Why are some int parameters Tensors for some op but simple int for others?

Here is is an example:

  • ConcatV2 has an axis Tensor parameter (as input)
  • Unpack has an axis int parameter (as parameter)

Why is that?

1 Like

It is not possible to have a Tensor with more than one data type. It is possible, however, to serialize arbitrary data structures as strings and store those in tensors.

How is that related to the axis parameter in the examples I gave?

This post was flagged by the community and is temporarily hidden.