Tensorflow.js tf.constraints.nonNeg() Function

Tensorflow.js is an open-source library that is being developed by Google for running machine learning models as well as deep learning neural networks in the browser or node environment.

The tf.constraints.nonNeg() function us used to create a nonNeg constraint. nonNeg is a non-negative weight constraint. It is inherited from constraint class. Constraints are the attributes of layers.

Syntax: 

 tf.constraints.nonNeg() 

Parameters: 

• w: It specifies the input weight variable. It is an optional parameter.

Return value: It returns tf.constraints.Constraint.

Example 1: 

Javascript

// Importing the tensorflow.Js library
import * as tf from "@tensorflow/tfjs"

// Use nonNeg() function
const constraint = tf.constraints.nonNeg( )

// Print
console.log(constraint)

Output 

{}

Example 2: In this example we will create a dense layer using nonNeg constraint and apply the layer formed to a tensor.

Javascript

// Importing the tensorflow.Js library
import * as tf from "@tensorflow/tfjs"

// Create nonNeg constraint using nonNeg() function
const constraint = tf.constraints.nonNeg()

// Create a new dense layer using nonNeg constraint
const denseLayer = tf.layers.dense({
    units: 4,
    kernelInitializer: 'heNormal',
    kernelConstraint: constraint ,
    biasConstraint: constraint ,
    useBias: true
});

// Create input
const input = tf.ones([2, 2]);

// Apply denseLayer to input
const output = denseLayer.apply(input);

// Print the output
output.print()

Output 

Tensor
    [[-0.7439224, -1.3572885, -1.2860565, 1.3913929],
     [-0.7439224, -1.3572885, -1.2860565, 1.3913929]]

Reference: https://js.tensorflow.org/api/1.0.0/#constraints.nonNeg