Python – tensorflow.eye()

TensorFlow is open-source Python library designed by Google to develop Machine Learning models and deep learning  neural networks. 

tensorflow.eye() is used to generate identity matrix.

Syntax: tensorflow.eye( num_rows, num_columns, batch_shape, dtype, name)

Parameters:

• num_rows: It is int32 scalar Tensor which defines the number of rows to be present in resulting matrix.
• num_columns(optional): It is int32 scalar Tensor which defines the number of columns to be present in resulting matrix. It’s default value is num_rows.
• batch_shape(optional): It is list or tuple of Python integers or a 1-D int32 Tensor. If it’s not none  the returned Tensor will have leading batch dimensions of this shape. 
• dtype(optional): It defines the dtype of returned tensor. Default value is float32.
• name(optional): It defines the name for the operation.

Return : It returns a Tensor of shape batch_shape + [num_rows, num_columns].

Example 1:

Python3

# Importing the library
import tensorflow as tf

# Initializing the input
num_rows = 5

# Printing the input
print('num_rows:', num_rows)

# Calculating result
res = tf.eye(num_rows)

# Printing the result
print('res: ', res)

Output:

num_rows: 5
res:  tf.Tensor(
[[1. 0. 0. 0. 0.]
 [0. 1. 0. 0. 0.]
 [0. 0. 1. 0. 0.]
 [0. 0. 0. 1. 0.]
 [0. 0. 0. 0. 1.]], shape=(5, 5), dtype=float32)

Example 2:

Python3

# Importing the library
import tensorflow as tf

# Initializing the input
num_rows = 5
num_columns = 6
batch_shape = [3]

# Printing the input
print('num_rows:', num_rows)
print('num_columns:', num_columns)
print('batch_shape:', batch_shape)

# Calculating result
res = tf.eye(num_rows, num_columns, batch_shape)

# Printing the result
print('res: ', res)

Output:

num_rows: 5
num_columns: 6
batch_shape: [3]
res:  tf.Tensor(
[[[1. 0. 0. 0. 0. 0.]
  [0. 1. 0. 0. 0. 0.]
  [0. 0. 1. 0. 0. 0.]
  [0. 0. 0. 1. 0. 0.]
  [0. 0. 0. 0. 1. 0.]]

 [[1. 0. 0. 0. 0. 0.]
  [0. 1. 0. 0. 0. 0.]
  [0. 0. 1. 0. 0. 0.]
  [0. 0. 0. 1. 0. 0.]
  [0. 0. 0. 0. 1. 0.]]

 [[1. 0. 0. 0. 0. 0.]
  [0. 1. 0. 0. 0. 0.]
  [0. 0. 1. 0. 0. 0.]
  [0. 0. 0. 1. 0. 0.]
  [0. 0. 0. 0. 1. 0.]]], shape=(3, 5, 6), dtype=float32)