Tensorflow.js tf.LayersModel class .predictOnBatch() Method

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

The .predictOnBatch() function is used to return expectancies for an individual group of instances.

Syntax:

predictOnBatch(x)

Parameters:

x: It is the stated input instances, like a Tensor i.e. the models that has precisely one input or else an array of Tensors i.e. models that has more than one input. It can be of type tf.Tensor, or tf.Tensor[].

Return Value: It returns the tf.Tensor object or tf.Tensor[].

Example 1:

Javascript

// Importing the tensorflow.js library 

import * as tf from "@tensorflow/tfjs"

  
// Defining model 
const Mod = tf.sequential({ 

   layers: [tf.layers.dense({units: 2, inputShape: [30]})] 
}); 

  
// Calling predictOnBatch() method and 
// Printing output 
Mod.predictOnBatch(tf.randomNormal([6, 30])).print();

Output:

Tensor
    [[-1.4716092, -1.8019401],
     [-1.0033149, -0.2789704],
     [-0.4451316, 0.2422157 ],
     [-0.1512984, -0.0726933],
     [2.1483333 , 2.4668102 ],
     [0.4091003 , 0.8335327 ]]

Example 2:

Javascript

// Importing the tensorflow.js library 

import * as tf from "@tensorflow/tfjs"

  
// Calling predictOnBatch() method and 
// Printing output 
tf.sequential({ 

   layers: [tf.layers.dense({units: 3, inputShape: [40]})] 
}).predictOnBatch(tf.truncatedNormal([5, 40])).print();

Output:

Tensor
    [[-1.5034456, -0.3429004, -0.2388536],
     [0.0083699 , -0.3176711, 2.1414554 ],
     [1.1850954 , -0.4481514, 1.1278313 ],
     [-0.1004405, 1.420954  , 0.4890856 ],
     [0.4184967 , 0.1191952 , -0.0936601]]

Reference: https://js.tensorflow.org/api/latest/#tf.LayersModel.predictOnBatch

Article Tags :

JavaScript

Web Technologies

Tensorflow.js