Last active
March 3, 2021 07:07
-
-
Save gSrikar/06f8152f2ade3f7dfc3bc0ede99b87b5 to your computer and use it in GitHub Desktop.
Tensors with different ranks shapes are created and printed out to the output. Tensor Ranks and Tensor Shapes are explained in detail at http://gsrikar.blogspot.com/2017/06/what-is-tensor-rank-and-tensor-shape.html
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
''' | |
Class creates tensors of different ranks and shapes | |
''' | |
import tensorflow as tf | |
def main(): | |
''' | |
Main method | |
''' | |
# Rank 0 tensor with 0-D shape and type float | |
scalar = tf.constant(48.3) | |
print('Scalar: ', scalar) # Scalar: Tensor("Const:0", shape=(), dtype=float32) | |
# Rank 1 tensor with 1-D shape and type int | |
vector = tf.constant([1, 9, -6, 7, 0]) | |
print('Vector: ', vector) # Vector: Tensor("Const_1:0", shape=(5,), dtype=int32) | |
# Rank 2 tensor with 2-D shape and type float | |
matrix = tf.constant([[2.4, 5.1], [3.3, 7.9], [8.5, 6.1]]) | |
print('Matrix: ', matrix) # Matrix: Tensor("Const_2:0", shape=(3, 2), dtype=float32) | |
# Rank 3 tensor with 3-D shape and type int | |
tensor = tf.constant([[[2, 5, 6], [5, 3, 3], [6, 7, 8]], | |
[[0, 0, 1], [9, 7, 9], [2, 3, 6]], | |
[[4, 8, 2], [1, 0, 8], [4, 4, 0]]]) | |
print('Tensor: ', tensor) # Tensor: Tensor("Const_3:0", shape=(3, 3, 3), dtype=int32) | |
# Start the session | |
with tf.Session() as sess: | |
# Print the values | |
print('Scalar value: ', sess.run(scalar)) # Scalar value: 48.3 | |
print('Vector value: ', sess.run(vector)) # Vector value: [ 1 9 -6 7 0] | |
print('Matrix value: ', sess.run(matrix)) # Matrix value: [[ 2.4000001 5.0999999 ] | |
# [ 3.29999995 7.9000001 ] | |
# [ 8.5 6.0999999 ]] | |
print('Tensor value: ', sess.run(tensor)) # Tensor value: [[[2 5 6] | |
# [5 3 3] | |
# [6 7 8]] | |
# | |
# [[0 0 1] | |
# [9 7 9] | |
# [2 3 6]] | |
# | |
# [[4 8 2] | |
# [1 0 8] | |
# [4 4 0]]] | |
# Print the ranks | |
print('Scalar Rank: ', sess.run(tf.rank(scalar))) # Scalar Rank: 0 | |
print('Vector Rank: ', sess.run(tf.rank(vector))) # Vector Rank: 1 | |
print('Matrix Rank: ', sess.run(tf.rank(matrix))) # Matrix Rank: 2 | |
print('Tensor Rank: ', sess.run(tf.rank(tensor))) # Tensor Rank: 3 | |
# Print the shapes | |
print('Scalar Shape: ', sess.run(tf.shape(scalar))) # Scalar Shape: [] | |
print('Vector Shape: ', sess.run(tf.shape(vector))) # Vector Shape: [5] | |
print('Matrix Shape: ', sess.run(tf.shape(matrix))) # Matrix Shape: [3 2] | |
print('Tensor Shape: ', sess.run(tf.shape(tensor))) # Tensor Shape: [3 3 3] | |
if __name__ == '__main__': | |
''' | |
Starting point | |
''' | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment