There's more…

Be sure to run the preceding script from the command line:

$ python3 using_tensorboard.py

Run the command: $tensorboard --logdir="tensorboard"   Then navigate to http://127.0.0.0:6006
Generation 10 of 100. Train Loss: 20.4, Test Loss: 20.5.
Generation 20 of 100. Train Loss: 17.6, Test Loss: 20.5.
Generation 90 of 100. Train Loss: 20.1, Test Loss: 20.5.
Generation 100 of 100. Train Loss: 19.4, Test Loss: 20.5.

We'll then run the preceding specified command to start Tensorboard:

$ tensorboard --logdir="tensorboard"
Starting TensorBoard b'29' on port 6006
(You can navigate to http://127.0.1.1:6006)

Here is a sample of what we can see in Tensorboard:

Figure 1: The scalar value, our slope estimate, visualized in Tensorboard.

Here, we can see a plot over the 100 generations of our scalar summary, the slope estimate. We can see that it does, in fact, approach the true value of 2:

Figure 2: Here we visualize histograms of the errors and residuals for our model.

The preceding graph shows one...