Tensorflow.js tf.clipByValue() Function

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

The .clipByValue() function is used to find the clip values of the stated tensor input and is done element wise.

Syntax :  

tf.clipByValue(x, clipValueMin, clipValueMax)

Parameters:  

• x: It is the tensor input whose values are to be clipped, and it can be of type tf.Tensor, or TypedArray, or Array.
• clipValueMin: It is the lower bound of the stated range which is to be clipped.
• clipValueMax: It is the upper bound of the stated range which is to be clipped.

Return Value: It returns the tf.Tensor object.

Example 1: In this example, we are defining an input tensor of integer type and then printing the clipped values of it. For creating an input tensor we are utilizing the .tensor1d() method and in order to print the output we are using the .print() method.  

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

// Defining tensor input elements
const y = tf.tensor1d([9, -10, -19, 12]);

// Calling clipByValue() method and
// printing output
y.clipByValue(-11, 11).print();

Output:

Tensor
    [9, -10, -11, 11]

Example 2: In this example, float type values are considered as tensor input and all the parameters are passed directly to the clipByValue function.

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

// Defining float values
var val = [.5, 4.9, .275, -1.46];

// Calling tensor1d method
const y = tf.tensor1d(val);

// Calling clipByValue() method
var res = tf.clipByValue(y, -1, 1);

// printing output
res.print();

Output:

Tensor
    [0.5, 1, 0.275, -1]

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