The trace of a matrix is basic operation in deep learning, in this tutorial, we will write some examples to show how to calculte it using tensorflow.
Understand The Trace of a Matrix for Beginners – Deep Learning Tutorial
Preliminaries
import tensorflow as tf import numpy as np
Create 3*3 matrix
A = tf.constant([[1,2,3],[4,5,6],[7,8,9]], dtype=tf.float32)
Compute matrix A trace
A_tr = tf.trace(A)
Output the trace value
init = tf.global_variables_initializer()
init_local = tf.local_variables_initializer()
with tf.Session() as sess:
sess.run([init, init_local])
np.set_printoptions(precision=4, suppress=True)
tr_a = (sess.run([A_tr]))
print tr_a
The value is: 15.0
If matrix A is not n * n?
Create a matrix A 3 * 4
A = tf.constant([[1,2,3,5],[4,5,6,7],[7,8,9,8]], dtype=tf.float32)
Compute its trace
A_tr_ = tf.trace(A)
The trace value is also 15.0
Meanwhile, if the dimension of matrix A is not 2, how about 2 * 3 * 4
Create a 2 * 3 * 4 matrix
B = tf.Variable(tf.random_uniform([2,3,4], -0.01, 0.01), dtype=tf.float32)
Compute matrix B trace
B_tr = tf.trace(B)
The value is:
[0.0093 0.0114]