CODEWITHSUNDEEP
pythontensorflow
Uvuvwevwevwe
Uvuvwevwevwe

Reputation: 1041

How to print more than 3 values inside a tensor by using tf.Print()?

I have a following simple code for the purpose of splitting a tensor into three parts along axis=1

sess = tf.Session()
a = tf.constant([[1, 2, 3], [4, 5, 6], [7, 8, 9], [10, 11, 12]])

x, y, z = tf.split(a, 3, axis=1)

print_out_x = tf.Print(x, [x], message='Value of x: ', name='x_value')
print_out_y = tf.Print(y, [y], message='Value of y: ', name='y_value')
print_out_z = tf.Print(z, [z], message='Value of z: ', name='z_value')

sess.run([print_out_x, print_out_y, print_out_z])

print(print_out_x)
print(print_out_y)
print(print_out_z)

I got the output like below

enter image description here

How can I get full of values inside x, y, z instead of ... by using tf.Print()? More sepcifically, I want output should be

Value of z: [[3][6][9][12]]
Value of y: [[2][5][8][11]]
Value of x: [[1][4][7][10]]

Upvotes: 0

Views: 1128

Answers (1)

Vijay Mariappan
Vijay Mariappan

Reputation: 17191

You can use the Args: summarize in the tf.Print() function to set the number of arguments that need to be printed per intput tensor.

For ex, 

tf.Print(x, [x],summarize=x.shape[0], message='Value of x: ', name='x_value')
#Value of x: [[1][4][7][10]]

Upvotes: 2

Related Questions