Tensorflow.js tf.metrics.categoricalAccuracy() 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.metrics.categoricalAccuracy() function is used to return categorical accuracy between two tensor. It takes two tensor as a parameter.

Syntax:

tf.metrics.categoricalAccuracy(a, b);

Parameters: 

• a: The first specified tensor.
• b: The second specified tensor. It must have same data type as “a”.

Return Value: It returns the categorical accuracy of the two specified tensors “a” and “b”.

Example 1:

Javascript

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

// Initializing two tensors
const a = tf.tensor2d([[1, 0, 0, 1], [0, 1, 0, 1]]);
const b = tf.tensor2d([
    [0.1, 0.6, 0.01, 0.05],
    [0.1, 0.02, 0.05, 0.3]
]);

// Calling the .categoricalAccuracy() function
const accuracy = tf.metrics.categoricalAccuracy(a, b);

// Print tensor
accuracy.print();

Output:

Tensor
    [0, 0]

Example 2:

Javascript

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

// Initializing two tensors
const a = tf.tensor([1, 0, 0, 1]);
const b = tf.tensor([1, 0.6, 0.01, 0.05]);

// Calling the .categoricalAccuracy() function
const accuracy = tf.metrics.categoricalAccuracy(a, b);

// Print tensor
accuracy.print();

Output:

Tensor
    1

Example 3:

Javascript

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

// Initializing two tensors
const a = tf.tensor([0, 0, 0, 1]);
const b = tf.tensor([0.1, 0.8, 0.05, 0.05]);

// Calling the .categoricalAccuracy() function
const accuracy = tf.metrics.categoricalAccuracy(a, b);

// Print tensor
accuracy.print();

Output:

Tensor
    0

Reference: https://js.tensorflow.org/api/latest/#metrics.categoricalAccuracy