JavaScript

The tf.squaredDifference() Function in TensorFlow.js

In the Tensorflow.js library, the tf.squaredDifference() function is used to return the square difference of each element present in two tensors/scalars.

Computation

(i-j)*(i-j)

Where “i” is the element in the first tensor, and “j” is the element in the second tensor.

Syntax:

tensor.squaredDifference()

Return:

It returns a positive integer even if the result is negative.

Example 1

Create two 1D tensors with 4 integers and get the squared difference.

<html>
<!--   CDN Link that delivers the Tensorflow.js framework -->
<script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs"></script>
 
<body>
<script>
 
 //tensor1
let values1 = tf.tensor1d([0,1,2,3]);
 //tensor2
let values2 = tf.tensor1d([1,4,1,5]);

//display
document.write("<b>Tensor-1: </b> "+values1);
document.write("<br>");
document.write("<b>Tensor-2: </b> "+values2);
document.write("<br>");

//return the squared difference
document.write("<b>Squared Difference: </b> "+values1.squaredDifference(values2));


</body>
</html>

Output:

Working:
(0-1)*(0-1) = 1
(1-4)*(1-4)=9
(2-1)*(2-1)=1
(3-5)*(3-5)=4

Example 2

Create two 1D tensors with 5 integers and get the squared difference.

<html>
<!--   CDN Link that delivers the Tensorflow.js framework -->
<script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs"></script>
 
<body>
<script>
 
 //tensor1
let values1 = tf.tensor1d([12,34,56,78,90]);
 //tensor2
let values2 = tf.tensor1d([12,34,56,78,90]);

//display
document.write("<b>Tensor-1: </b> "+values1);
document.write("<br>");
document.write("<b>Tensor-2: </b> "+values2);
document.write("<br>");

//return the squared difference
document.write("<b>Squared Difference: </b> "+values1.squaredDifference(values2));
</script>

</body>
</html>

Output:

If the values in a tensor are the same, then the squared difference is 0.

It is also possible to get the squared difference with a tensor and a scalar.

Example 3

Create a1D tensor with 5 integers, a scalar with a value of 2 and get the squared difference.

<html>
<!--   CDN Link that delivers the Tensorflow.js framework -->
<script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs"></script>
 
<body>
<script>
 
 //tensor
let values = tf.tensor1d([12,34,56,78,90]);
 //scalar
let value = tf.scalar(2);

//display
document.write("<b>Tensor: </b> "+values);
document.write("<br>");
document.write("<b>Scalar: </b> "+value);
document.write("<br>");

//return the squared difference
document.write("<b>Squared Difference: </b> "+values.squaredDifference(value));
</script>

</body>
</html>

Output:

Working:

  1. (12-2)*(12-2)=100
  2. (34-2)*(34-2)=1024
  3. (56-2)*(56-2)=2916
  4. (78-2)*(78-2)=5776
  5. (90-2)*(90-2)=7744

Conclusion

In this article, we saw how to return the square difference of elements in tensors using tf.squaredDifference(). It performs computation using the formula (i-j)*(i-j), where “i” is the element in the first tensor and “j” is the element in the second tensor. Also, we discussed the different examples with different cases

About the author

Gottumukkala Sravan Kumar

B tech-hon's in Information Technology; Known programming languages - Python, R , PHP MySQL; Published 500+ articles on computer science domain