In a python environment, I'm following along with "Getting Started With TensorFlow".
>>> tf.constant(3.0, dtype=tf.float32)
# Tensor("Const:0", shape=(), dtype=float32)
>>> tf.constant(4.0) # also tf.float32 implicitly
# Tensor("Const_1:0", shape=(), dtype=float32)
I am wondering what the meaning and purpose is of ":0" as the visualization of these nodes will be just "Const" and "Const_1". Is there ":1" for tf.constant()?
I searched in the source tensorflow/python/framework/constant_op.py, but it doesn't seem to have my answer.
Edit for duplicate questions on How does TensorFlow name tensors?
My bad for not concerning tf.Variable and having forgotten constant is also a case of variables.
