A dataflow graph is the representation of a computation where the nodes represent units of computation, and the edges represent the data consumed or produced by the computation.
In the context of tf.Graph, every API call defines tf.Operation (node) that can have multiple inputs and outputs tf.Tensor (edges). For instance, referring to our main example, when calling tf.constant([[1, 2], [3, 4]], dtype=tf.float32), a new node (tf.Operation) named Const is added to the default tf.Graph inherited from the context. This node returns a tf.Tensor (edge) named Const:0.
Since each node in a graph is unique, if there is already a node named Const in the graph (that is the default name given to all the ...
