Which of the following options is not the Tensorflow build process?

Question

Options

• ABuilding a calculation chart
• BInput tensor
• CGenerate a session
• DUpdate weights I Right Answers)

• A
  3% (2)
• B
  7% (4)
• C
  12% (7)
• D
  78% (46)

Explanation

Option D is wrong because updating weights is a training operation, not a step in TensorFlow's graph build process. Weight updates occur during the execution/training phase when the session runs backpropagation - they happen after the graph is built and a session is active.

Why the distractors are wrong (i.e., why A, B, and C ARE part of the build process):

• A (Building a calculation graph): This is the defining step of TensorFlow's build phase - you construct a static computational graph of nodes and operations before any data flows through it.
• B (Input tensor): Defining input tensors (or placeholders in TF 1.x) is how you specify the graph's entry points - a necessary part of constructing the graph.
• C (Generate a session): Creating a tf.Session() is the final build-phase step that prepares the runtime environment to execute the graph.

Memory tip: Think of TensorFlow's classic workflow as Build → Run → Train. Steps A, B, and C all belong to Build (define the graph and session). Weight updates belong to Train (run the optimizer inside the session) - so D is the odd one out.

No community discussion yet for this question.