Artificial Intelligence By Example

Displaying the data flow graph in TensorBoard

A practical way to obtain a visual display of your architecture and results is to have your own little TensorBoard_reader.py, as shown in the following lines of code:

def launchTensorBoard(): import os os.system('tensorboard --logdir=' + './logs/tensorboard_logs') returnimport threadingt = threading.Thread(target=launchTensorBoard, args=([]))t.start()#In your browser, enter http://localhost:6006 as the URL#and then click on Graph tab. You will then see the full graph.

All you have to do is git the output directory to the log directory path in your program. It can be included in the main source code.

Once launched, you will see images of the previous section. You are now ready to enter corporate ...