Understand tf.add_n(): Add a List of Tensors – TensorFlow Tutorial

TensorFlow tf.add_n() function can allow us to add a list of tensors. In this tutorial, we will introduce how to use this function using some examples.

Syntax

tf.add_n(
    inputs,
    name=None
)

Add a list of tensors and return a tensor.

Parameter

inputs: A list of tensors

How to get a list of tensors?

There are two ways to get a list of tensors:

(1) Use python list comprehension

Understand Python List Comprehension for Beginners – Python Tutorial

For example:

lossL2 = tf.add_n([tf.nn.l2_loss(v) for v in vars if v.name not
                           in ['bias', 'gamma', 'b', 'g', 'beta']]) * l2_coef

(2) Use tf.split(), tf.unstack() function and so on. These tensorflow function will return a list of tensors.

TensorFlow tf.split(): Splits a Tensor into Sub Tensors

How to use tf.add_n() in tensorflow?

We will use an example to show you how to use it.

Create a tensor with the shape (2, 3)

import tensorflow as tf
import numpy as np

x1 = tf.convert_to_tensor(np.array([[1, 2, 3],[2, 3, 4]]), dtype = tf.float32)

x1 is a (2, 3) tensor.

Split x1 to a list of tensors

z1 = tf.unstack(x1)

z1 will be [tensor(1, 2, 3), tensor([2, 3, 4])]

Add z1 using tf.add_n()

y1 = tf.add_n(z1)
init = tf.global_variables_initializer()
with tf.Session() as sess:
    sess.run(init)
    print(sess.run(y1))

Run this code, y1 will be:

tensor([3, 5, 7])