tf.get_session_handle(data, name=None)

tf.get_session_handle(data, name=None)

See the guide: Tensor Handle Operations > Tensor Handle Operations

Return the handle of data.

This is EXPERIMENTAL and subject to change.

Keep data "in-place" in the runtime and create a handle that can be used to retrieve data in a subsequent run().

Combined with get_session_tensor, we can keep a tensor produced in one run call in place, and use it as the input in a future run call.


  • data: A tensor to be stored in the session.
  • name: Optional name prefix for the return tensor.


A scalar string tensor representing a unique handle for data.


  • TypeError: if data is not a Tensor.


c = tf.multiply(a, b)
h = tf.get_session_handle(c)
h =

p, a = tf.get_session_tensor(h.handle, tf.float32)
b = tf.multiply(a, 10)
c =, feed_dict={p: h.handle})

Defined in tensorflow/python/ops/