课程: Building and Deploying Deep Learning Applications with TensorFlow

今天就学习课程吧！

今天就开通帐号，24,100 门业界名师课程任您挑！

Visualize the computational graph

Visualize the computational graph - TensorFlow教程

课程: Building and Deploying Deep Learning Applications with TensorFlow

开始一个月试用 购买团队方案

Visualize the computational graph

“

- [Narrator] It's always helpful to visualize what's happening with your data. This is where TensorBoard comes in. It takes what we do in TensorFlow and creates a graphical representation of it. Before we can open up TensorBoard, we need some log files to look at. Let's open up train_model.py and let's run it. When we run it, this will train the neural network and save log files that we'll be able to view with TensorBoard. Right click, choose run. You can see here in Pycharm the new logs folder has been created with the log files. Alright, lets run TensorBoard. To run TensorBoard, let's open up a Terminal window. In Pycharm you can hover your mouse on the bottom left and then click Terminal, but if you prefer, you can also open the standard operating system Terminal window outside of Pycharm. To run TensorBoard, type TensorBoard and then --logdir. And we'll give the folder where our logs are written, in this case, 05/logs, note that on Windows, you'll use backslashes instead of…

内容