I have attached an image of a scalar summary plot in Tensorflow. As you can see, there are two orange lines plotted. According to the legend, one corresponds to the actual value (the ghost line) and the other corresponds to the Smoothed Value (the main line). I do not understand why these two lines are plotted as the loss is a single value in each iteration of the training process.
I hope someone can help me. Thank you
Loss during Training
The smoothing is over time (the x-axis), not over multiple values for each iteration. This produces a more consistent curve when the loss can be quite erratic. Turn the smoothing slider down if you want to see the real value of the loss at each time step.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With