The computations you'll use TensorFlow for—such as training a massive deep neural network—can be complex and confusing, and its corresponding computational graph will be complex as well. To make it easier to understand, debug, and optimize TensorFlow programs, the TensorFlow team have included a suite of visualization tools called TensorBoard, which is a suite of web applications that can run through your browser. TensorBoard can be used to visualize your TensorFlow graph, plot quantitative metrics about the execution of your graph, and show additional data such as images that pass through it. When TensorBoard is fully configured, it looks like this:
To understand how TensorBoard works, we are going to build a computational graph which will act as a classifier for the MNIST dataset, which is a dataset of handwritten...
