JIYIK CN >

This tutorial will show you how to convert a Tensor to a NumPy array in Python.

Use the function in Python Tensor.numpy()to convert a tensor to a NumPy array

Eager Execution of TensorFlow library can be used to convert tensor to NumPy array in Python. With eager execution Eager Execution, the behavior of TensorFlow library operations changes and these operations are executed immediately. We can also use to Eager Executionperform NumPy operations on Tensor objects. Tensor.numpy()The function converts a Tensor to a NumPy array in Python. In TensorFlow 2.0, eager execution is enabled by default Eager Execution. Therefore, this method works best with TensorFlow version 2.0. See the following code example.

import tensorflow as tf

tensor = tf.constant([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
print("Tensor = ", tensor)
array = tensor.numpy()
print("Array = ", array)

Output:

Tensor =  tf.Tensor(
[[1 2 3]
 [4 5 6]
 [7 8 9]], shape=(3, 3), dtype=int32)
Array =  [[1 2 3]
 [4 5 6]
 [7 8 9]]

tf.constant()In the code above, we first created and initialized a Tensor object in Python using the tensor.numpy() function tensor. We printed it in Python using the tensor.numpy() function tensorand converted it to a NumPy array array. Finally, we printed it 数组.

Use the function in Python Tensor.eval()to convert a tensor to a NumPy array

We can also Tensor.eval()convert a Tensor to a NumPy array in Python using the function. This method is not supported in TensorFlow version 2.0. Therefore, we have to keep the previous version 1.0 of TensorFlow or disable all the behaviors of version 2.0 of the TensorFlow library. See the following code example.

import tensorflow.compat.v1 as tf

tf.disable_v2_behavior()
tensor = tf.constant([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
print("Tensor = ", tensor)
array = tensor.eval(session=tf.Session())
print("Array = ", array)

Output:

Tensor =  Tensor("Const_1:0", shape=(3, 3), dtype=int32)
Array =  [[1 2 3]
 [4 5 6]
 [7 8 9]]

In the code above, we use tensor.eval()the function in Python to convert the Tensor object tensorto a NumPy array array. We first imported version 1.0 of the TensorFlow library and disabled all behaviors of version 2.0. Then, we tf.constant()created and initialized using the function tensorand tensorprinted the value in . We then executed tensor.eval()the function and saved the returned value arrayin and printed the value in array.

Use the function in Python TensorFlow.Session()to convert a tensor to a NumPy array

TensorFlow.Session()is another method that can be used to convert a Tensor to a NumPy array in Python. This method Tensor.eval()is very similar to the previous method with the function. Version 2.0 of the TensorFlow library also does not support this method. We either have to install version 1.0 of the TensorFlow library or disable all the behaviors of version 2.0 of the TensorFlow library. We can pass a Tensor object to TensorFlow.Session().run()the function to convert that Tensor object to a NumPy array in Python. See the following code example.

import tensorflow.compat.v1 as tf

tf.disable_v2_behavior()
tensor = tf.constant([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
print("Tensor = ", tensor)
array = tf.Session().run(tensor)
print("Array = ", array)

Output:

Tensor =  Tensor("Const_6:0", shape=(3, 3), dtype=int32)
Array =  [[1 2 3]
 [4 5 6]
 [7 8 9]]

In the code above, we use tf.Session.run(tensor)the function in Python to convert the Tensor object tensorto a NumPy array array. We first imported the 1.0 compatible TensorFlow library and disabled all the behaviors of version 2.0. Then, we created the tensor object tensorand printed tensorthe value of . Then, we used tf.Session.run(tensor)the function to tensorconvert the tensor to arraya NumPy array and printed the value in array.