18. Quiz: TensorFlow Softmax Workspaces

TensorFlow Softmax

Now that you've built a softmax function from scratch, let's see how softmax is done in TensorFlow.

x = tf.nn.softmax([2.0, 1.0, 0.2])

Easy as that! tf.nn.softmax() implements the softmax function for you. It takes in logits and returns softmax activations.

Quiz

Use the softmax function in the quiz below to return the softmax of the logits.

Workspace

This section contains either a workspace (it can be a Jupyter Notebook workspace or an online code editor work space, etc.) and it cannot be automatically downloaded to be generated here. Please access the classroom with your account and manually download the workspace to your local machine. Note that for some courses, Udacity upload the workspace files onto https://github.com/udacity , so you may be able to download them there.

Workspace Information:

  • Default file path:
  • Workspace type: jupyter
  • Opened files (when workspace is loaded): n/a

Quiz

Answer the following 2 questions about softmax.

What happens to the softmax probabilities when you multiply the logits by 10?

SOLUTION: Probabilities get close to 0.0 or 1.0

What happens to the softmax probabilities when you divide the logits by 10?

SOLUTION: The probabilities get close to the uniform distribution