TF2.0: How to create a tensor with numpy attribute

Question

I am trying to create a tensor with numpy attribute and tried this:

tensor1 = tf.reshape(tf.constant([100, 100, 100], dtype=tf.int64), (-1, 1))

print(tensor1)

 #tensor1
 tf.Tensor(
[[100]
 [100]
 [100]], shape=(3, 1), dtype=int64)

But the output does not show the tensor ID or the numpy attribute where the values will be an array. How do I create a tensor of that format

simon · Accepted Answer · 2019-10-16 16:51:51Z

1

It still has the numpy method, just call it:

tensor1 = tf.reshape(tf.constant([100, 100, 100], dtype=tf.int64), (-1, 1))

print(tensor1)
print(tensor1.numpy())

Also, look at the difference between

print(str(tensor1))

and

print(repr(tensor1))

answered Oct 16, 2019 at 16:51

simon

8161 gold badge8 silver badges12 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

TF2.0: How to create a tensor with numpy attribute

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related