Reference Guide

42 Dell EMC Ready Solutions for AI Deep Learning with NVIDIA | v1.0
Figure 35: Stop the server
5.2 Tensorboard Integration
Besides the TensorFlow framework, the portal also provides the Tensorboard visualization tool. The
Tensorboard is used to visualize the TensorFlow computation graph, plot quantitative metrics about the
execution of a graph, and show some other additional data. To use Tensorboard, the user needs to use
TensorFlow FileWriter API to serialize the wanted data into a directory. Figure 36 shows an example in which
Figure 36: An example of using Tensorboard
button will be shown up. Tensorboard will output all information in a
separate tab as shown in Figure 37. The output result in Figure 37 is only an example. A user may get a different
result depending on what information are written to the logs.