Matrix operations using TensorFlow (Python)

01 August 2017 By Bhavyanshu Parasher

Operations

Below are some of the examples that you can use to learn TensorFlow. There are some basic matrix and vector operations. You can check out the generated data flow graphs using the tensorboard command.

Matrix Addition

Very basic addition of two matrices.

import tensorflow as tf

a = tf.Variable([[0,1], [2,3]], name="matrix_a")
b = tf.Variable([[2,4], [8,9]], name="matrix_b")
init = tf.variables_initializer([a, b], name="init")
x = tf.add(a, b)

with tf.Session() as s:
    writer = tf.summary.FileWriter('graphs', s.graph)
    s.run(init)
    print(s.run(x))

writer.close()

Sequential Operations

Result of addition of two matrices multiplied with Matrix B.

import tensorflow as tf

a = tf.Variable([[0,1], [2,3]], name="matrix_a")
b = tf.Variable([[2,4], [8,9]], name="matrix_b")
init = tf.variables_initializer([a, b], name="init")
add = tf.add(a, b)
final = tf.multiply(add, b)

with tf.Session() as s:
    writer = tf.summary.FileWriter('graphs', s.graph)
    s.run(init)
    print(s.run(final))

writer.close()

Launch TensorBoard

You can launch TensorBoard using the following command. Open your browser and go to localhost:8082

tensorard --logdir="./graphs" --port 8082

blog comments powered by Disqus

Share

Published

01 August 2017 By Bhavyanshu Parasher - [email protected]

Tags