How to get or convert contents of Keras Tensor using tf.keras.Input()?

[shahrokh nasseri]

Hello Dears Keras Developers and users

In the example given about the classification of structural data (Structured data classification from scratch in Code examples) on the web page of Keras, I want to get/see the contents of objects such as age_encoded or trestbps_encoded.

These objects are tensorflow.python.framework.ops.Tensor.

How can I get or view the contents of these tensors?

I googled very much, but could not find correct solution for doing it. I noticed that I can view the content of eager tensor, for example:

import tensorflow as tf
a = tf.constant([[1, 2], [3, 4]])
print(a)
tf.Tensor(
[[1 2]
 [3 4]], shape=(2, 2), dtype=int32)

I can not get the contents of age_encoded:

print(age_encoded)
Tensor("normalization/truediv:0", shape=(None, 1), dtype=float32)

In other words, can I convert tensorflow.python.framework.ops.Tensor to tensorflow.python.framework.ops.EagerTensor and how to do it?

Specifications Tensorflow and Keras installed in Anaconda (ipython):

Keras: 2.3.0-tf

Tensorflow: 2.2.0

Specifications Tensorflow and Keras installed Pycharm:

Keras: 2.4.0

Tensorflow: 2.3.0

Please guide me.

I am waiting for your help.

Best regards.

Shahrokh

[Lance Norskog]

Try eval().

https://www.tensorflow.org/api_docs/python/tf/Tensor#eval

--
You received this message because you are subscribed to the Google Groups "Keras-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to keras-users...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/keras-users/43556053-8d41-4a50-8b97-b8f10b29c64an%40googlegroups.com.

--

Lance Norskog
lance....@gmail.com
Redwood City, CA

[shahrokh nasseri]

Dear Lance

Thanks a lot for your guidance. Unfortunately I got another error after using the eval function.

I run the model provided in Keras until after the object of age_encoded, and then run the following commands and get error.

I found a solution in stackoverflow, but again I get error:

So I still can not see the contents of the tensor object.

Please guide me.

Best regards.

Shahrokh