of(weights) → {Perceptron}

Creates a Perceptron object
Source:
Parameters:
Name Type Description
weights array Weights for predicting
Returns:
Type:
Perceptron
Example
const weights = [0.123456, 0.458789, 0.9987454]
const m =  Perceptron.of(weights)

predict(row) → {Boolean}

Source:
Parameters:
Name Type Description
row Array Array of values to predict
Returns:
Type:
Boolean
Example
const m =  Perceptron.of()
const train  = [...] // dataset: [[0,1,2], [1,2,3]]
m.train(train, 0.1, 10000)
m.predict([1,2,1]) // returns 1 or 0

train(train, lRate, nEpoch)

Source:
Parameters:
Name Type Description
train Array Dataset to train the preceptron
lRate Number Set the learning rate of the perceptron
nEpoch Number Set the number of iterations to train the perceptron
Example
const m =  Perceptron.of()
const train  = [...] // dataset: [[0,1,2], [1,2,3]]
m.train(train, 0.1, 10000)
m.weights = [0.213131, -0.45464, 0.789797]