Tensorflow.js tf.clone() Function

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

The tf.clone() function is used to create a copy of a tensor. The tf.clone() function creates a new tensor of the same shape and value of another tensor.

Syntax:

tf.clone( x )

Parameters:

• x: It is the tensor which we want to clone. Its value can be tensor, Array, TypedArray type.

Returns: It returns a Tensor Object.

Example 1: In this example, we are creating a copy of the tensor x and storing it into y.

Javascript

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

// Creating a tensor object
const x = tf.tensor([6, 1]);

// Cloning the tensor x and
// storing it into y
const y = tf.clone(x);

// Printing the tensor
y.print();

Output :

[6, 1]

Example 2: In this example, we cloned the tensor x using x.clone() instead of tf.clone(x).

Javascript

import * as tf from "@tensorflow/tfjs"

// Creating a tensor object
const x = tf.tensor([12, 2]);

// Cloning the tensor x
const y = x.clone();

// Printing the tensor
y.print();

Output :

[12, 2]

Example 3: Have a look at the example below.

Javascript

// Creating a tensor x
const x = tf.tensor([2, 2]);

// Creating a clone of tensor x
// using clone() function
const y = x.clone();

// Creating a tensor a and
// storing the same value as x
const a = tf.tensor([2, 2]);

// Copying the value of a into
// b using assignment operator
const b = a;

console.log(x == y);  //  false
console.log(x === y); //  false
console.log(a == x);  //  false
console.log(a == b);  //  true
console.log(a === b); //  true

Output :

false
false
false
true
true

Reference: https://js.tensorflow.org/api/latest/#clone